- Email: [email protected]
Logarithmic mean Divisia index (LMDI) decomposition of coal consumption in China based on the energy allocation diagram of coal flows
Energy 85 (2015) 366e378
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Logarithmic mean Divisia index (LMDI) decomposition of coal consumption in China based on the energy allocation diagram of coal flows ChinHao Chong, Linwei Ma*, Zheng Li, Weidou Ni, Shizhong Song State Key Laboratory of Power Systems, Department of Thermal Engineering, Tsinghua-BP Clean Energy Centre, Tsinghua University, Beijing 100084, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 December 2014 Received in revised form 17 March 2015 Accepted 20 March 2015 Available online 2 May 2015
This manuscript attempted to analyze the influencing factors of coal consumption growth in China using the logarithmic mean Divisia index (LMDI) decomposition method developed based on the physical processes of coal utilization from raw coal to the end-use sectors. By mapping the energy allocation diagram of coal flows, we built a method to balance the energy allocation of coal flows and derived several technical influencing factors. These factors were used to develop an LMDI decomposition method suitable for analyzing the coal consumption growth of complex coal-use systems, such as that of China. The method is subsequently applied to analyze the influencing factors of China's coal consumption growth from 2001 to 2011. The results indicate the rapid growth of GDP (gross domestic production) per capita, which heavily relied on the expansion of heavy industry as the dominant factor driving coal consumption growth. Improvement in the energy efficiency of coal power generation and coal end-use combustion were the primary factors reducing coal consumption. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Coal consumption LMDI Sankey diagram Coal flow Influencing factor
1. Introduction China now is the largest coal producer and coal consumer in the world, accounting for 46.6% of the global coal production and 50.3% of the global coal consumption in 2013 [1]. Faced with challenges of global climate change and domestic environment protection, controlling the rapidly increasing coal consumption in China has become an important issue, not only for China but also for the world [2e4]. Referring to IEA and BP, policy choices in China, which has outlined plans to cap the share of coal regarding total energy use, will be particularly important as China now consumes as much coal as the rest of the world combined [5]. In addition, the trends of global coal growth before 2035 will be greatly influenced by China's profile [6]. However, most of the discussions presented in previous studies have focused on the coal needs and targets for controlling the total coal consumption [7e9]. Quantitative analyses on the influencing factors of coal consumption growth that consider the rapid and dynamical development of energy systems in China are lacking, which are essential for developing effective policies to control coal consumption and deserve further study.
* Corresponding author. E-mail addresses: [email protected], [email protected] (L. Ma). http://dx.doi.org/10.1016/j.energy.2015.03.100 0360-5442/© 2015 Elsevier Ltd. All rights reserved.
The index decomposition analysis (IDA) method has been widely applied to analyze the influencing factors of energy consumption growth, including those based on the Laspeyres index and the Divisia index [10]. Ang et al. [11e13] presented a review of the development and applications of IDA methods and recommended the logarithmic mean Divisia index I (LMDI) method because it is robust and convenient for many applications. Many studies have utilized the LMDI method to decompose the total energy consumption growth of various regions and countries [14e16], and the commonly considered factors are population growth, GDP (gross domestic production), economic structure, energy intensity and energy mix. However, to apply the LMDI method to coal consumption growth in China, for which research is lacking, the factors that are considered need to be expanded because coal is only one part of the energy system. In addition, China consumes a huge amount of coal, and the uses of coal have been diversified for various sectors, including power and heat generation, coking and end-use combustion, etc. Therefore, the structural changes through energy use systems, including the stages of primary energy sources, energy conversion and energy end-use, in addition to technically detailed factors such as energy conversion efficiency and energy end-use efficiency, should be carefully considered to further develop the LMDI method to decompose the technical details of coal consumption growth.
C. Chong et al. / Energy 85 (2015) 366e378
Abbreviations Subscript i ith industry involved Subscript j jth energy involved Subscript k kth influencing factors involved IDA index decomposition analysis LMDI logarithmic mean Divisia index SQ standard quantity PEQ primary energy quantity CBPEQ coal-based primary energy quantity ESQ,j jth energy expressed in SQ ESQ,ij jth energy expressed in SQ in the ith industry EPEQ,j jth energy expressed in PEQ ECBPEQ,j jth energy expressed in CBPEQ C total raw coal consumption C0 total raw coal consumption at time 0 CT total raw coal consumption at time T P population GDP gross domestic production GDPi value added of the ith sector Q GDP per capita Si proportion of the ith industry Ii energy intensity of the ith industry Mij proportion of the jth fuel in the ith industry Kcombustion,j end-use combustion factor of energy j KPEQ,j primary energy quantity converted factor of energy j
This study aims to develop an LMDI method suitable for analyzing the coal consumption growth of countries with complex coal use systems, such as in China, and to apply the method to analyze the influencing factors of coal consumption growth in China. First, by mapping China's coal flows to Sankey diagrams, we studied the physical processes of coal use from the raw coal supply to the end-uses as electricity, heat and coke, and thus derive key influencing factors that need to be further considered. Then, using these key influencing factors, we develop an LMDI method suitable for analyzing such coal use systems and apply it the coal consumption growth in China from 2001 to 2011, during which China's coal consumption increased by an average growth ratio of 12.5%. The contents of this paper are organized as follows: an introduction of the methodology and data input are described in Section 2, the results and discussion are given in Section 3, and conclusions and suggestions are summarized in Section 4. 2. Methodology and data input 2.1. Sankey diagram and technical influencing factors Sankey diagrams are popular in energy system analyses [17e19] and coal system analyses [20e22]. In this work, an energy allocation Sankey diagram, which presents the energy balance from raw coal supplies to the final end-uses, is used as a tool to understand the physical process of coal systems and to derive the key influencing factors of coal consumption growth. 2.1.1. Diagram structure and energy balance In this study, the energy allocation Sankey diagram of coal was divided into three stages: 1.) raw coal supply 2.) coal transformation and 3.) end-use sector. The diagram was mapped according to the first law of thermodynamics. The width of the flow signified the quantity of the energy, and the colour of the flow signified the type of energy.
Kcoal,j Kfossil,j DCpop
DCaff DCstr DCint DCmix DCcom DCpeq DCcoal DCfos F0k FTk DCFk
lk
367
coal component factor of energy j fossil fuel component factor of energy j increment of raw coal consumption caused by change of P increment of raw coal consumption caused by change of Q increment of raw coal consumption caused by changes of Si increment of raw coal consumption caused by changes of Ii increment of raw coal consumption caused by changes of Mij increment of raw coal consumption caused by changes of Kcombustion,j increment of raw coal consumption caused by changes of KPEQ,j increment of raw coal consumption caused by changes of Kcoal,j increment of raw coal consumption caused by changes of Kfossil,j value of the kth influencing factor at time 0 value of the kth influencing factor at time T increment of raw coal consumption caused by the kth influencing factor elastic efficiency of the kth influencing factor
No energy loss is reflected in energy allocation Sankey diagrams. Therefore, we trace the energy of raw coal as it is exploited. After production and importation, the raw coal is processed and converted into secondary energy sources, such as coke, heat and electricity, in transformation sector and finally consumed at the end-use sectors. By compensating for all of the energy losses that occur during energy processing and transformation, we can express the secondary energy consumption as a coal-based primary energy quantity (CBPEQ), which indicates the amount of raw coal consumption required to produce a secondary energy source. 2.1.2. Original data e energy balance table The Energy Balance of China and the Final Energy Consumption by Industrial Sector, are two statistical tables in the China Energy Statistical Yearbook, are used as the original data sources for mapping the Sankey diagram. The data in the Energy Balance of China is vertically divided into 6 parts: A) primary energy supply, B) input () and output (þ) of the transformation, C) loss, D) final consumption, E) statistical difference, and F) total energy consumption, and horizontally divided into 30 types of energy, including raw coal, various coal products, crude oil, various oil products, natural gas, electricity, heat and recovery energy. The details of the final energy consumption of industrial sectors are further presented in another table named Final Energy Consumption by Industrial Sector, which includes more than 40 subsectors. To illustrate the structure of the final consumption of the industrial sectors in the diagram, we re-categorized the subsectors in the industrial sector, as shown in Table 1. Because the coal preparation ratio according to the Energy Balance of China is much lower than the actual data, we adjusted the ratio according to another study [23] in the mapping. However, it became too complex to apply this adjusted ratio because we are unable to obtain more accurate information to estimate the losses in the coal preparation process after the revision. Hence, the original ratio is kept in the LMDI decomposition.
368
C. Chong et al. / Energy 85 (2015) 366e378
Table 1 The re-categorized industrial subsectors. Industrial subsector 1 Non-energy industrial subsector 1.1 Non-energy mining Mining and processing of ferrous metal ores Mining and processing of non-ferrous metal ores Mining and processing of non-metal ores Mining of other ores 1.2 Smelting and pressing of ferrous metals 1.3 Smelting and pressing of non-ferrous metals 1.4 Manufacture of non-metallic mineral and chemical products 1.5 Manufacture of raw chemical materials and chemical products 1.6 Manufacture of food and beverages Processing of food from agricultural products Manufacture of food Manufacture of liquor, beverages and refined tea 1.7 Manufacture of textiles Manufacture of textiles Manufacture of textiles, wearable apparel and accessories 1.8 Manufacture of machinery and vehicles Manufacture of general purpose machinery Manufacture of special purpose machinery Manufacture of automobiles Manufacture of railways, ships, aerospace and other transportation equipment 1.9 Manufacture of paper and paper products 1.10 Other non-energy industrial subsector Manufacture of tobacco Manufacture of leather, fur, feather and related products and footwear Processing of timber, manufacture of wood, bamboo, rattan, palm and straw products Manufacture of furniture Printing and reproduction of recording media Manufacture of articles for culture, education, arts and crafts, sports and entertainment activities Manufacture of medicines Manufacture of chemical fibres Manufacture of rubber and plastics products Manufacture of metal products Manufacture of electrical machinery and apparatuses Manufacture of computers, communication and other electronic equipment Manufacture of measuring instrument and machinery Utilization of waste resources Repair service of metal products, machinery and equipment Production and supply of water Others manufacture 2 Energy industry subsector Production and supply of electric power and heat power Production and supply of gas Mining and preparation of coal Extraction of petroleum and natural gas Processing of petroleum, coking and Processing of nuclear fuel
2.1.3. Data processing and diagram mapping In our previous study [24], we presented a method that can quickly generate the data required for mapping an energy allocation diagram of the coal in China, which is based on a series of standardized steps and rigid equations that can be programmed. The original data from the statistical tables are processed using this method to prepare a balanced dataset for the mapping. We define 9 energy types related to coal use in the mapping, as illustrated in Table 2, and assume that all end-use fuels are burned at an average combustion efficiency except coke. 2.1.4. Technical influencing factors derived in the mapping process In the process of mapping the energy allocation diagram of coal, we [24] defined 4 factors to express the end-use energy carriers in the CBPEQ type, as shown in equation (1). These factors include: 1) the end-use combustion factor, 2) the primary energy quantity converted factor, 3) the coal component factor, and 4) the fossil fuel component factor, as illustrated in Table 3. These factors are used to develop the LMDI decomposition method of coal consumption growth in Section 2.2.
