- Email: [email protected]
Comprehensive evaluation index system for wind power utilization levels in wind farms in China
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
Contents lists available at ScienceDirect
Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser
crossmark
Comprehensive evaluation index system for wind power utilization levels in wind farms in China ⁎
Rui-jing Shia,b, Xiao-chao Fanb, , Ying Hea a b
Xinjiang Institute of Engineering, No. 1350, Ayding Lake Road, Toutun River District, Urumqi 830000, PR China College of Electrical Engineering, Xinjiang University, No. 1230, Yanan Road, Tianshan District, Urumqi 830000, PR China
A R T I C L E I N F O
A BS T RAC T
Keywords: Wind power farm in China Wind power curtailment Comprehensive evaluation index system Wind power utilization level Improved analytic hierarchy process (IAHP) Fuzzy comprehensive evaluation method (FCEM)
China's wind power installed capacity is the largest in the world, but the utilization of wind power equipment is not very good, far behind USA. In this paper, the development of China's wind power is reviewed, and the present wind power curtailment restricts the sound development of China's wind power industry. The characteristics of China's wind farm are summered. With the insatiability and intermittence, wind power is not welcome to China's electric grid, and large-scale wind power construction does not match with the existing power grid, therefore, wind power curtailment is serious and the level of wind power utilization is very low. To solve the wind power curtailment rationing problem, in this paper, combined with the characteristics of China's wind farm operation, the wind power utilization level evaluation index system has been built, reflecting the wind resource characteristics, wind power equipment type, wind power output, wind power curtailment, grid technology, operation management and so on. Taking Hami wind farm in Xinjiang province as an example, wind power utilization level is evaluated comprehensively, combined the improved analytic hierarchy process (IAHP) analysis and fuzzy comprehensive evaluation method (FCEM). The results show that the establishment of wind power utilization level comprehensive evaluation index system is helpful to find the main factors which effect the level of wind power utilization and improve the wind power field operation, which can provide reference for the planning and design of wind farm, and the results have certain value on theoretical significance and engineering application.
1. Introduction The wind power industry in China has been rapidly developing in the past 10 years under the favorable new energy policy support of the country, which offers unique advantages [1,2]. With the considerable development of the population and economy, China now faces a serious energy crisis and severe environmental pollution [3,4]. According to the Global Wind Energy Association statistics, the new wind power installed capacity of China in 2014 was 23,196 MW, which represents a 44.2% increase relative to the recorded value in 2013 [5]. Despite this large installed capacity, the country's wind power utilization rate remains low. The year 2010 was an important turning point in China's wind power industry. In 2010, China's total installed capacity of wind power surpassed that of the USA, thereby ranking China first in the world. However, the phenomenon of wind power curtailment appeared. The accumulative power of wind turbines in the country reached 44.7 GW, but 31% of these wind turbines could not be interconnected. The total amount of wind power curtailment reached 3.94×109 kW h at the ⁎
curtailment rate of 10%. Since then, China's wind power industry has been struggling between distress and glory [2,4,6]. The wind power curtailment of China from 2010 to 2014 is shown in Fig. 1. The most serious case of wind power curtailment occurred in 2012. From 2010 to 2014, the total wind power curtailment reached 20.82×109 kW h and the curtailment rate was 17.12%. Consequently, the direct economic losses reached 11.4 billion yuan [2,5,6]. China's wind resources are concentrated in large scales and are far away from the load center. Wind power curtailment is mainly concentrated in North, Northeast, and Northwest China, the southeast coast, and other remote areas. Most of these areas are located at the end of the power grid [2,7,8]. Thus, the power grid is relatively weak there. The wind power construction is faster than the growth of the local power accommodation capacity, and the wind power integration scale exceeds the power export capacity of the power grid [7–9]. The actual output of wind farms in China has yet to match expectations because of certain factors, such as wind resources, performance of wind turbine equipment, management of wind farm operations, system disposal capacity, one-sided pursuit of installed
Corresponding author. E-mail addresses: [email protected], [email protected] (X.-c. Fan).
http://dx.doi.org/10.1016/j.rser.2016.11.168 Received 10 September 2015; Received in revised form 11 May 2016; Accepted 12 November 2016 1364-0321/ © 2016 Elsevier Ltd. All rights reserved.
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
power industry. 2. Wind power in China China is a large country with excellent wind energy production that reaches approximately 3.226 billion kW. With technological advancement, the wind energy reserves of the country amount to 1 billion kW, which is close to that of the USA. Hence, China is regarded as one of the five major wind power producers in the world [14–16]. The distribution of the annual duration of wind speed above 3 m/s in China is shown in Fig. 2 [1,2]. Wind energy resources depend on wind energy density and the annual cumulative hours of wind energy. The distribution of the average wind power density in areas with a height of 70 m is shown in Fig. 3 [17]. Wind energy resources are greatly influenced by the terrain. In China, wind energy is mainly distributed in the following areas [18– 20]:
Fig. 1. The wind power curtailment of China from 2010 to 2014 [2,4,5]. (Source: CWEA, GWEC. CWP: Curtailed Wind Power, PCWP:Proportion of Curtailed Wind Power, Economic loss: ￥109 yuan).
capacity and the efficiency ignored [10–12]. The method of multiple indexes is used to analyze the utilization level of wind power, because it is closely related with many factors. In the literature [2,13], three levels of wind power utilization index system is constructed, and the system includes basic indicators, the development of scale indicators and the use of efficiency indicators. Among them, the annual wind power generation accounted for the proportion of power consumption and wind power installed capacity accounted for the proportion of total net installed capacity highlight the scale of the development of wind power in various regions of our country, and the annual utilization hours index is a comprehensive embodiment of accommodating wind power in the power system. For the evaluation methods, domestic and foreign scholars have proposed dozens, and several methods have been widely used, such as principal component analysis method (PCAM), Delphi method, the grey comprehensive evaluation method (GCEM), artificial neural network method (ANNM), analytic hierarchy process (AHP) and fuzzy comprehensive evaluation method (FCEM), and so on, the advantages and disadvantages of these methods as shown in Table 1. Therefore, in this paper, the wind farm operation comprehensive evaluation model is established, combined the improved analytic hierarchy process (IAHP) and fuzzy comprehensive evaluation method (FCEM), the status of the wind energy farm operation is comprehensively evaluated. The results can facilitate the operation and planning of wind farms and promote the scientific development of the wind
(1) Southeast coast and its islands, which serve as the largest wind energy resources; (2) In ner Mongolia and Gansu in the north, which are major wind energy resources; (3) Heilongjiang and Eastern Jilin and the Liaodong Peninsula coast, which also provide a considerable amount of wind energy; (4) The Qinghai Tibet Plateau and the three northern regions of the northern coastal area, which serve as a large wind energy sources; (5) Yunnan, Guizhou, Sichuan, Gansu, Southern Shaanxi, Henan, Western Hunan, the mountainous areas of Fujian, Guangdong, Guangxi, and the Tarim Basin, which feature the smallest wind area. With China's recent focus on the development of nine 10 million kW class wind power bases, nine wind power bases have been built in Hami in Xinjiang, Jiuquan in Gansu, and in other areas, as shown in Fig. 4. The development of these wind power bases follows the wind resource distribution in China and the layout involving the “building of a large base for a large power grid.” The nine large wind farms are expected to reach an installed capacity of more than 79 GW in late 2015; this value should account for more than 75% of the total wind power of the country [21,22]. At present, China's wind power industry is a global leader in the field of wind power development. Since 2010, the total wind power installed capacity of the country has been ranked first [2]. By the end of 2012, the wind power of China had reached 13.5 TWh, thereby making wind power the third largest type of power supplied by the country, with thermal power and hydropower topping the list [3,4]. By the end of 2014, the cumulative wind power installed capacity had reached nearly 114.6 million kW, which accounted for 7% of the total power installed capacity during that period. Moreover, the grid connected capacity reached nearly 1 million kW, which corresponded to the operation of nearly 7 million wind power units, i.e., more than 1500 wind farms. Fig. 5 shows the wind power development in China from 2001 to 2014 [2–4]. According to the China Wind Energy Association statistics [23], the most pressing problem of the Chinese wind power industry is wind power curtailment. The details of the wind power curtailment of China in 2014 are illustrated in Fig. 1. During the said period, the total wind power curtailment reached 70 billion kWh, which corresponded to a direct economic loss of nearly 40 billion yuan over the past five years. As a result of the rejection of wind power turbines for use in the power grid, the poor power supply structure, and the limited regulation capacity of the power system, the output of wind power turbines is restricted, and wind power equipment is even shut down. These conditions increase the severity of the rationing of disposable wind power. Table 2 presents the wind power curtailment in major Chinese provinces in 2014 [24]. The rate of wind power curtailment reached
Table 1 Comparison of several common comprehensive evaluation methods. Evaluation methods
Advantages
Disadvantages
PCAM
Effective reduction of the number of original variables by using dimension reduction ideas, to achieve the effect of fast convergence speed. Give full play to the role of experts, benefit by mutual discussion，high accuracy This method is applicable to the problem of accurate and objective index. This method can deal with nonlinear and non local large scale complex systems, and has strong adaptability and fault tolerance. Level of clarity and ease of analysis
When the principal component factor is positive and negative, the comprehensive evaluation is not clear. The process is complex, and the time is long.
Delphi
GCEM
ANNM
AHP
FCEM
This method can solve the problem of fuzziness and uncertainty.