ECBPEQ ;j ¼ ESQ ;j $Kcombustion;j $KPEQ ;j $Kcoal;j $Kfossil;j
(1)
2.2. LMDI decomposition approach Most previous studies have adopted equation (2) to express the energy consumption as standard quantity (ESQ) and consider influencing factors such as population (P), GDP per capita (GDP/P), economic structure (GDPi/GDP), energy intensity (ESQ,i/GDPi) and Table 2 The 9 coal-derived energy types involved in the energy allocation diagram of coal. No.
Energy types
1 2 3 4 5 6 7 8 9
Raw coal Cleaned coal: cleaned coal, other washed coal Coke Gas: coke oven gas, other gas Briquettes Gangue Other coking products Heat Electricity
C. Chong et al. / Energy 85 (2015) 366e378
369
Table 3 Description of the technical influencing factors. Factor
Description
Kcombustion,j
1 To consider the impact of the end-use combustion efficiency, we assume that all of the fuels except coke, heat and electricity consumed in the end-use sector are converted into heat, with an average end-use combustion efficiency, 4. 2 Kcombustion,j is defined as the unit of energy input to the boiler or furnace required to produce 1 unit of heat. 3 Kcombustion of coke, heat and electricity are 1. 1 The primary energy quantity type transformation factor, defined as the total number of units of primary energy that are consumed to produce 1 unit of secondary energy. This value is used to amplify secondary energy to primary energy required to produce the secondary energy. 2 As the processing and transportation losses can be neglected, KPEQ of raw coal, natural gas, crude oil and recovery gas are set as 1. 1 The coal component factor, representing the share of coal in primary energy that is consumed to produce 1 unit of secondary energy. 2 Kcoal of raw coal is 1. Kcoal values of natural gas, crude oil and recovery gas are 0.
KPEQ,j
Kcoal,j
Kfossil,j
Equation
1 The Kcoal of heat and electricity should be revised using the fossil fuel component factor Kfossil, as the calculated processes only considered the transformation sectors mentioned in the China Energy Balance and did not include renewable energy and nuclear power. 2 Kfossil shows the share of fossil fuel consumed to produce 1 unit of secondary energy. 3 Kfossil of other energy sources are 1.
energy mix (ESQ,ij/ESQ,i), in which i represents economic sectors and j represents energy types.
ESQ ¼
X
P$
ij
GDP GDPi ESQ ;i ESQ ;ij $ $ $ P GDP GDPi ESQ ;i
(2)
Further considering the technical influencing factors derived in Section 2.1.4, we expand equation (2) to equation (3) to express the total amount of raw coal consumption of China's entire industry, including primary industries, secondary industries, and tertiary industries, in the CBPEQ. The limitation of this method is that it does not include the residential consumption of raw coal, but fortunately this is generally a small portion (for instance, 10.7% in 2011 for China) of the entire raw coal consumption. The symbol items in equation (3) are explained in Table 4. The LMDI formulae for decomposing coal consumption growth in the entire industry of China are presented in Table 5.
ECBPEQ ¼
X ij
P$
GDP GDPi ESQ ;i ESQ ;ij $ $ $ $K $K P GDP GDPi ESQ ;i combustion;j PEQ ;j
(3)
$Kcoal;j $Kfossil;j 2.3. Data input for LMDI decomposition We obtained the energy data from the China Energy Statistical Yearbook [25e27] and the economic data from the China Statistical Yearbook [28e30]. The end use combustion efficiency referred to Refs. [31,32,22], as shown in Table 6. Using our data-processing method previously developed [24], we calculate Kcombustion (Table 7), KPEQ (Table 8), Kcoal (Table 9) and Kfossil (Table 10) of each fuel in 2001, 2006 and 2011. 3. Results and discussions 3.1. Energy allocation diagram of coal use in China The energy allocation diagrams of coal in China for 2001, 2006 and 2011 are presented in Figs. 1e3. Referring to these diagrams,
Kcombustion;j ¼ f1 4: end-use combustion efficiency in average
P KPEQ ;j ¼
n
KPEQ;n $SQn;input
P
SQj;output
j
KPEQ,n: PEQ transformation factor of the nth fuel input to produce fuel j SQn,input: Input of the nth fuel to produce fuel j expressed in SQ SQj,ouput: Output of fuel j expressed in SQ P Kcoal;j ¼
Kcoal;n $PEQn;input
n
PEQj;output
Kcoal,n: Coal component factor of the nth fuel input to produce fuel j PEQn,input: Input of the nth fuel to produce fuel j expressed in PEQ type PEQn,output: Output of fuel j expressed in PEQ type SQ
fossil;j Kfossil;j ¼ SQfossil;j þSQ nonfossil;j
Subscript j: the jth fuel type involved SQfossil,j: Fuel j produced by using fossil fuel expressed in SQ SQnon-fossil,j: Fuel j produced by using non-fuel fuel expressed in SQ
the main features and dynamics of China's coal use system are as follows: (1) Coal supply primarily relies on domestic production. Although coal import in China increased rapidly from 2009 after it became a net coal importer, the self-satisfaction ratio of coal was still as high as 96.8% in 2011. (2) The structure of coal transformation is quite diversified. From 2001 to 2011, the proportion of coal used for electricity generation changed from 55.1% to 47.3%, for coking from 13.4% to 16.4%, for heat generation from 6.8% to 4.6%, and for end-use combustion from 15.1 to 31.7%, as outlined in Fig. 4. During this period, the trend for coal use shifted from electricity generation and heat generation toward end-use combustion and coking. (3) The end-use of coal-derived energy carriers is primarily consumption, and a greater percentage of coal being consumed by the industrial sector (manufacturing), which was responsible for 60.7% and 68.3% of total coal consumption in 2001 and 2011, especially heavy industrial sectors, as shown in Fig. 5. Referring to Fig. 6, coal consumption for the manufacturing of ferrous metals, non-ferrous metals, non-metallic mineral and chemicals increased from 35% to 51% between 2001 and 2011.
3.2. LMDI decomposition results Through LMDI decomposition, we acquire the increment of consumption of coal of China's entire industry and its primary industry, secondary industry and tertiary industry caused by each influencing factor between 2001 and 2006, 2006e2011 and 2001e2011. The decomposition results are given in Tables 11e14, and the summarized the decomposition results of the entire industry are shown in Fig. 7. The growth of GDP per capita is the dominant factor driving the coal consumption growth in 2001e2011, whereas the influences of other factors are relatively small. In the following sections, we discuss each of these influencing factors.
370
C. Chong et al. / Energy 85 (2015) 366e378
Table 4 The symbol items in equation (3). Items
Descriptions
Subscription i
1 Primary industries 2 Secondary industries 3 Tertiary industries Raw coal, cleaned coal, briquettes, gangue, coke, coke oven gas, blast furnace gas, converter gas, other gas, other coking products, crude oil, oil products (gasoline, kerosene, diesel oil, fuel oil, naphtha, lubricants, paraffin waxes, white spirit, bitumen asphalt, petroleum coke LPG, refinery gas and other petroleum products), natural gas, LNG, heat and electricity. Population Gross domestic production Value added of the ith sector Energy consumption of the ith sector expressed in SQ Fuel j consumption of the ith sector expressed in SQ Combustion factor of fuel j Primary energy quantity type conversion factor of fuel j Coal component factor of fuel j Fossil fuel component factor of fuel j
Subscription j
P GDP GDPi ESQ,i ESQ,ij Kcombustion,j KPEQ,j Kcoal,j Kfossil,j
Table 5 LMDI formulae for decomposing coal consumption growth. IDA identify
C¼
P
Cij ¼
PP
ij
Change scheme
i
j
E
E
GDPi SQ ;i SQ ;ij P$GDP P $ GDP $GDPi $ ESQ ;i $Kcombustion;j $KPEQ ;j $Kcoal;j $Kfossil;j
DCtot ¼ C T C 0 ¼ DCpop þ DCaff þ DCstr þ DCInt þ DCmix þ DCcom þ DCpeq þ DCcoal þ DCfos P P CijT Cij0 T DCpop ¼ ln PP 0 ln C T ln C 0
LMDI formulae
ij
ij
Q ¼ GDP P
P
DCaff ¼
i Si ¼ GDP GDP
DCstr ¼
ij E
SQ ;i Ii ¼ GDP i
DCint ¼
QT Q0
CijT Cij0 ln CijT ln Cij0
ln
STi S0i
CijT Cij0 ln CijT ln Cij0
ln
IiT Ii0
P ij
Mij ¼
ESQ ;ij Ei;SQ
DCmix ¼
P
DCcom ¼
ln
CijT Cij0 ln CijT ln Cij0
ln
CijT Cij0 ln CijT ln Cij0
ln
P ij
KPEQ,j
P
DCpeq ¼
ij
Kcoal,j
DCcoal ¼
Kfossil,j
DCfos ¼
P ij
P ij
!
CijT Cij0 ln CijT ln Cij0
ij
Kcombustion,j
ln
ij
P
ij
CijT Cij0 ln CijT ln Cij0
CijT Cij0 ln CijT ln Cij0
CijT Cij0 ln CijT ln Cij0
! !
MijT Mij0
! T Kcombustion;j 0 Kcombustion;j
! T KPEQ;j 0 KPEQ;j
! ln
T Kcoal;j 0 Kcoal;j
! ln
T Kfossil;j 0 Kfossil;j
Note: the superscripts 0 and T give the parameter in period 0 and period T, respectively.
Table 8 KPEQ of each fuel type. Table 6 The end-use combustion efficiency 4. Fuel type
2001
2006
2011
End use combustion efficiency
0.477
0.514
0.552
Note: the combustion efficiency of the furnace and kiln were 62.5% and 35% in 2005 [22]; and 67.5% and 37% in 2010 [31], respectively. We assumed that the combustion efficiency increased linearly from 2001 to 2011, and the ratio of the furnace and kiln remained 1.33 [32].
Fuel type
2001
2006
2011
Raw coal Cleaned coal Briquette Gangue
1.00 1.06 1.01
1.00 1.04 1.11
e
e
1.00 1.07 1.15 1.07
Coke Coke oven gas Blast furnace gas
1.07 1.07
1.06 1.06
e
e
Converter gas
Fuel type
2001
2006
2011
Other gas Other coking products Crude oil Oil products Natural gas LNG
Coke, heat and electricity Others
1.00 1.71
1.00 1.57
1.00 1.46
Heat Electricity
Table 7 Kcombustion of each fuel type.
e
e
1.31 1.09 1.00 1.02 1.00
1.56 1.07 1.00 1.03 1.00
e
e
1.19 3.04
1.35 2.86
1.10 1.11 1.00 1.00 1.43 1.11 1.00 1.03 1.00 1.16 1.32 2.56
C. Chong et al. / Energy 85 (2015) 366e378 Table 9 Kcoal of each fuel type. Fuel type
2001
2006
2011
Heat Electricity Other gas
0.84 0.95 1.00
0.91 0.97 0.94
0.91 0.97 0.95
Note: (1) raw coal, cleaned coal, briquette, gangue, coke, coke oven gas, other coking products: 1.00 or almost 1.00; (2) blast furnace gas, converter gas, crude oil, oil products, natural gas, LNG: 0.00 or almost 0.00. Table 10 Kfossil of each fuel type. Fuel type
2001
2006
2011
Heat Electricity Others
1.00 0.80 1.00
1.00 0.83 1.00
0.89 0.81 1.00
371
3.2.1. The influence of population The population growth in China contributed to coal consumption growth during both 2001e2006 and 2006e2011, although the contribution is quite limited. The reason may be that the population growth of China is relative small, with a total increase of 6.2% in 2011 compared to 2001 due to the rigid population restriction policy that have been in place since the 1970s in China.