Only judge the pros and cons, do not reflect the absolute level. Requires a large number of samples and marginal conditions
Evaluation of the object of the factors can not be too much, generally not more than 9 If the indicators do not have mutual independence, it is difficult to solve the information related issues
462
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Fig. 2. Distribution of annual duration of wind speed above 3 m/s in China [1,2]. (Source: CWEA, China Meteorological Station).
the least fluctuation. (2) The construction of large-scale wind farms does not match existing power systems, and such condition leads to serious wind power curtailment. Wind farm construction is faster than grid planning for thermal power and hydropower. In some instances, wind power and other power sources are inadequate, and peak shaving is weak. When the acceptance ability of a power grid is limited, the dispatching department remotely controls the wind turbine output. In such a case, a large amount of wind power is curtailed. Examples of wind power turbine output in a full load condition and in a curtailment case are shown in Figs. 9 and 10, respectively. A high wind power curtailment clearly equates to low wind power generating capacity and significantly low wind power utilization rate.
more than 20% in Jilin, Heilongjiang, and Inner Mongolia (IM). Wind power curtailment was also serious in Xinjiang, Liaoning, and Hebei, with the rate exceeding 16%. 3. Characteristics of an operating wind farm in China According to the statistics of the National Energy Administration and China Wind Energy Association (2010–2014) [24,25], the characteristics of an operating wind farm in China are as follows: (1) The randomness of the wind power output is strong, and intermittence is obvious. The curve of the daily power output of a wind farm in Hami in Xinjiang is shown in Fig. 6. The power output of the wind farm is obviously unstable, and the intermittence and volatility are evident. The power output of a wind farm is influenced by wind speed and the type of wind power unit. Under different wind conditions, the power output of wind turbines of the same type may vary, as shown in Fig. 7. A strong wind speed equates to a large power output. Under the same wind conditions, wind power capacity of different type wind turbine may also vary, as shown in Fig. 8. At 0.5 s, the power output of all wind turbines fluctuates. Specifically, the power output of direct-drive wind turbines shows the most obvious fluctuation, that of doubly fed wind turbines fluctuates frequently, and that of asynchronous wind turbines is relatively stable. After 0.5 s, the power output of direct-drive wind turbines becomes stable and reaches its peak, that of doubly fed wind turbines shows slight fluctuations, and that of asynchronous wind turbines shows
In summary, a large-scale wind power-connected grid influences the secure operation of a power system because of the randomness and volatility of wind energy. Thus, wind farms in China should be beneficial and sustainable to ensure the healthy development of the wind power industry. Along with the establishment of friendly wind farms, provincial power grids must also be operated to improve the level of wind power utilization, reduce the influence of wind farm operations on power systems, and promote a positive interaction between grid networks and wind farms.
4. Index system for wind power utilization level In order to reasonably evaluate the development of wind power in china and fully grasp the level of wind power utilization, a diversified 463
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Fig. 3. Distribution of the average wind power density in China's land 70 m height [16]. (Source: CWEA, China Meteorological Station).
wind power utilization level index comparison system is constructed in the literature [2]. The future wind power utilization level is forecasted and analyzed by using the overall optimization model of power system. The development level of China's wind power should be comprehensive study from multiple aspects. We should focus on the coordinated development of wind power and power system level and improve the overall power system planning and operation level, which is the key to realize wind power scientific and efficient development. In the literature [2], the overall utilization level of wind power industry in China is studied, and this paper focuses on the utilization level of a certain wind farm. In addition to the object of study, the research methods are also different, in literature [2] the future wind power utilization level in China is forecasted and analyzed by using the overall optimization model of power system, however, in this paper the wind farm operation comprehensive evaluation model is established by using the method IAHP and FCEM. Wind power penetration is used in a global scale to indicate the level of wind power farm utilization. The level of wind power farm utilization is related not only to the size of wind resources and the installed capacity but also to the power system. In the present work, we consider the characteristics of wind power farm operations and determine the factors that affect the level of wind power utilization. These factors include wind resource characteristics, type of wind turbine, wind power productivity, equipment operation, wind power curtailment, and friendly wind farm conditions [26].
P=
π ρDW2 v 3CP. 8
(1)
The electric energy is written as
W = PT =
π ρDW2 v 3CP T , 8
(2)
where ρ is the air density, v is the input wind speed, DW is the diameter of the wind wheel, CP is the power factor of the wind turbine, and T denotes the effective wind speed hours. As shown, the factors that affect conversion of energy from wind electricity in wind resources are the wind speed, air density, and effective wind speed hours. 4.2. Type of wind power turbine At present, wind farms in China come in the form of squirrel cage induction generators, doubly fed induction generators, and direct-drive permanent magnet wind generators [9]. Given that wind turbines have several types, their features tend to differ. Specifically, their abilities to capture wind energy, the quality of their power output, and the conditions that stabilize the power grid after the occurrence of a fault are not identical. Their effects also differ even with the use of the same model produced by different manufacturers. The types and characteristics of wind turbines are presented in Table 3.
4.1. Wind resource characteristics
4.3. Operation of wind power equipment
The power output of a wind turbine is written as
With the development of the manufacturing industry dedicated to 464
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Fig. 4. Nine wind power bases in China.
wind power equipment and the improvement of wind power technology, the cost of wind power has been reduced, but the quality of equipment during operations remains problematic. The rate of wind power equipment utilization and the repair duration in cases of equipment failure are used to measure the quality of a wind turbine. The rate of wind turbine utilization can be expressed as
η=
T−A × 100%. T−B
(3)
In Eq. (3), A is the downtime for equipment failure or routine maintenance, B is the non-equipment-related downtime, and T denotes the statistical hours.
Fig. 5. The development of China's wind power from 2001 to 2014. (NIC: New Installed Capacity, CIC: Cumulative Installed Capacity, GR NIC: Growth NIC %, GR CIC: Growth CIC %).
465
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Table 2 Curtailed wind power of China's key provinces in 2014. (CIC: Cumulative Installed Capacity, NGC: New Grid Capacity, CGC: Cumulative Grid Capacity, GE: Grid Electricity, CWP: Curtailed Wind Power, PCWP: Proportion of Curtailed Wind Power, PCWP=CWP/(CWP+GE)). Source: The National Energy Administration (http://www.nea.gov.cn/) [24]. Rank
Province
CIC/GW
NGC/MW
CGC/MW
GE/TWh
CWP/TWh
PCWP/%
1 2 3 4 5 6 7 8 9 10
Jilin Heilongjiang IM Xinjiang Liaoning Hebei Gansu Shandong Shanxi Ningxia
4.65 5.53 22.31 9.67 7.11 9.87 10.73 8.26 5.86 6.14
689.2 1336.3 2309.8 248.9 1091.5 1729.8 2541.3 1282.8 1921.0 2501.3
4256.6 5109.7 20185.6 9012.0 6953.8 9056.6 9768.7 8005.1 5673.6 5893.2
9.87 15.57 83.22 25.32 20.02 39.22 27 26.17 17.15 15.13
5.37 5.14 16.9 5.88 4.24 7.89 1.77 0.7 0.33 0.16
25.24 24.82 23.03 18.85 17.48 16.75 6.15 2.61 1.89 1.06
Fig. 9. Wind power turbine output in a full load condition. Fig. 6. The curve of the daily power output of a wind farm in Hami in Xinjiang.
Fig. 10. Wind power turbine output in a curtailment case. Table 3 Comparison of the types and characteristics of wind turbines.
Fig. 7. The same type of wind turbine output power in different wind conditions.
Characteristic
Asynchronous type
Doubly fed type
Direct-drive type
Generator type
Excitation
Excitation
Fault frequency System reliability Influence on power grid Recovery power generation after fault Reactive absorption Low-voltage ride through
High Low Serious Difficult
Higher Higher Serious Easy
Permanent magnet Low High Minimal Easy
Yes Weak
No Strong
No Stronger
Fig. 8. Under the same wind conditions, output of different type wind turbines.
4.4. Wind power productivity
where Ei is the power output of each wind turbine at the outlet and n is the number of wind turbines in a wind farm.
4.4.1. Wind power generation capacity The wind power generation capacity refers to the sum of the power output per wind power unit during typhoon periods, i.e.,
4.4.2. Wind power installed capacity The wind power installed capacity is written as n
P=
n
E=
∑ Ei, i =1
∑ Pi, i =1
(4)
(5)
where Pi is the capacity of each wind turbine and n is the number of 466
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
power utilization, an evaluation index system for wind power utilization levels is established, as shown in Fig. 11. The proposed system includes comprehensive factors, including the production of wind power and the interaction between wind power and power grids, which affect the level of wind power utilization in the process.
wind turbines in a wind farm. 4.5. Loss of wind power Power plants experience a certain amount of loss in the power generation and transmission process. Wind power loss mainly involves the internal consumption of wind power (electric rate of comprehensive field and loss resulting from equipment failure and maintenance) and the loss resulting from limited power networks. In the case of wind power curtailment, a wind farm is capable of sending electricity but fails to do so because of the limitation in the power grid transmission channel and other factors, including safety issues. These factors do not include loss of electricity as a result of equipment failure. The calculation method for wind power generation in a wind farm is released by an electrical supervisor when the output of the wind farm is limited; the model machine method is used to calculate wind power.