3.2.2. The influence of GDP per capita The increase in GDP per capita, whose average annual growth rate was approximately 10%, is the main contributor of coal consumption growth from 2001 to 2011. The growth of the GDP per capita will facilitate additional services that are required for people, thus promoting more energy consumption, especially during this special development stage of China. Currently, China is
Fig. 1. Energy allocation diagram of coal, China 2001.
Fig. 2. Energy allocation diagram of coal, China 2006.
372
C. Chong et al. / Energy 85 (2015) 366e378
simultaneously experiencing rapid industrialization, urbanization and motorization [33], and the new added demand of services primarily pertain to energy-intensive products, such as houses, cars, machines and infrastructures. These changes have not only rapidly increased of the energy consumption of domestic consumption expenditures, but they have also increased the energy consumption embedded in gross capital formation and exported products. Referring to Fig. 8, gross capital formation was a main driving force of economic growth in China. According to Ma et al. [19], domestic consumption expenditures accounted for 38.8%, gross capital formation (infrastructure construction) accounted for 32.3% and export products accounted for 28.9% of China's total energy consumption in 2005. Therefore, the coal consumption growth from 2001 to 2011 was due largely to rapid GDP growth, which was accompanied by a huge demand for energy-intensive products and a rapid expansion of energy-intensive industries that caused a large increase of coal consumption.
3.2.3. The influence of economic structure and energy intensity Changes in economic structure and energy intensity both increased coal consumption in 2001e2006 but also decreased coal consumption in 2006e2011. This occurred because the proportion and energy intensity of secondary industries increased in 2001e2006, whereas both decreased in 2006e2011. These influencing factors did not change in the primary and tertiary industries, and therefore did not have significant impacts on coal consumption compared to secondary industries, as shown in Figs. 9 and 10. Although energy intensity and economic structures both impact coal consumption in the same direction, the impact of energy intensity is more significant than the economic changes. The expansion of secondary industries in 2001e2006 was primarily caused by a rapid development of ferrous metal, nonmetallic mineral and chemical manufacturing, which are highly energy intensive [34]. To restrict the extensive development and energy consumption of these industries, energy intensity was set as a constraint in the 11th Five Year Plan [35]. In addition, the
Fig. 3. Energy allocation diagram of coal, China 2011.
Fig. 4. The consumption of coal-derived energy carriers expressed in CBPEQ of China.
C. Chong et al. / Energy 85 (2015) 366e378
373
Fig. 5. The coal consumption expressed in CBPEQ of the end-use sectors of China.
government of China also targets to reduce the energy intensity of each industrial subsector by improving the energy efficiency of production equipment by 2010. This led to a great decrease in coal consumption in 2006e2011.
Fig. 6. The coal consumption expressed in CBPEQ of non-energy industrial subsectors of China.
3.2.4. The influence of energy structure in end-use The changing energy structures in end-use sectors contributed to coal consumption growth from 2001e2006 and 2006e2011 due to an increased proportion of coal and coke in the end-use sectors, as shown in Fig. 11. According to the decomposition results, increases in the amount of end-use coal and coke consumption increased the total coal consumption while decreases in the amount of other energy carriers decreased the total coal consumption, as shown in Fig. 12. The consumption growth of end-use coal and coke is primarily due to the rapid expansion of heavy industries, such as non-ferrous metal (iron & steel) and non-metallic material (cement, etc.) industries.
Table 11 LMDI decomposition result of the entire industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCfos
DCtot
2001e2006 2006e2011 2001e2011
4320 3072 7514
58,468 109,552 149,181
3199 2851 2311
15,630 30,517 2171
1171 8194 10,527
2236 8053 9341
1046 20 1345
2217 3765 5311
1775 2427 355
81,159 73,224 154,412
Table 12 LMDI decomposition result of the primary industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCrew
DCtot
2001e2006 2006e2011 2001e2011
158 76 213
2151 2710 4237
1002 416 1295
359 2884 1968
20 719 620
157 299 421
36 2 36
85 116 151
88 45 44
1568 253 1315
Table 13 LMDI decomposition result of the secondary industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCrew
DCtot
2001e2006 2006e2011 2001e2011
3819 2752 6660
51,688 98,142 132,213
4208 3365 2917
15,129 25,195 1435
872 5735 8082
1733 6553 7576
904 12 1168
2016 3441 4808
1470 2146 202
74,342 65,939 140,296
374
C. Chong et al. / Energy 85 (2015) 366e378
Table 14 LMDI decomposition result of the tertiary industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCrew
DCtot
2001e2006 2006e2011 2001e2011
342 243 641
4628 8698 12,729
6 930 689
142 2437 1638
278 1739 1825
345 1200 1342
106 5 139
114 207 351
216 235 108
5247 7538 12,800
Fig. 7. The LMDI additive decomposition result of the entire industry's coal consumption in China.
3.2.5. The influence of energy transformation and end-use efficiency The increased energy transformation efficiency, represented by KPEQ,j, was primarily due to improved electricity supply efficiency and was is the main factor that decreased coal consumption in 2001e2011. During this period, by shutting down coal power plants with low efficiencies and small capacities and building advanced coal power plants with high efficiencies and large capacities, as outlined in Table 15, the coal consumption per unit of supplied electricity was improved continuously. For instance, the coal consumption per unit of supplied electricity of thermal (mainly coal) power plants was improved from 345 gce/kWh in 2008 [36] to 329 gce/kWh in 2011 [37]. It is forecasted to be further reduced to 310 g/kWh by 2020 [38]. Improvements in the end-use combustion efficiency also helped reduce coal consumption. In 2010, the boiler efficiency and kiln efficiency in China were still lower than the average efficiencies for the world [40] due to the following reasons: 1) most of the boiler and kiln capacities in China are quite small and have low-rated
efficiencies, and 2) most of the boiler and kiln are not operated under ideal conditions, leading to low performing efficiencies [41]. Therefore, it remains possible to further improve the end-use combustion efficiency to reduce coal consumption.
3.2.6. The influence of primary energy structure The primary energy structure is represented by a coal component factor that indicates the coal proportion of the total fossil energy that is consumed for electricity, heat and coke production, and the fossil fuel component factor indicates the proportion of fossil energy in the total primary energy that is consumed for electricity, heat and coke production. Both factors increased coal consumption in 2001e2011, but the fossil fuel component factor decreased coal consumption in 2006e2011. The reason lies in that the coal component factor remained unchanged between 2001 and 2011, while the proportion of renewable energy and nuclear power remained unchanged between 2001 and 2006 (from 7.9% to 7.5%) but markedly increased between 2006 and 2011 (from 7.5% to 8.8%). The development of renewable energy and nuclear power was stimulated by the policies that were targeted to increase the proportion of non-fossil energy in the total primary energy consumption in the 11th Five Year plan [35], which seeks to increase this proportion to 11.4% by 2015 and to reach 15% by 2020 [42].
3.3. The elasticity of the influencing factors
Fig. 8. The contribution of main driving forces to economic growth in China.
Based on the LMDI decomposition approach, we can identify the extent to which the total changes of these influencing factors contributed to the total growth of coal consumption. To further identify the sensitivities of these factors, we used the elastic efficiency method to calculate the elasticity of these factors based on equation (4):
C. Chong et al. / Energy 85 (2015) 366e378
375
Fig. 9. Economic structure in China and its impact on coal consumption.
Fig. 10. Energy intensity in China and its impact to coal consumption.
Fig. 11. Energy consumption structure in entire industry of China in SQ.
Fig. 12. The increment of coal consumption in China caused by DCmix expressed according to fuel type.
376
C. Chong et al. / Energy 85 (2015) 366e378
Table 15 The capacity of electricity generation units and its electricity supply efficiency in 2010 [39]. Generation unit
Quantity (unit)
Total capacity (GW)
Electricity supplication efficiency (gce/kWh)
Capacity ¼ 1000 MW 600 MW Capacity<1000 MW 300 MW Capacity<600 MW 200 MW Capacity<300 MW 100 MW Capacity<200 MW 6 MW Capacity<100 MW Total
31 361 775 250 463 4491 6371
31.16 224.31 248.57 52.01 60.72 76.72 693.49
293 317 330 348 358 363 331
lk ¼
DCFk
(4)
FkT Fk0
where lk is the elastic efficiency of the kth influencing factor; FTk and F0k are the values of the kth influencing factor at time T and time 0, respectively; and DCFk is the increment of coal consumption caused by the kth influencing factor. Table 16 lists the elasticity of each influencing factor. For the end-use energy structure, we include only coke, heat and electricity in the calculation, considering their relative importance.