5. Comprehensive evaluation method 5.1. Improved analytic hierarchy process In the improved analytic hierarchy process, a judgment matrix that uses three scales instead of the traditional nine scales is constructed with a high degree of convergence speed and consistency [27]. The steps are as follows: 1. To construct a hierarchy, divide the evaluation object into a target layer, a criterion layer, and an index layer. 2. Calculate the weight of the criterion layer as follows:
4.6. Friendly wind farm conditions (1) Establish the comparison matrix according to the relative importance of two indicators:
To improve the positive interaction between wind farms and power grids and the safe operation level of power networks, provincial power grid operations and wind farm management must meet the following requirements:
⎡0 ⎢ a1 A = ⎢⎢ a2 ⎢⋮ ⎣ an
(1) Basic wind farm management The basic management of a wind farm includes data management, personnel management, and regulatory systems. (2) Grid technical conditions According to the “anti-accident measures for wind power grid operations” and the “technology requirements for wind farm access power systems,” wind farms must be equipped with a dynamic reactive power compensation device and an on-line power quality monitoring device. Moreover, wind farms must be checked for their low-voltage ride through capability, accuracy of the relay protection device, and equipped fault recording and high-precision forecasting devices. (3) Operations management
a1 a11 a21 ⋮ an1
a2 a12 a22 ⋮ a n2
⋯ ⋯ ⋯ ⋯ ⋯
an ⎤ a1n ⎥ a2n ⎥ , ⎥ ⋮ ⎥ ann ⎦
(6)
where a1,…,an are the indexes of the criterion layer; aij can be expressed as follows:
⎧ 2 ai is more important than a j; ⎪ aij = ⎨1 ai and a j are equally important; . ⎪ 0 a is more important than a . j i ⎩
(7)
(2) Establish the structure judgment matrix B, i.e.
⎡0 ⎢ ⎢ a1 B = ⎢ a2 ⎢⋮ ⎢ ⎣ an
Operations management mainly involves thorough examinations and repairs. It is best utilized when dealing with accidents and ensuring safe operations. According to the analysis of the factors affecting the level of wind
a1 b11 b 21 ⋮ bn1
a2 b12 b 22 ⋮ bn2
Fig. 11. Index system for wind power utilization level.
467
⋯ ⋯ ⋯ ⋯ ⋯
an ⎤ ⎥ b1n ⎥ b 2n ⎥ ⋮ ⎥ ⎥ bnn ⎦
(8)
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
⎧ ci − cj + 1 ci ≥ cj ⎪ ⎪ cmin −1 ⎨ bij = ⎡ c − c ⎤ ⎪ ⎢ j i + 1⎥ ci < cj ⎪ ⎣ cmin ⎦ ⎩
Table 4 The definition of the corresponding relation.
n
ci =
∑ aij j =1
(9)
cmin = min{c1, c2, …, cn}.
Grade
Excellent
Good
Moderate
Qualified
Poor
Symbol Quantized value
q1 0.9
q2 0.7
q3 0.5
q4 0.3
q5 0.1
(3) Calculate the weight, and determine the consistency. To calculate the maximum eigenvalue of matrix B, γmax, and the corresponding feature vector C, which is the weight vector, γmax is integrated into the consistency index I=(γmax−n)/(n−1). If I is less than 0.1, then the judgment matrix is in conformity with the requirements. Otherwise, the comparison matrix needs to be recalculated, and the consistency must be checked until the requirements are met. (4) Calculate the weight of each index. The weight vector formed from the judgment matrix is obtained by m experts by repeating processes ①–③, i.e.,
C (k ) = {c1k , c2k , …, cnk }, (k = 1, 2 … m)
E=
1 m
(10)
m
∑ C (k ),
(k = 1, 2 … m). (11)
k =1
In the formula, m is the number of participants, C(k) is the weight vector formed by the k judgment matrix, and E is the weight vector of expectations. Each index weight vector P can be obtained by normalizing weight vector E.
Fig. 12. UHV transmission projects in Hami.
5.2. Comprehensive fuzzy evaluation method
In this way, the level of wind power utilization is converted to the value of W, and wk is the degree in which the wind power utilization level belongs to the k level in the evaluation.
5.2.1. Determining the evaluation level According to the national standard and the feasibility report of the wind farm, the wind power utilization level of a wind farm is divided into five evaluation grades: D={excellent, good, moderate, qualified, poor}, and Dk∈[0,1].
5.2.4. Determining the level of the evaluation object according to the evaluation grade The evaluation level is a non-number; thus, it should be quantified. The definition of the corresponding relation is shown in Table 4. Let Q be the evaluation value of the wind power utilization level. The weighted average method is used to determine the wind power utilization level Q, i.e.,
5.2.2. Building the single factor evaluation matrix According to Fig. 11, the evaluation index system for wind power utilization level is given as A={A1, A2, A3, A4, A5, A6}. Let the i evaluation of a single factor be Ri=(ri1, ri2, ri3, ri4, ri5). In the formula, rik is the degree in which the i factor belongs to the k level in the evaluation.
⎡ R1 ⎤ ⎢ R ⎥ ⎡ r11 ⎢ 2 ⎥ ⎢ r21 ⎢ R ⎥ ⎢r R = ⎢ 3 ⎥ = ⎢ r31 41 R4 ⎢ ⎥ ⎢ r51 R ⎢ 5 ⎢ ⎥ ⎣ r61 ⎣ R6 ⎦
r12 r22 r32 r42 r52 r62
r13 r23 r33 r43 r53 r63
r14 r24 r34 r44 r54 r64
r15 ⎤ r25 ⎥ r35 ⎥ r45 ⎥ r55 ⎥⎥ r65 ⎦
Q = (w10 , w20 , w30 , w40 , w50 )⋅( q1 q2 q3 q4 q5 )T
Thus, the comprehensive evaluation of wind power utilization levels is transformed into a quantitative analysis. 6. Case application The Xinjiang grid connected to the northwest 750 kV channel began its operations on November 3, 2010. The Xinjiang power grid was officially connected to the national grid during this period [28]. Hami, the output port of electric power, is the east gate of Xinjiang. The electric power output under high pressure is shown in Fig. 12. The A line is the AC project of Golmud–Hami at 750 kV. The B line is ± 800 kV between Sourthern Hami and Zhengzhou. The C line is ± 1,100 kV between Northern Hami and Chongqing. Built outside Hami–Zhengzhou, the Hami–Chongqing UHV DC transmission project, and the Hami–Sand–Golmud 750 kV second channel, Hami is the first to house a strong and smart power grid and is thus a fire, wind, and light cluster power energy base. Take for example an operating wind farm in Hami. The main parameters according to the evaluation index system of 2014 obtained from the Hami Statistics Bureau are shown in Table 5.
(12)
5.2.3. Determining the fuzzy evaluation set W, the comprehensive fuzzy evaluation set for wind power utilization levels, represents the product of the normalized weight vector P and the single factor evaluation matrix R, i.e.,
W = ( w1 w2 w3 w4 w5 w6 ) = P⋅R ⎡ r11 r12 r13 ⎢ r21 r22 r23 ⎢r r r p p p p p p = ( 1 2 3 4 5 6 )⋅⎢ r31 r32 r33 ⎢ r41 r42 r43 ⎢ 51 52 53 ⎣ r61 r62 r63
(14)
r14 r24 r34 r44 r54 r64
r15 ⎤ r25 ⎥ r35 ⎥ r45 ⎥ r55 ⎥⎥ r65 ⎦
(13)
468
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
A21 1 A22 3 A23 5
Table 5 Main parameters obtained from Hami wind farm in 2014. Type
Direct-drive type
Unit
Cumulative Year
Wind resource
Average wind speed Effective wind speed hours Average air density
m/s h kg/m3
8.21 8,306.22 1.08
Wind power productivity
Generating capacity Installed capacity
MkWh MW
152.43 50.5
Loss of wind power
Farm consumption Outage loss rate Wind power curtailment
% % %
0.74 11.1 5.1
Equipment operation
Wind turbine utilization Repair time for cases of equipment failure
% h
94.45 3,718.12
A3
A2 A3 A4 A5 A6
5
3
1/ 5 1/ 1 1/ 1/ 5 3 9 1/ 3 1 1/ 3 7 5 9 7 1 1/ 1/ 1/ 1/ 7 3 5 11 3 7 5 1/ 3
7
1/ 6.40 3 3 1/ 7 5 1/ 5 11 3 1 1/ 9 9 1
0.04
2
0.75 0.25
I=0 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A4.
A4
0.14
I=0.02 & $2lt;0.1 Uniform convergence
A31 A32 γmax Weight P I=(γmax−n)/(n−1)
A31 1 3 A32 1/3 1
A41 A42 γmax Weight P I=(γmax−n)/(n−1)
3 A41 1 A42 1/3 1
2
0.75 0.25
I=0 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A5.
A5
A51 A52 A53 γmax Weight P I=(γmax−n)/(n−1)
A51 1 A52 3 A53 5
1/3 1/5 3.04 1 1/3 3 1
0.1 0.26 0.64
I=0.02 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A6.
A6
A61 A62 A63 γmax Weight P I=(γmax−n)/(n−1)
A61 1 A62 5 A63 3
A1 A2 A3 A4 A5 A6 γmax Weight P I=(γmax−n)/ (n−1)
A1 1
0.1 0.26 0.64
The judgment matrix and weight based on A3.
(1) Determining the evaluation factor set As shown in Fig. 11, the evaluation factor set of the first subtarget A is A={A1, A2, A3, A4, A5, A6}. A1={A11, A12, A13}, A2={A21, A22, A23}, A3={A31, A32}, A4={A41, A42}, A5={A51, A52, A53}, and A6={A61, A62, A63}. The evaluation factor sets of the third subtarget A are A61={A611, A612, A613}, A62={A621, A622, A623, A624}, and A63={A631, A632, A633}. (2) Determining the weight set Ten experts were invited to participate in the review of this project. According to the feasibility report and relevant national standards for wind farm operations, the experts provided scores on the basis of the comparison of two indexes that represented the importance of the various factors shown in Fig. 11. The judgment matrix and the weight of each index are obtained using Eqs. (6)– (11). The judgment matrix and weight based on A.
A
1/3 1/5 3.04 1 1/3 3 1
1/5 1/3 3.04 1 3 1/3 1
0.1 0.64 0.26
I=0.02 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A61.