The elasticity results provide a quick impression of the effect of these factors on coal consumption growth. For example, in 2001e2001, each person contributed to an increment of 1 tce in coal consumption growth, which increased the GDP per capita by 1 RMB per capita and resulted in a total coal consumption growth of 75,100 tce. In addition, the per capita increment also increased the proportion of secondary industries by 1 percentage point, resulting in a total coal consumption growth of 23.80 Mtce. These results also highlight the importance of the proportion, intensity and energy structure of secondary industries compared to primary and tertiary industries and the importance of energy source
Table 16 The elasticity of the influencing factors on coal consumption growth. FTk
FTk F0k
Unit
DCFk (100 Mtce)
P Q
1.253 8622
1.331 28,488
0.78 19,866
billion persons RMB per capita
0.752 14.923
0.97 7.51
Primary industries
S1
0.146
0.101
0.045
0.130
2.90
1 tce/1 person 10 thousand tce/RMB per capita 1 Mtce/Percentage point
Secondary industries
S2
0.458
0.470
0.012
0.292
23.83
1 Mtce/Percentage point 1 Mtce/Percentage point
Population GDP per capita Proportion
Tertiary industries
S3
0.410
0.438
0.027
Primary industries Secondary industries Tertiary industries Primary industries
I1 I2 I3 M1,coke
0.1794 0.8065 0.2199 0.047
0.1029 0.8170 0.1886 0.014
0.0764 0.0105 0.0314 0.033
Secondary industries
M2,coke
0.231
0.252
0.021
Tertiary industries
M3,coke
0.006
0.000
0.006
Primary industries
M1,heat
0.001
0.001
Secondary industries
M2,heat
0.107
0.054
Tertiary industries
M3,heat
0.016
0.016
Proportion of electricity in end use consumption
Primary industries
M1,electricity
0.356
0.334
0.022
Secondary industries
M2,electricity
0.316
0.291
0.025
Tertiary industries
M3,electricity
0.195
0.213
0.017
Primary energy quantity converted factor
Coke
KPEQ,coke
1.073
1.102
0.029
Heat
KPEQ,heat
1.192
1.316
0.124
Electricity
KPEQ,electricity
3.040
2.563
0.477
Coke
Kcoal,coke
1.000
1.000
0.0002
Heat
Kcoal,heat
0.843
0.906
0.063
Electricity
Kcoal,electricity
0.952
0.967
0.014
Heat
Kfossil,heat
1.000
0.886
0.114
Kfossil,electricity
0.798
0.813
0.015
Kcombustion
1.710
1.460
0.250
Energy intensity
Proportion of coke in end use consumption
Proportion of heat in end use consumption
Coal component factor
Fossil fuel component factor
lk
F0k
kth Influencing factors
Electricity
End-use combustion factor
0.0001 0.053 0.0001
e e e tce/10,000 RMB tce/10,000 RMB tce/10,000 RMB e e e e e e e e e e e e e e e e e e
0.069
2.52
0.197 0.144 0.164 0.012
2.58 13.63 5.23 0.35
Unit
100 tce/tce per 10,000 RMB 100 tce/tce per 10,000 RMB 100 tce/tce per 10,000 RMB 1 Mtce/Percentage point
0.192
9.00
1 Mtce/Percentage point
0.009
1.55
1 Mtce/Percentage point
0.000
0.36
1 Mtce/Percentage point
0.412
7.86
1 Mtce/Percentage point
0.000
1.92
1 Mtce/Percentage point
0.016
0.72
1 Mtce/Percentage point
0.434
17.26
1 Mtce/Percentage point
0.068
3.97
1 Mtce/Percentage point
0.059
2.01
1 Mtce/0.01 factor change
0.063
0.51
1 Mtce/0.01 factor change
1.075
2.25
1 Mtce/0.01 factor change
0.000
2.18
1 Mtce/0.01 factor change
0.046
0.73
1 Mtce/0.01 factor change
0.092
6.57
1 Mtce/0.01 factor change
0.077
0.68
1 Mtce/0.01 factor change
0.113
7.82
1 Mtce/0.01 factor change
0.532
2.13
1 Mtce/0.01 factor change
C. Chong et al. / Energy 85 (2015) 366e378
structures for electricity production compared to those of coke and heat. 3.4. Uncertainties Although the methods and data used in this work represent the best attempts of the authors, uncertainties do exist. For example, we assumed that all of the end-use fuels except coke are combusted at an averaged combustion efficiency estimated from several studies, which may introduce some uncertainties into the results. However, this combustion efficiency considers the importance of using end-use efficiency for total energy consumption [43]. If more indicators of end-use energy efficiency are introduced in the future, such as the efficiencies of electric motors and other devices, the LMDI method could be further expanded to incorporate the impact of end-use efficiencies of primary energy consumption. Another uncertainty is that the coal preparation rate provided in the statistical data is lower than the actual rate. We adopted this adjusted data for mapping the Sankey diagram, but we did not considered it in the LMDI decomposition. Furthermore, the LMDI decomposition approach adopted in this study possesses a limitation in that it can only analyze the influencing factors of coal consumption in the entire industry, which does not included residential consumption. In the future, if residential consumption is to be considered, the LMDI decomposition approach should be applied to residential coal consumption independently. 4. Conclusions and suggestions In this work, we mapped an energy allocation diagram of the coal use in China to present the energy balance of coal flows from raw coal supply to end-use sectors in the years 2001, 2006 and 2011. We conclude that the main features of coal use systems in China from 2001 to 2011 are: 1) China is a growing coal importer, but coal supply still relies on domestic production; 2) China's coal transformation is quite diversified in electricity and heat generation, coking and end-use combustion; and 3) the industrial sector (manufacturing), especially heavy industries, such as iron and steel and cement industries, are responsible for the majority of coal consumption and its growth in China. Based on the technical influencing factors derived from the mapping process, we developed an LMDI decomposition approach of coal consumption growth and applied it to analyze the influencing factors of coal consumption growth in China's entire industry. The results indicated that: 1) the growth of GDP per capita is the major factor driving coal consumption growth in China from 2001 to 2011, which is mainly supported by the expansion of coal intensive industries; 2) the changes in economic structure and energy intensity first increase coal consumption and then reduce coal consumption, which was facilitated by policy adjustments aimed at enhancing structural changes and energy savings after 2006; 3) the improvement of the energy transformation efficiency is the primarily factor responsible for reducing coal consumption from 2001 to 2011, which was primarily realized by the improvement in the electricity supply efficiency, and the improvement in the end-use combustion efficiency also helped reduce coal consumption; and 4) the improvement in the primary energy structure, especially the proportion of non-fossil energy, began to reduce coal consumption only from 2006 due to policy adjustments. The elasticity of these factors is further analyzed to observe their sensitivity, which highlights the importance of the proportion, intensity and energy structures of secondary industries and also the importance of the energy source structures for electricity generation.
377
The policy implications of this study are summarized as follows for controlling total coal consumption in China. First, the prioritized solution is to change the mode of economic growth, as this study shows that a rapid GDP growth based on energy intensive production and consumption will bring about a rapid growth of coal consumption. The emphasis is to reduce the proportion and energy intensity of secondary industries throughout the entire economy, especial heavy industries, and to improve the economy's energy structure by promoting energy that is derived from sources other than coal. Second, the energy transformation efficiency and the end-use combustion efficiency of coal must be continually improved to ensure that coal is used at the best available efficiency, which has already played an important role in reducing coal consumption from 2001 to 2011. Third, the utilization scale of renewable energy, nuclear and natural gas must be rapidly increased to substitute coal, especially for electricity generation. While the total amount of coal needs to be controlled, the majority of new electricity demands should be satisfied by new non-coal electricity generation. Acknowledgements The authors gratefully acknowledge the financial support from the BP Company regarding the Phase II Collaboration between BP and Tsinghua University, and the support from Tsinghua-Rio Tinto Joint Research Centre for Resources, Energy and Sustainable Development for the work on the energy allocation diagram of coal flows. References [1] BP. BP statistical review of world energy June 2013. BP; 2013. [2] China State Council. Energy conservation and emission reduction regulation and plan in 12th five-year plan. The Central People's Government of the People's Republic of China; 2011. [3] China State Council. Energy development regulation and plan in 12th fiveyear plan. The Central People's Government of the People's Republic of China; 2013. [4] Chinese State Council. Chinese State Council plan to restrict China energy consumption. The Central People's Government of the People's Republic of China; 2013. [5] International Energy Agency (IEA). World energy outlook 2013. International Energy Agency; 2013. [6] BP. BP energy outlook 2035. BP; 2014. [7] Fang WZ. The government should restrict China energy consumption. China Econ Trade Her 2008;2:14e6. [8] Ni WD, Chen Z, Li Z. China current energy utilization and key energy strategy. Energy China 2008;12:6e9. [9] Ni WD. Controlling total energy production and consumption in imperative. Sci Technol Rev 2008:24. [10] Ang BW, Zhang FQ. A survey of index decomposition analysis in energy and environmental studies. Energy 2000;25(12):1149e76. [11] Ang BW. Decomposition analysis for policymaking in energy: which is the preferred method? Energy Policy 2004;32(9):1131e9. [12] Ang BW, Liu FL. A new energy decomposition method: perfect in decomposition and consistent in aggregation. Energy 2001;26(6):537e48. [13] Ang BW. The LMDI approach to decomposition analysis: a practical guide. Energy Policy 2005;33(7):867e71. [14] Wang W, Liu X, Zhang M, Song X. Using a new generalized LMDI (logarithmic mean Divisia index) method to analyze China's energy consumption. Energy 2014;67(0):617e22. [15] Fernandez Gonzalez P, Landajo M, Presno MJ. The Divisia real energy intensity indices: evolution and attribution of percent changes in 20 European countries from 1995 to 2010. Energy 2013;58:340e9. [16] Chontanawat J, Wiboonchutikula P, Buddhivanich A. Decomposition analysis of the change of energy intensity of manufacturing industries in Thailand. Energy 2014;77(0):171e82. [17] Cullen JM, Allwood JM. Theoretical efficiency limits for energy conversion devices. Energy 2010;35(5):2059e69. [18] Cullen JM, Allwood JM. The efficient use of energy: tracing the global flow of energy from fuel to service. Energy Policy 2010;38(1):75e81. [19] Ma LW, Allwood JM, Cullen JM, Li Z. The use of energy in China: tracing the flow of energy from primary source to demand drivers. Energy 2012;40(1): 174e88.
378
C. Chong et al. / Energy 85 (2015) 366e378
[20] Yu FF. Coal flow diagram based on coal supply chain: methodology and application. Beijing, China: Tsinghua University; 2013. [21] Yu FF, Ma LW, Jiang LX, Li Z, Ni WD. Mapping the energy flows of coal utilization in China: the methodology and case studies for they year 2005 and 2010. Adv Mater Res 2013;724e725:1234e41. [22] Pu HJ. The sustainable utilization of coal in China. China University of Mining and Technology Press; 2010. [23] Ma J. Present status in China of coal preparation industry & consideration on its development in the twelfth five-year plan. China Coal Process Util 2011;4: 1e5. [24] Chong CH, Ma LW, Li Z, Geng J, Zhang TK. A programmed data-processing method for mapping energy allocation Sankey diagram of China. In: 2014 International conference on power and energy. Shanghai, China: Taylor & Francis Group; 2014. [25] China's National Bureau of Statistics. China energy statistical yearbook 2002. Beijing: China Statistics Press; 2003. [26] China's National Bureau of Statistics. China energy statistical yearbook 2007. Beijing: China Statistics Press; 2008. [27] China's National Bureau of Statistics. China energy statistical yearbook 2012. Beijing: China Statistics Press; 2013. [28] China's National Bureau of Statistics. China statistical yearbook 2002. Beijing: China Statistics Press; 2003. [29] China's National Bureau of Statistics. China statistical yearbook 2007. Beijing: China Statistics Press; 2008. [30] China's National Bureau of Statistics. China statistical yearbook 2012. Beijing: China Statistics Press; 2013. [31] National Development and Reform Commission of China (NDRC). Middle and long term energy conservation plan. Beijing: The Central People's Government of the People's Republic of China; 2004.
[32] National Development and Reform Commission of China (NDRC). Major energy conservation project in eleventh five year plan in China. Beijing: The Central People's Government of the People's Republic of China; 2006. [33] Ma LW, Liu P, Fu F, Li Z, Ni WD. Integrated energy strategy for the sustainable development of China. Energy 2011;36(2):1143e54. [34] Zhao X, Ma C, Hong D. Why did China's energy intensity increase during 1998e2006: decomposition and policy analysis. Energy Policy 2010;38(3): 1379e88. [35] National Development and Reform Commission of China (NDRC). Energy development plan in 11th five year plan. Beijing: The Central People's Government of the People's Republic of China; 2007. [36] Department of Regulation and Statistic in CEC. Statistical data of electricity in China 2009. China Electricity Council (CEC); 2010. [37] Department of Regulation and Statistic in CEC. Statistical data of electricity in China 2012. China Electricity Council (CEC); 2013. [38] Yang Y, Yang Z, Xu G, Wang N. Situation and prospect of energy consumption for China's thermal power generation. Proc CSEE 2013;33:1e11. [39] Wang Y, Gu A, Zhang A. Recent development of energy supply and demand in China, and energy sector prospects through 2030. Energy Policy 2011;39(11): 6745e59. [40] Yu J. Status and transformation measures of industrial coal-fired boiler in China. Clean Coal Technol 2012;3:89e91. 113. [41] Xie Q. Analysis of coal-fired industrial boiler energy consumption status and energy-saving measures. Value Eng 2011;34:45e6. [42] Li Z, Chang SY, Ma LW, Liu P, Zhao LX, Yao Q. The development of low-carbon towns in China: concepts and practices. Energy 2012;47(1):590e9. [43] Sun JX, Ma LW, Ni WD, Li Z. Reducing China's energy consumption and CO2 emissions by more efficient use of energy. In: International conference on materials for renewable energy and environment 2013; 2013. p. 802e8.