I=0.08 & $2lt;0.1 Uniform convergence
A61
A611 A612 A613 γmax Weight P I=(γmax−n)/(n−1)
A611 1 A612 5 A613 3
0.07 0.47 0.02
1/5 1 1/3
1/3 3 1
3.04
0.1 0.64 0.26
I=0.02 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A62. 0.26 A62
A621 A622 A623 A624 γmax Weight P I=(γmax−n)/ (n−1)
A621 A622 A623 A624
1 3 5 7
The judgment matrix and weight based on A1.
A1
A11 A12 A13 γmax Weight P I=(γmax−n)/(n−1)
3 A11 1 A12 1/3 1 A13 3 5
1/3 3.04 1/5 1
0.26 0.1 0.64
I=0.02 & $2lt;0.1 Uniform convergence
1/5 1/3 1 3
1/7 1/5 1/3 1
4.12
0.06 0.12 0.26 0.56
I=0.04 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A63.
The judgment matrix and weight based on A2.
A2
1/3 1 3 5
A63
A631 A632 A633 γmax Weight P I=(γmax−n)/(n−1)
A631 1 A632 3
A21 A22 A23 γmax Weight P I=(γmax−n)/(n−1) 469
1/3 1
1/5 1/3
3.04
0.1 0.26
I=0.02 & $2lt;0.1 Uniform con-
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
A633 5
3
1
0.64
7. Result analysis
vergence
(3) Determining the evaluation set The evaluation set is D={excellent, good, moderate, qualified, poor}, and Dk∈[0,1]. (4) Single-factor evaluation
The results of the first-stage subtarget evaluation are shown Table 6. Table 6 the results of the first-stage subtarget evaluation.
(a) Third-stage subtarget evaluation The fuzzy evaluation matrix of the third-level subtargets is obtained using the evaluation criteria based on actual wind farm configurations, “anti-accident measures for wind power grid operations,” and “technical requirements for wind farms connected to power systems.”
⎡1 ⎡ 0.7 0.3 0 0 0 ⎤ ⎢ ⎢ ⎥ RA61 = 0.4 0.5 0.1 0 0 RA62 = ⎢1 ⎢⎣ ⎥⎦ ⎢1 0.3 0.6 0.1 0 0 ⎣1 ⎡ 0.2 0.8 0 0 0 ⎤ = ⎢ 0.4 0.5 0.1 0 0 ⎥ ⎢⎣ ⎥ 0.7 0.2 0.1 0 0 ⎦
0 0 0 0
0 0 0 0
0 0 0 0
0⎤ 0⎥ R ⎥ 0 ⎥ A63 0⎦
(15)
(16)
0.7832 0.6536 0.7850 0.6050 0.5720 0.8599
Good Moderate Good Moderate Moderate Good
A6
A11 A12 A13 A41 A42 A61 A62 A63
QA4 = ( 0.1750 0.4250 0.2250 0.1000 0.0750 ) ⋅( 0.9 0.7 0.5 0.3 0.1)T = 0.6050
Weight
Result
Grade
0.26 0.1 0.64 0.75 0.25 0.1 0.64 0.26
0.80 0.76 0.78 0.60 0.62 0.7628 0.90 0.7964
Good Good Good Moderate Moderate Good Excellent Good
On the basis of the evaluation of these indicators, the corresponding conclusions can be drawn: (a) The wind resources are in good condition (0.7832). In 2014, the average wind speed of the evaluated wind farm was 8.21 m/s, which accounted for 86.99% of its annual wind power. The wind farm can thus be considered as a good wind resource. (b) The power production is not high (0.6050). With the selected wind farm being new, the unit operation necessitates a few adjustments, especially in terms of unit fault number. The fan equipment has an average utilization rate of only 94.45%, which is lower than the average utilization rate of a Chinese wind turbine equipment. Moreover, many people have abandoned wind power rationing, which leads to low electrical energy productivity. (c) Wind farms have certain characteristics (0.8599). Wind turbines are large-capacity direct-drive units with full power during A-D-A transformation and grid operation. These devices do not require reactive power compensation, and they feature a low-voltage ride through capability, which greatly improves the stability of wind farm operations under certain friendly conditions. Thus, the evaluation results are in agreement with the actual situation.
The fuzzy comprehensive evaluation matrix of the second-level subtarget RA is obtained with the formula F = P⋅R .