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Logarithmic mean Divisia index (LMDI) decomposition of coal consumption in China based on the energy allocation diagram of coal flows ChinHao Chong, Linwei Ma*, Zheng Li, Weidou Ni, Shizhong Song State Key Laboratory of Power Systems, Department of Thermal Engineering, Tsinghua-BP Clean Energy Centre, Tsinghua University, Beijing 100084, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 December 2014 Received in revised form 17 March 2015 Accepted 20 March 2015 Available online 2 May 2015
This manuscript attempted to analyze the influencing factors of coal consumption growth in China using the logarithmic mean Divisia index (LMDI) decomposition method developed based on the physical processes of coal utilization from raw coal to the end-use sectors. By mapping the energy allocation diagram of coal flows, we built a method to balance the energy allocation of coal flows and derived several technical influencing factors. These factors were used to develop an LMDI decomposition method suitable for analyzing the coal consumption growth of complex coal-use systems, such as that of China. The method is subsequently applied to analyze the influencing factors of China's coal consumption growth from 2001 to 2011. The results indicate the rapid growth of GDP (gross domestic production) per capita, which heavily relied on the expansion of heavy industry as the dominant factor driving coal consumption growth. Improvement in the energy efficiency of coal power generation and coal end-use combustion were the primary factors reducing coal consumption. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Coal consumption LMDI Sankey diagram Coal flow Influencing factor
1. Introduction China now is the largest coal producer and coal consumer in the world, accounting for 46.6% of the global coal production and 50.3% of the global coal consumption in 2013 [1]. Faced with challenges of global climate change and domestic environment protection, controlling the rapidly increasing coal consumption in China has become an important issue, not only for China but also for the world [2e4]. Referring to IEA and BP, policy choices in China, which has outlined plans to cap the share of coal regarding total energy use, will be particularly important as China now consumes as much coal as the rest of the world combined [5]. In addition, the trends of global coal growth before 2035 will be greatly influenced by China's profile [6]. However, most of the discussions presented in previous studies have focused on the coal needs and targets for controlling the total coal consumption [7e9]. Quantitative analyses on the influencing factors of coal consumption growth that consider the rapid and dynamical development of energy systems in China are lacking, which are essential for developing effective policies to control coal consumption and deserve further study.
* Corresponding author. E-mail addresses: [email protected], [email protected] (L. Ma). http://dx.doi.org/10.1016/j.energy.2015.03.100 0360-5442/© 2015 Elsevier Ltd. All rights reserved.
The index decomposition analysis (IDA) method has been widely applied to analyze the influencing factors of energy consumption growth, including those based on the Laspeyres index and the Divisia index [10]. Ang et al. [11e13] presented a review of the development and applications of IDA methods and recommended the logarithmic mean Divisia index I (LMDI) method because it is robust and convenient for many applications. Many studies have utilized the LMDI method to decompose the total energy consumption growth of various regions and countries [14e16], and the commonly considered factors are population growth, GDP (gross domestic production), economic structure, energy intensity and energy mix. However, to apply the LMDI method to coal consumption growth in China, for which research is lacking, the factors that are considered need to be expanded because coal is only one part of the energy system. In addition, China consumes a huge amount of coal, and the uses of coal have been diversified for various sectors, including power and heat generation, coking and end-use combustion, etc. Therefore, the structural changes through energy use systems, including the stages of primary energy sources, energy conversion and energy end-use, in addition to technically detailed factors such as energy conversion efficiency and energy end-use efficiency, should be carefully considered to further develop the LMDI method to decompose the technical details of coal consumption growth.
C. Chong et al. / Energy 85 (2015) 366e378
Abbreviations Subscript i ith industry involved Subscript j jth energy involved Subscript k kth influencing factors involved IDA index decomposition analysis LMDI logarithmic mean Divisia index SQ standard quantity PEQ primary energy quantity CBPEQ coal-based primary energy quantity ESQ,j jth energy expressed in SQ ESQ,ij jth energy expressed in SQ in the ith industry EPEQ,j jth energy expressed in PEQ ECBPEQ,j jth energy expressed in CBPEQ C total raw coal consumption C0 total raw coal consumption at time 0 CT total raw coal consumption at time T P population GDP gross domestic production GDPi value added of the ith sector Q GDP per capita Si proportion of the ith industry Ii energy intensity of the ith industry Mij proportion of the jth fuel in the ith industry Kcombustion,j end-use combustion factor of energy j KPEQ,j primary energy quantity converted factor of energy j
This study aims to develop an LMDI method suitable for analyzing the coal consumption growth of countries with complex coal use systems, such as in China, and to apply the method to analyze the influencing factors of coal consumption growth in China. First, by mapping China's coal flows to Sankey diagrams, we studied the physical processes of coal use from the raw coal supply to the end-uses as electricity, heat and coke, and thus derive key influencing factors that need to be further considered. Then, using these key influencing factors, we develop an LMDI method suitable for analyzing such coal use systems and apply it the coal consumption growth in China from 2001 to 2011, during which China's coal consumption increased by an average growth ratio of 12.5%. The contents of this paper are organized as follows: an introduction of the methodology and data input are described in Section 2, the results and discussion are given in Section 3, and conclusions and suggestions are summarized in Section 4. 2. Methodology and data input 2.1. Sankey diagram and technical influencing factors Sankey diagrams are popular in energy system analyses [17e19] and coal system analyses [20e22]. In this work, an energy allocation Sankey diagram, which presents the energy balance from raw coal supplies to the final end-uses, is used as a tool to understand the physical process of coal systems and to derive the key influencing factors of coal consumption growth. 2.1.1. Diagram structure and energy balance In this study, the energy allocation Sankey diagram of coal was divided into three stages: 1.) raw coal supply 2.) coal transformation and 3.) end-use sector. The diagram was mapped according to the first law of thermodynamics. The width of the flow signified the quantity of the energy, and the colour of the flow signified the type of energy.
Kcoal,j Kfossil,j DCpop
DCaff DCstr DCint DCmix DCcom DCpeq DCcoal DCfos F0k FTk DCFk
lk
367
coal component factor of energy j fossil fuel component factor of energy j increment of raw coal consumption caused by change of P increment of raw coal consumption caused by change of Q increment of raw coal consumption caused by changes of Si increment of raw coal consumption caused by changes of Ii increment of raw coal consumption caused by changes of Mij increment of raw coal consumption caused by changes of Kcombustion,j increment of raw coal consumption caused by changes of KPEQ,j increment of raw coal consumption caused by changes of Kcoal,j increment of raw coal consumption caused by changes of Kfossil,j value of the kth influencing factor at time 0 value of the kth influencing factor at time T increment of raw coal consumption caused by the kth influencing factor elastic efficiency of the kth influencing factor
No energy loss is reflected in energy allocation Sankey diagrams. Therefore, we trace the energy of raw coal as it is exploited. After production and importation, the raw coal is processed and converted into secondary energy sources, such as coke, heat and electricity, in transformation sector and finally consumed at the end-use sectors. By compensating for all of the energy losses that occur during energy processing and transformation, we can express the secondary energy consumption as a coal-based primary energy quantity (CBPEQ), which indicates the amount of raw coal consumption required to produce a secondary energy source. 2.1.2. Original data e energy balance table The Energy Balance of China and the Final Energy Consumption by Industrial Sector, are two statistical tables in the China Energy Statistical Yearbook, are used as the original data sources for mapping the Sankey diagram. The data in the Energy Balance of China is vertically divided into 6 parts: A) primary energy supply, B) input () and output (þ) of the transformation, C) loss, D) final consumption, E) statistical difference, and F) total energy consumption, and horizontally divided into 30 types of energy, including raw coal, various coal products, crude oil, various oil products, natural gas, electricity, heat and recovery energy. The details of the final energy consumption of industrial sectors are further presented in another table named Final Energy Consumption by Industrial Sector, which includes more than 40 subsectors. To illustrate the structure of the final consumption of the industrial sectors in the diagram, we re-categorized the subsectors in the industrial sector, as shown in Table 1. Because the coal preparation ratio according to the Energy Balance of China is much lower than the actual data, we adjusted the ratio according to another study [23] in the mapping. However, it became too complex to apply this adjusted ratio because we are unable to obtain more accurate information to estimate the losses in the coal preparation process after the revision. Hence, the original ratio is kept in the LMDI decomposition.
368
C. Chong et al. / Energy 85 (2015) 366e378
Table 1 The re-categorized industrial subsectors. Industrial subsector 1 Non-energy industrial subsector 1.1 Non-energy mining Mining and processing of ferrous metal ores Mining and processing of non-ferrous metal ores Mining and processing of non-metal ores Mining of other ores 1.2 Smelting and pressing of ferrous metals 1.3 Smelting and pressing of non-ferrous metals 1.4 Manufacture of non-metallic mineral and chemical products 1.5 Manufacture of raw chemical materials and chemical products 1.6 Manufacture of food and beverages Processing of food from agricultural products Manufacture of food Manufacture of liquor, beverages and refined tea 1.7 Manufacture of textiles Manufacture of textiles Manufacture of textiles, wearable apparel and accessories 1.8 Manufacture of machinery and vehicles Manufacture of general purpose machinery Manufacture of special purpose machinery Manufacture of automobiles Manufacture of railways, ships, aerospace and other transportation equipment 1.9 Manufacture of paper and paper products 1.10 Other non-energy industrial subsector Manufacture of tobacco Manufacture of leather, fur, feather and related products and footwear Processing of timber, manufacture of wood, bamboo, rattan, palm and straw products Manufacture of furniture Printing and reproduction of recording media Manufacture of articles for culture, education, arts and crafts, sports and entertainment activities Manufacture of medicines Manufacture of chemical fibres Manufacture of rubber and plastics products Manufacture of metal products Manufacture of electrical machinery and apparatuses Manufacture of computers, communication and other electronic equipment Manufacture of measuring instrument and machinery Utilization of waste resources Repair service of metal products, machinery and equipment Production and supply of water Others manufacture 2 Energy industry subsector Production and supply of electric power and heat power Production and supply of gas Mining and preparation of coal Extraction of petroleum and natural gas Processing of petroleum, coking and Processing of nuclear fuel
2.1.3. Data processing and diagram mapping In our previous study [24], we presented a method that can quickly generate the data required for mapping an energy allocation diagram of the coal in China, which is based on a series of standardized steps and rigid equations that can be programmed. The original data from the statistical tables are processed using this method to prepare a balanced dataset for the mapping. We define 9 energy types related to coal use in the mapping, as illustrated in Table 2, and assume that all end-use fuels are burned at an average combustion efficiency except coke. 2.1.4. Technical influencing factors derived in the mapping process In the process of mapping the energy allocation diagram of coal, we [24] defined 4 factors to express the end-use energy carriers in the CBPEQ type, as shown in equation (1). These factors include: 1) the end-use combustion factor, 2) the primary energy quantity converted factor, 3) the coal component factor, and 4) the fossil fuel component factor, as illustrated in Table 3. These factors are used to develop the LMDI decomposition method of coal consumption growth in Section 2.2.
ECBPEQ ;j ¼ ESQ ;j $Kcombustion;j $KPEQ ;j $Kcoal;j $Kfossil;j
(1)
2.2. LMDI decomposition approach Most previous studies have adopted equation (2) to express the energy consumption as standard quantity (ESQ) and consider influencing factors such as population (P), GDP per capita (GDP/P), economic structure (GDPi/GDP), energy intensity (ESQ,i/GDPi) and Table 2 The 9 coal-derived energy types involved in the energy allocation diagram of coal. No.
Energy types
1 2 3 4 5 6 7 8 9
Raw coal Cleaned coal: cleaned coal, other washed coal Coke Gas: coke oven gas, other gas Briquettes Gangue Other coking products Heat Electricity
C. Chong et al. / Energy 85 (2015) 366e378
369
Table 3 Description of the technical influencing factors. Factor
Description
Kcombustion,j
1 To consider the impact of the end-use combustion efficiency, we assume that all of the fuels except coke, heat and electricity consumed in the end-use sector are converted into heat, with an average end-use combustion efficiency, 4. 2 Kcombustion,j is defined as the unit of energy input to the boiler or furnace required to produce 1 unit of heat. 3 Kcombustion of coke, heat and electricity are 1. 1 The primary energy quantity type transformation factor, defined as the total number of units of primary energy that are consumed to produce 1 unit of secondary energy. This value is used to amplify secondary energy to primary energy required to produce the secondary energy. 2 As the processing and transportation losses can be neglected, KPEQ of raw coal, natural gas, crude oil and recovery gas are set as 1. 1 The coal component factor, representing the share of coal in primary energy that is consumed to produce 1 unit of secondary energy. 2 Kcoal of raw coal is 1. Kcoal values of natural gas, crude oil and recovery gas are 0.
KPEQ,j
Kcoal,j
Kfossil,j
Equation
1 The Kcoal of heat and electricity should be revised using the fossil fuel component factor Kfossil, as the calculated processes only considered the transformation sectors mentioned in the China Energy Balance and did not include renewable energy and nuclear power. 2 Kfossil shows the share of fossil fuel consumed to produce 1 unit of secondary energy. 3 Kfossil of other energy sources are 1.
energy mix (ESQ,ij/ESQ,i), in which i represents economic sectors and j represents energy types.
ESQ ¼
X
P$
ij
GDP GDPi ESQ ;i ESQ ;ij $ $ $ P GDP GDPi ESQ ;i
(2)
Further considering the technical influencing factors derived in Section 2.1.4, we expand equation (2) to equation (3) to express the total amount of raw coal consumption of China's entire industry, including primary industries, secondary industries, and tertiary industries, in the CBPEQ. The limitation of this method is that it does not include the residential consumption of raw coal, but fortunately this is generally a small portion (for instance, 10.7% in 2011 for China) of the entire raw coal consumption. The symbol items in equation (3) are explained in Table 4. The LMDI formulae for decomposing coal consumption growth in the entire industry of China are presented in Table 5.
ECBPEQ ¼
X ij
P$
GDP GDPi ESQ ;i ESQ ;ij $ $ $ $K $K P GDP GDPi ESQ ;i combustion;j PEQ ;j
(3)
$Kcoal;j $Kfossil;j 2.3. Data input for LMDI decomposition We obtained the energy data from the China Energy Statistical Yearbook [25e27] and the economic data from the China Statistical Yearbook [28e30]. The end use combustion efficiency referred to Refs. [31,32,22], as shown in Table 6. Using our data-processing method previously developed [24], we calculate Kcombustion (Table 7), KPEQ (Table 8), Kcoal (Table 9) and Kfossil (Table 10) of each fuel in 2001, 2006 and 2011. 3. Results and discussions 3.1. Energy allocation diagram of coal use in China The energy allocation diagrams of coal in China for 2001, 2006 and 2011 are presented in Figs. 1e3. Referring to these diagrams,
Kcombustion;j ¼ f1 4: end-use combustion efficiency in average
P KPEQ ;j ¼
n
KPEQ;n $SQn;input
P
SQj;output
j
KPEQ,n: PEQ transformation factor of the nth fuel input to produce fuel j SQn,input: Input of the nth fuel to produce fuel j expressed in SQ SQj,ouput: Output of fuel j expressed in SQ P Kcoal;j ¼
Kcoal;n $PEQn;input
n
PEQj;output
Kcoal,n: Coal component factor of the nth fuel input to produce fuel j PEQn,input: Input of the nth fuel to produce fuel j expressed in PEQ type PEQn,output: Output of fuel j expressed in PEQ type SQ
fossil;j Kfossil;j ¼ SQfossil;j þSQ nonfossil;j
Subscript j: the jth fuel type involved SQfossil,j: Fuel j produced by using fossil fuel expressed in SQ SQnon-fossil,j: Fuel j produced by using non-fuel fuel expressed in SQ
the main features and dynamics of China's coal use system are as follows: (1) Coal supply primarily relies on domestic production. Although coal import in China increased rapidly from 2009 after it became a net coal importer, the self-satisfaction ratio of coal was still as high as 96.8% in 2011. (2) The structure of coal transformation is quite diversified. From 2001 to 2011, the proportion of coal used for electricity generation changed from 55.1% to 47.3%, for coking from 13.4% to 16.4%, for heat generation from 6.8% to 4.6%, and for end-use combustion from 15.1 to 31.7%, as outlined in Fig. 4. During this period, the trend for coal use shifted from electricity generation and heat generation toward end-use combustion and coking. (3) The end-use of coal-derived energy carriers is primarily consumption, and a greater percentage of coal being consumed by the industrial sector (manufacturing), which was responsible for 60.7% and 68.3% of total coal consumption in 2001 and 2011, especially heavy industrial sectors, as shown in Fig. 5. Referring to Fig. 6, coal consumption for the manufacturing of ferrous metals, non-ferrous metals, non-metallic mineral and chemicals increased from 35% to 51% between 2001 and 2011.
3.2. LMDI decomposition results Through LMDI decomposition, we acquire the increment of consumption of coal of China's entire industry and its primary industry, secondary industry and tertiary industry caused by each influencing factor between 2001 and 2006, 2006e2011 and 2001e2011. The decomposition results are given in Tables 11e14, and the summarized the decomposition results of the entire industry are shown in Fig. 7. The growth of GDP per capita is the dominant factor driving the coal consumption growth in 2001e2011, whereas the influences of other factors are relatively small. In the following sections, we discuss each of these influencing factors.
370
C. Chong et al. / Energy 85 (2015) 366e378
Table 4 The symbol items in equation (3). Items
Descriptions
Subscription i
1 Primary industries 2 Secondary industries 3 Tertiary industries Raw coal, cleaned coal, briquettes, gangue, coke, coke oven gas, blast furnace gas, converter gas, other gas, other coking products, crude oil, oil products (gasoline, kerosene, diesel oil, fuel oil, naphtha, lubricants, paraffin waxes, white spirit, bitumen asphalt, petroleum coke LPG, refinery gas and other petroleum products), natural gas, LNG, heat and electricity. Population Gross domestic production Value added of the ith sector Energy consumption of the ith sector expressed in SQ Fuel j consumption of the ith sector expressed in SQ Combustion factor of fuel j Primary energy quantity type conversion factor of fuel j Coal component factor of fuel j Fossil fuel component factor of fuel j
Subscription j
P GDP GDPi ESQ,i ESQ,ij Kcombustion,j KPEQ,j Kcoal,j Kfossil,j
Table 5 LMDI formulae for decomposing coal consumption growth. IDA identify
C¼
P
Cij ¼
PP
ij
Change scheme
i
j
E
E
GDPi SQ ;i SQ ;ij P$GDP P $ GDP $GDPi $ ESQ ;i $Kcombustion;j $KPEQ ;j $Kcoal;j $Kfossil;j
DCtot ¼ C T C 0 ¼ DCpop þ DCaff þ DCstr þ DCInt þ DCmix þ DCcom þ DCpeq þ DCcoal þ DCfos P P CijT Cij0 T DCpop ¼ ln PP 0 ln C T ln C 0
LMDI formulae
ij
ij
Q ¼ GDP P
P
DCaff ¼
i Si ¼ GDP GDP
DCstr ¼
ij E
SQ ;i Ii ¼ GDP i
DCint ¼
QT Q0
CijT Cij0 ln CijT ln Cij0
ln
STi S0i
CijT Cij0 ln CijT ln Cij0
ln
IiT Ii0
P ij
Mij ¼
ESQ ;ij Ei;SQ
DCmix ¼
P
DCcom ¼
ln
CijT Cij0 ln CijT ln Cij0
ln
CijT Cij0 ln CijT ln Cij0
ln
P ij
KPEQ,j
P
DCpeq ¼
ij
Kcoal,j
DCcoal ¼
Kfossil,j
DCfos ¼
P ij
P ij
!
CijT Cij0 ln CijT ln Cij0
ij
Kcombustion,j
ln
ij
P
ij
CijT Cij0 ln CijT ln Cij0
CijT Cij0 ln CijT ln Cij0
CijT Cij0 ln CijT ln Cij0
! !
MijT Mij0
! T Kcombustion;j 0 Kcombustion;j
! T KPEQ;j 0 KPEQ;j
! ln
T Kcoal;j 0 Kcoal;j
! ln
T Kfossil;j 0 Kfossil;j
Note: the superscripts 0 and T give the parameter in period 0 and period T, respectively.
Table 8 KPEQ of each fuel type. Table 6 The end-use combustion efficiency 4. Fuel type
2001
2006
2011
End use combustion efficiency
0.477
0.514
0.552
Note: the combustion efficiency of the furnace and kiln were 62.5% and 35% in 2005 [22]; and 67.5% and 37% in 2010 [31], respectively. We assumed that the combustion efficiency increased linearly from 2001 to 2011, and the ratio of the furnace and kiln remained 1.33 [32].
Fuel type
2001
2006
2011
Raw coal Cleaned coal Briquette Gangue
1.00 1.06 1.01
1.00 1.04 1.11
e
e
1.00 1.07 1.15 1.07
Coke Coke oven gas Blast furnace gas
1.07 1.07
1.06 1.06
e
e
Converter gas
Fuel type
2001
2006
2011
Other gas Other coking products Crude oil Oil products Natural gas LNG
Coke, heat and electricity Others
1.00 1.71
1.00 1.57
1.00 1.46
Heat Electricity
Table 7 Kcombustion of each fuel type.
e
e
1.31 1.09 1.00 1.02 1.00
1.56 1.07 1.00 1.03 1.00
e
e
1.19 3.04
1.35 2.86
1.10 1.11 1.00 1.00 1.43 1.11 1.00 1.03 1.00 1.16 1.32 2.56
C. Chong et al. / Energy 85 (2015) 366e378 Table 9 Kcoal of each fuel type. Fuel type
2001
2006
2011
Heat Electricity Other gas
0.84 0.95 1.00
0.91 0.97 0.94
0.91 0.97 0.95
Note: (1) raw coal, cleaned coal, briquette, gangue, coke, coke oven gas, other coking products: 1.00 or almost 1.00; (2) blast furnace gas, converter gas, crude oil, oil products, natural gas, LNG: 0.00 or almost 0.00. Table 10 Kfossil of each fuel type. Fuel type
2001
2006
2011
Heat Electricity Others
1.00 0.80 1.00
1.00 0.83 1.00
0.89 0.81 1.00
371
3.2.1. The influence of population The population growth in China contributed to coal consumption growth during both 2001e2006 and 2006e2011, although the contribution is quite limited. The reason may be that the population growth of China is relative small, with a total increase of 6.2% in 2011 compared to 2001 due to the rigid population restriction policy that have been in place since the 1970s in China.
3.2.2. The influence of GDP per capita The increase in GDP per capita, whose average annual growth rate was approximately 10%, is the main contributor of coal consumption growth from 2001 to 2011. The growth of the GDP per capita will facilitate additional services that are required for people, thus promoting more energy consumption, especially during this special development stage of China. Currently, China is
Fig. 1. Energy allocation diagram of coal, China 2001.
Fig. 2. Energy allocation diagram of coal, China 2006.
372
C. Chong et al. / Energy 85 (2015) 366e378
simultaneously experiencing rapid industrialization, urbanization and motorization [33], and the new added demand of services primarily pertain to energy-intensive products, such as houses, cars, machines and infrastructures. These changes have not only rapidly increased of the energy consumption of domestic consumption expenditures, but they have also increased the energy consumption embedded in gross capital formation and exported products. Referring to Fig. 8, gross capital formation was a main driving force of economic growth in China. According to Ma et al. [19], domestic consumption expenditures accounted for 38.8%, gross capital formation (infrastructure construction) accounted for 32.3% and export products accounted for 28.9% of China's total energy consumption in 2005. Therefore, the coal consumption growth from 2001 to 2011 was due largely to rapid GDP growth, which was accompanied by a huge demand for energy-intensive products and a rapid expansion of energy-intensive industries that caused a large increase of coal consumption.
3.2.3. The influence of economic structure and energy intensity Changes in economic structure and energy intensity both increased coal consumption in 2001e2006 but also decreased coal consumption in 2006e2011. This occurred because the proportion and energy intensity of secondary industries increased in 2001e2006, whereas both decreased in 2006e2011. These influencing factors did not change in the primary and tertiary industries, and therefore did not have significant impacts on coal consumption compared to secondary industries, as shown in Figs. 9 and 10. Although energy intensity and economic structures both impact coal consumption in the same direction, the impact of energy intensity is more significant than the economic changes. The expansion of secondary industries in 2001e2006 was primarily caused by a rapid development of ferrous metal, nonmetallic mineral and chemical manufacturing, which are highly energy intensive [34]. To restrict the extensive development and energy consumption of these industries, energy intensity was set as a constraint in the 11th Five Year Plan [35]. In addition, the
Fig. 3. Energy allocation diagram of coal, China 2011.
Fig. 4. The consumption of coal-derived energy carriers expressed in CBPEQ of China.
C. Chong et al. / Energy 85 (2015) 366e378
373
Fig. 5. The coal consumption expressed in CBPEQ of the end-use sectors of China.
government of China also targets to reduce the energy intensity of each industrial subsector by improving the energy efficiency of production equipment by 2010. This led to a great decrease in coal consumption in 2006e2011.
Fig. 6. The coal consumption expressed in CBPEQ of non-energy industrial subsectors of China.
3.2.4. The influence of energy structure in end-use The changing energy structures in end-use sectors contributed to coal consumption growth from 2001e2006 and 2006e2011 due to an increased proportion of coal and coke in the end-use sectors, as shown in Fig. 11. According to the decomposition results, increases in the amount of end-use coal and coke consumption increased the total coal consumption while decreases in the amount of other energy carriers decreased the total coal consumption, as shown in Fig. 12. The consumption growth of end-use coal and coke is primarily due to the rapid expansion of heavy industries, such as non-ferrous metal (iron & steel) and non-metallic material (cement, etc.) industries.
Table 11 LMDI decomposition result of the entire industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCfos
DCtot
2001e2006 2006e2011 2001e2011
4320 3072 7514
58,468 109,552 149,181
3199 2851 2311
15,630 30,517 2171
1171 8194 10,527
2236 8053 9341
1046 20 1345
2217 3765 5311
1775 2427 355
81,159 73,224 154,412
Table 12 LMDI decomposition result of the primary industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCrew
DCtot
2001e2006 2006e2011 2001e2011
158 76 213
2151 2710 4237
1002 416 1295
359 2884 1968
20 719 620
157 299 421
36 2 36
85 116 151
88 45 44
1568 253 1315
Table 13 LMDI decomposition result of the secondary industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCrew
DCtot
2001e2006 2006e2011 2001e2011
3819 2752 6660
51,688 98,142 132,213
4208 3365 2917
15,129 25,195 1435
872 5735 8082
1733 6553 7576
904 12 1168
2016 3441 4808
1470 2146 202
74,342 65,939 140,296
374
C. Chong et al. / Energy 85 (2015) 366e378
Table 14 LMDI decomposition result of the tertiary industry's coal consumption in China (unit: Mtce). Year
DCpop
DCaff
DCstr
DCint
DCmix
DCpeq
DCcoal
DCcom
DCrew
DCtot
2001e2006 2006e2011 2001e2011
342 243 641
4628 8698 12,729
6 930 689
142 2437 1638
278 1739 1825
345 1200 1342
106 5 139
114 207 351
216 235 108
5247 7538 12,800
Fig. 7. The LMDI additive decomposition result of the entire industry's coal consumption in China.
3.2.5. The influence of energy transformation and end-use efficiency The increased energy transformation efficiency, represented by KPEQ,j, was primarily due to improved electricity supply efficiency and was is the main factor that decreased coal consumption in 2001e2011. During this period, by shutting down coal power plants with low efficiencies and small capacities and building advanced coal power plants with high efficiencies and large capacities, as outlined in Table 15, the coal consumption per unit of supplied electricity was improved continuously. For instance, the coal consumption per unit of supplied electricity of thermal (mainly coal) power plants was improved from 345 gce/kWh in 2008 [36] to 329 gce/kWh in 2011 [37]. It is forecasted to be further reduced to 310 g/kWh by 2020 [38]. Improvements in the end-use combustion efficiency also helped reduce coal consumption. In 2010, the boiler efficiency and kiln efficiency in China were still lower than the average efficiencies for the world [40] due to the following reasons: 1) most of the boiler and kiln capacities in China are quite small and have low-rated
efficiencies, and 2) most of the boiler and kiln are not operated under ideal conditions, leading to low performing efficiencies [41]. Therefore, it remains possible to further improve the end-use combustion efficiency to reduce coal consumption.
3.2.6. The influence of primary energy structure The primary energy structure is represented by a coal component factor that indicates the coal proportion of the total fossil energy that is consumed for electricity, heat and coke production, and the fossil fuel component factor indicates the proportion of fossil energy in the total primary energy that is consumed for electricity, heat and coke production. Both factors increased coal consumption in 2001e2011, but the fossil fuel component factor decreased coal consumption in 2006e2011. The reason lies in that the coal component factor remained unchanged between 2001 and 2011, while the proportion of renewable energy and nuclear power remained unchanged between 2001 and 2006 (from 7.9% to 7.5%) but markedly increased between 2006 and 2011 (from 7.5% to 8.8%). The development of renewable energy and nuclear power was stimulated by the policies that were targeted to increase the proportion of non-fossil energy in the total primary energy consumption in the 11th Five Year plan [35], which seeks to increase this proportion to 11.4% by 2015 and to reach 15% by 2020 [42].
3.3. The elasticity of the influencing factors
Fig. 8. The contribution of main driving forces to economic growth in China.
Based on the LMDI decomposition approach, we can identify the extent to which the total changes of these influencing factors contributed to the total growth of coal consumption. To further identify the sensitivities of these factors, we used the elastic efficiency method to calculate the elasticity of these factors based on equation (4):
C. Chong et al. / Energy 85 (2015) 366e378
375
Fig. 9. Economic structure in China and its impact on coal consumption.
Fig. 10. Energy intensity in China and its impact to coal consumption.
Fig. 11. Energy consumption structure in entire industry of China in SQ.
Fig. 12. The increment of coal consumption in China caused by DCmix expressed according to fuel type.
376
C. Chong et al. / Energy 85 (2015) 366e378
Table 15 The capacity of electricity generation units and its electricity supply efficiency in 2010 [39]. Generation unit
Quantity (unit)
Total capacity (GW)
Electricity supplication efficiency (gce/kWh)
Capacity ¼ 1000 MW 600 MW Capacity<1000 MW 300 MW Capacity<600 MW 200 MW Capacity<300 MW 100 MW Capacity<200 MW 6 MW Capacity<100 MW Total
31 361 775 250 463 4491 6371
31.16 224.31 248.57 52.01 60.72 76.72 693.49
293 317 330 348 358 363 331
lk ¼
DCFk
(4)
FkT Fk0
where lk is the elastic efficiency of the kth influencing factor; FTk and F0k are the values of the kth influencing factor at time T and time 0, respectively; and DCFk is the increment of coal consumption caused by the kth influencing factor. Table 16 lists the elasticity of each influencing factor. For the end-use energy structure, we include only coke, heat and electricity in the calculation, considering their relative importance.
The elasticity results provide a quick impression of the effect of these factors on coal consumption growth. For example, in 2001e2001, each person contributed to an increment of 1 tce in coal consumption growth, which increased the GDP per capita by 1 RMB per capita and resulted in a total coal consumption growth of 75,100 tce. In addition, the per capita increment also increased the proportion of secondary industries by 1 percentage point, resulting in a total coal consumption growth of 23.80 Mtce. These results also highlight the importance of the proportion, intensity and energy structure of secondary industries compared to primary and tertiary industries and the importance of energy source
Table 16 The elasticity of the influencing factors on coal consumption growth. FTk
FTk F0k
Unit
DCFk (100 Mtce)
P Q
1.253 8622
1.331 28,488
0.78 19,866
billion persons RMB per capita
0.752 14.923
0.97 7.51
Primary industries
S1
0.146
0.101
0.045
0.130
2.90
1 tce/1 person 10 thousand tce/RMB per capita 1 Mtce/Percentage point
Secondary industries
S2
0.458
0.470
0.012
0.292
23.83
1 Mtce/Percentage point 1 Mtce/Percentage point
Population GDP per capita Proportion
Tertiary industries
S3
0.410
0.438
0.027
Primary industries Secondary industries Tertiary industries Primary industries
I1 I2 I3 M1,coke
0.1794 0.8065 0.2199 0.047
0.1029 0.8170 0.1886 0.014
0.0764 0.0105 0.0314 0.033
Secondary industries
M2,coke
0.231
0.252
0.021
Tertiary industries
M3,coke
0.006
0.000
0.006
Primary industries
M1,heat
0.001
0.001
Secondary industries
M2,heat
0.107
0.054
Tertiary industries
M3,heat
0.016
0.016
Proportion of electricity in end use consumption
Primary industries
M1,electricity
0.356
0.334
0.022
Secondary industries
M2,electricity
0.316
0.291
0.025
Tertiary industries
M3,electricity
0.195
0.213
0.017
Primary energy quantity converted factor
Coke
KPEQ,coke
1.073
1.102
0.029
Heat
KPEQ,heat
1.192
1.316
0.124
Electricity
KPEQ,electricity
3.040
2.563
0.477
Coke
Kcoal,coke
1.000
1.000
0.0002
Heat
Kcoal,heat
0.843
0.906
0.063
Electricity
Kcoal,electricity
0.952
0.967
0.014
Heat
Kfossil,heat
1.000
0.886
0.114
Kfossil,electricity
0.798
0.813
0.015
Kcombustion
1.710
1.460
0.250
Energy intensity
Proportion of coke in end use consumption
Proportion of heat in end use consumption
Coal component factor
Fossil fuel component factor
lk
F0k
kth Influencing factors
Electricity
End-use combustion factor
0.0001 0.053 0.0001
e e e tce/10,000 RMB tce/10,000 RMB tce/10,000 RMB e e e e e e e e e e e e e e e e e e
0.069
2.52
0.197 0.144 0.164 0.012
2.58 13.63 5.23 0.35
Unit
100 tce/tce per 10,000 RMB 100 tce/tce per 10,000 RMB 100 tce/tce per 10,000 RMB 1 Mtce/Percentage point
0.192
9.00
1 Mtce/Percentage point
0.009
1.55
1 Mtce/Percentage point
0.000
0.36
1 Mtce/Percentage point
0.412
7.86
1 Mtce/Percentage point
0.000
1.92
1 Mtce/Percentage point
0.016
0.72
1 Mtce/Percentage point
0.434
17.26
1 Mtce/Percentage point
0.068
3.97
1 Mtce/Percentage point
0.059
2.01
1 Mtce/0.01 factor change
0.063
0.51
1 Mtce/0.01 factor change
1.075
2.25
1 Mtce/0.01 factor change
0.000
2.18
1 Mtce/0.01 factor change
0.046
0.73
1 Mtce/0.01 factor change
0.092
6.57
1 Mtce/0.01 factor change
0.077
0.68
1 Mtce/0.01 factor change
0.113
7.82
1 Mtce/0.01 factor change
0.532
2.13
1 Mtce/0.01 factor change
C. Chong et al. / Energy 85 (2015) 366e378
structures for electricity production compared to those of coke and heat. 3.4. Uncertainties Although the methods and data used in this work represent the best attempts of the authors, uncertainties do exist. For example, we assumed that all of the end-use fuels except coke are combusted at an averaged combustion efficiency estimated from several studies, which may introduce some uncertainties into the results. However, this combustion efficiency considers the importance of using end-use efficiency for total energy consumption [43]. If more indicators of end-use energy efficiency are introduced in the future, such as the efficiencies of electric motors and other devices, the LMDI method could be further expanded to incorporate the impact of end-use efficiencies of primary energy consumption. Another uncertainty is that the coal preparation rate provided in the statistical data is lower than the actual rate. We adopted this adjusted data for mapping the Sankey diagram, but we did not considered it in the LMDI decomposition. Furthermore, the LMDI decomposition approach adopted in this study possesses a limitation in that it can only analyze the influencing factors of coal consumption in the entire industry, which does not included residential consumption. In the future, if residential consumption is to be considered, the LMDI decomposition approach should be applied to residential coal consumption independently. 4. Conclusions and suggestions In this work, we mapped an energy allocation diagram of the coal use in China to present the energy balance of coal flows from raw coal supply to end-use sectors in the years 2001, 2006 and 2011. We conclude that the main features of coal use systems in China from 2001 to 2011 are: 1) China is a growing coal importer, but coal supply still relies on domestic production; 2) China's coal transformation is quite diversified in electricity and heat generation, coking and end-use combustion; and 3) the industrial sector (manufacturing), especially heavy industries, such as iron and steel and cement industries, are responsible for the majority of coal consumption and its growth in China. Based on the technical influencing factors derived from the mapping process, we developed an LMDI decomposition approach of coal consumption growth and applied it to analyze the influencing factors of coal consumption growth in China's entire industry. The results indicated that: 1) the growth of GDP per capita is the major factor driving coal consumption growth in China from 2001 to 2011, which is mainly supported by the expansion of coal intensive industries; 2) the changes in economic structure and energy intensity first increase coal consumption and then reduce coal consumption, which was facilitated by policy adjustments aimed at enhancing structural changes and energy savings after 2006; 3) the improvement of the energy transformation efficiency is the primarily factor responsible for reducing coal consumption from 2001 to 2011, which was primarily realized by the improvement in the electricity supply efficiency, and the improvement in the end-use combustion efficiency also helped reduce coal consumption; and 4) the improvement in the primary energy structure, especially the proportion of non-fossil energy, began to reduce coal consumption only from 2006 due to policy adjustments. The elasticity of these factors is further analyzed to observe their sensitivity, which highlights the importance of the proportion, intensity and energy structures of secondary industries and also the importance of the energy source structures for electricity generation.
377
The policy implications of this study are summarized as follows for controlling total coal consumption in China. First, the prioritized solution is to change the mode of economic growth, as this study shows that a rapid GDP growth based on energy intensive production and consumption will bring about a rapid growth of coal consumption. The emphasis is to reduce the proportion and energy intensity of secondary industries throughout the entire economy, especial heavy industries, and to improve the economy's energy structure by promoting energy that is derived from sources other than coal. Second, the energy transformation efficiency and the end-use combustion efficiency of coal must be continually improved to ensure that coal is used at the best available efficiency, which has already played an important role in reducing coal consumption from 2001 to 2011. Third, the utilization scale of renewable energy, nuclear and natural gas must be rapidly increased to substitute coal, especially for electricity generation. While the total amount of coal needs to be controlled, the majority of new electricity demands should be satisfied by new non-coal electricity generation. Acknowledgements The authors gratefully acknowledge the financial support from the BP Company regarding the Phase II Collaboration between BP and Tsinghua University, and the support from Tsinghua-Rio Tinto Joint Research Centre for Resources, Energy and Sustainable Development for the work on the energy allocation diagram of coal flows. References [1] BP. BP statistical review of world energy June 2013. BP; 2013. [2] China State Council. Energy conservation and emission reduction regulation and plan in 12th five-year plan. The Central People's Government of the People's Republic of China; 2011. [3] China State Council. Energy development regulation and plan in 12th fiveyear plan. The Central People's Government of the People's Republic of China; 2013. [4] Chinese State Council. Chinese State Council plan to restrict China energy consumption. The Central People's Government of the People's Republic of China; 2013. [5] International Energy Agency (IEA). World energy outlook 2013. International Energy Agency; 2013. [6] BP. BP energy outlook 2035. BP; 2014. [7] Fang WZ. The government should restrict China energy consumption. China Econ Trade Her 2008;2:14e6. [8] Ni WD, Chen Z, Li Z. China current energy utilization and key energy strategy. Energy China 2008;12:6e9. [9] Ni WD. Controlling total energy production and consumption in imperative. Sci Technol Rev 2008:24. [10] Ang BW, Zhang FQ. A survey of index decomposition analysis in energy and environmental studies. Energy 2000;25(12):1149e76. [11] Ang BW. Decomposition analysis for policymaking in energy: which is the preferred method? Energy Policy 2004;32(9):1131e9. [12] Ang BW, Liu FL. A new energy decomposition method: perfect in decomposition and consistent in aggregation. Energy 2001;26(6):537e48. [13] Ang BW. The LMDI approach to decomposition analysis: a practical guide. Energy Policy 2005;33(7):867e71. [14] Wang W, Liu X, Zhang M, Song X. Using a new generalized LMDI (logarithmic mean Divisia index) method to analyze China's energy consumption. Energy 2014;67(0):617e22. [15] Fernandez Gonzalez P, Landajo M, Presno MJ. The Divisia real energy intensity indices: evolution and attribution of percent changes in 20 European countries from 1995 to 2010. Energy 2013;58:340e9. [16] Chontanawat J, Wiboonchutikula P, Buddhivanich A. Decomposition analysis of the change of energy intensity of manufacturing industries in Thailand. Energy 2014;77(0):171e82. [17] Cullen JM, Allwood JM. Theoretical efficiency limits for energy conversion devices. Energy 2010;35(5):2059e69. [18] Cullen JM, Allwood JM. The efficient use of energy: tracing the global flow of energy from fuel to service. Energy Policy 2010;38(1):75e81. [19] Ma LW, Allwood JM, Cullen JM, Li Z. The use of energy in China: tracing the flow of energy from primary source to demand drivers. Energy 2012;40(1): 174e88.
378
C. Chong et al. / Energy 85 (2015) 366e378
[20] Yu FF. Coal flow diagram based on coal supply chain: methodology and application. Beijing, China: Tsinghua University; 2013. [21] Yu FF, Ma LW, Jiang LX, Li Z, Ni WD. Mapping the energy flows of coal utilization in China: the methodology and case studies for they year 2005 and 2010. Adv Mater Res 2013;724e725:1234e41. [22] Pu HJ. The sustainable utilization of coal in China. China University of Mining and Technology Press; 2010. [23] Ma J. Present status in China of coal preparation industry & consideration on its development in the twelfth five-year plan. China Coal Process Util 2011;4: 1e5. [24] Chong CH, Ma LW, Li Z, Geng J, Zhang TK. A programmed data-processing method for mapping energy allocation Sankey diagram of China. In: 2014 International conference on power and energy. Shanghai, China: Taylor & Francis Group; 2014. [25] China's National Bureau of Statistics. China energy statistical yearbook 2002. Beijing: China Statistics Press; 2003. [26] China's National Bureau of Statistics. China energy statistical yearbook 2007. Beijing: China Statistics Press; 2008. [27] China's National Bureau of Statistics. China energy statistical yearbook 2012. Beijing: China Statistics Press; 2013. [28] China's National Bureau of Statistics. China statistical yearbook 2002. Beijing: China Statistics Press; 2003. [29] China's National Bureau of Statistics. China statistical yearbook 2007. Beijing: China Statistics Press; 2008. [30] China's National Bureau of Statistics. China statistical yearbook 2012. Beijing: China Statistics Press; 2013. [31] National Development and Reform Commission of China (NDRC). Middle and long term energy conservation plan. Beijing: The Central People's Government of the People's Republic of China; 2004.
[32] National Development and Reform Commission of China (NDRC). Major energy conservation project in eleventh five year plan in China. Beijing: The Central People's Government of the People's Republic of China; 2006. [33] Ma LW, Liu P, Fu F, Li Z, Ni WD. Integrated energy strategy for the sustainable development of China. Energy 2011;36(2):1143e54. [34] Zhao X, Ma C, Hong D. Why did China's energy intensity increase during 1998e2006: decomposition and policy analysis. Energy Policy 2010;38(3): 1379e88. [35] National Development and Reform Commission of China (NDRC). Energy development plan in 11th five year plan. Beijing: The Central People's Government of the People's Republic of China; 2007. [36] Department of Regulation and Statistic in CEC. Statistical data of electricity in China 2009. China Electricity Council (CEC); 2010. [37] Department of Regulation and Statistic in CEC. Statistical data of electricity in China 2012. China Electricity Council (CEC); 2013. [38] Yang Y, Yang Z, Xu G, Wang N. Situation and prospect of energy consumption for China's thermal power generation. Proc CSEE 2013;33:1e11. [39] Wang Y, Gu A, Zhang A. Recent development of energy supply and demand in China, and energy sector prospects through 2030. Energy Policy 2011;39(11): 6745e59. [40] Yu J. Status and transformation measures of industrial coal-fired boiler in China. Clean Coal Technol 2012;3:89e91. 113. [41] Xie Q. Analysis of coal-fired industrial boiler energy consumption status and energy-saving measures. Value Eng 2011;34:45e6. [42] Li Z, Chang SY, Ma LW, Liu P, Zhao LX, Yao Q. The development of low-carbon towns in China: concepts and practices. Energy 2012;47(1):590e9. [43] Sun JX, Ma LW, Ni WD, Li Z. Reducing China's energy consumption and CO2 emissions by more efficient use of energy. In: International conference on materials for renewable energy and environment 2013; 2013. p. 802e8.