0 0 ⎤ 0.0720 0.0720 ⎥ ⎥ 0.0500 0 ⎥ 0.1000 0.0750 ⎥ 0.2820 0 ⎥ 0 0 ⎥⎦
0.14 0.04 0.07 0.47 0.02 0.26
A4
⎡ 0.2 0.2 0.2 0.2 0.2 ⎤ RA2 = ⎢ 0.2 0.2 0.2 0.2 0.2 ⎥ ⎢⎣ ⎥ 0.3 0.6 0.1 0 0 ⎦ 0.2 0.1 0 0 ⎤ R = ⎡ 0.2 0.4 0.2 0.1 0.1⎤ ⎥ ⎢ ⎥ A4 0.3 0.1 0.2 0 ⎦ ⎣ 0.1 0.5 0.3 0.1 0 ⎦ ⎡ 0.4040 0.5060 0.0900 0 0 ⎤ 0.3 0.1 0 0 ⎤ 0 0 0 0⎥ 0.2 0.1 0.1 0 ⎥ RA6 = ⎢1 ⎥ ⎢⎣ ⎥ 0.2 0.4 0.4 0 ⎦ 0.5720 0.3380 0.0900 0 0 ⎦
0 0.1360 0.1000 0.2250 0.2920 0.0324
Wind resource characteristics A1 Type of wind turbine A2 Equipment operation A3 Wind power productivity A4 Wind power loss A5 Friendly wind farm conditions A6
A1
0.5 0 0 0 ⎤ 0.7 0 0 0 ⎥ ⎥ 0.6 0 0 0 ⎦
0.5840 0.4560 0.2250 0.4250 0.2100 0.1385
Grade
Index
(17)
⎡ 0.4160 ⎢ 0.2640 ⎢ RA = ⎢ 0.6250 ⎢ 0.1750 ⎢ 0.2160 ⎢⎣ 0.8297
Result
Q A1 = ( 0.4160 0.5840 0 0 0 ) ⋅( 0.9 0.7 0.5 0.3 0.1)T = 0.7832 QA6 = ( 0.8297 0.1385 0.0324 0 0 ) ⋅ ( 0.9 0.7 0.5 0.3 0.1)T = 0.8599
(b) Second-stage subtarget evaluation Combined with the given comment set for evaluate indexes A1, A2, A3, A4, A5, and A6, the fuzzy statistical method is used to obtain the following fuzzy evaluation matrices:
⎡ 0.5 RA1 = ⎢ 0.3 ⎢⎣ 0.4 ⎡ 0.7 RA3 = ⎢ ⎣ 0.4 ⎡ 0.6 RA5 = ⎢ 0.6 ⎢⎣ 0
Weight
In the first level, the main factors influencing the wind power utilization level are indexes A6, A4, and A1, which are evaluated. The comprehensive evaluation of each index can be calculated with a fuzzy evaluation vector:
The fuzzy evaluation matrix of the third-level subtargets is obtained with the formula F = P⋅R .
⎡ 0.404 0.506 0.09 0 0 ⎤ RA6 = ⎢ 1 0 0 0 0⎥ ⎢⎣ ⎥ 0.572 0.338 0.09 0 0 ⎦
Index
(18)
(c) First-stage subtarget evaluation The fuzzy evaluation vector of A is obtained with the formula FA = WA⋅RA , i.e.,
FA = ( 0.4167 0.3557 0.1325 0.0590 0.0381).
In summary, the evaluation index system and method proposed in this study are scientific in nature and can thus serve as references in the planning and design of wind farms. The depth of this research is to be improved, in this paper we only select a wind farm to study, which is not able to fully verify the wind power farm operation comprehensive evaluation model rationality. We should select different types of wind farm to verify, in order to improve
Then, the comprehensive evaluation of A is performed as follows:
QA = [ 0.4167 0.3557 0.1325 0.0590 0.0381]⋅( 0.9 0.7 0.5 0.3 0.1)T = 0.7118 The result shows the good grade of the wind power utilization level of the wind farm. 470
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
of wind power in China. Renew Sustain Energy Rev 2015;41:338–46. [9] Li X, Huang Y, Ma S, Li C, Ma X. Analysis on the Influence Factors of Wind Power Accommodation. J Power Energy Eng 2015;3(04):162. [10] Pei W, Chen Y, Sheng K, Deng W, Du Y, Qi Z. Temporal-spatial analysis and improvement measures of Chinese power system for wind power curtailment problem. Renew Sustain Energy Rev 2015;49:148–68. [11] Kane L, Ault GW. Evaluation of wind power curtailment in active network management schemes. IEEE Trans Power Syst 2015;30(2):672–9. [12] He Y, Pang Y, Zhang J, Xia T, Zhang T. Feed-in tariff mechanisms for large-scale wind power in China. Renew Sustain Energy Rev 2015;51:9–17. [13] Bai JH, Zhang Q, Xin SX, Xie GH. Research on Key indexes Comparison of wind. Power Util Lev Electr Power 2012;45(11):7–12. [14] World Wind Energy Association (WWEA) . Small wind world report 2014. Germany: WWEA; 2014. [15] Sun S, Liu F, Xue S, Zeng M, Zeng F. Review on wind power development in China: current situation and improvement strategies to realize future development. Renew Sustain Energy Rev 2015;45:589–99. [16] Guo Y, Ru P, Su J, Anadon LD. Not in my backyard, but not far away from me: local acceptance of wind power in China. Energy 2015;82:722–33. [17] CMA, wind and solar energy resources center. China wind and solar energy resources bulletin 2014. 〈http://cwera.cma.gov.cn〉. [18] Twidell J, Weir T. Renewable energy resources. Routledge; 2015. [19] Xiao L, Wang J, Dong Y, Wu J. Combined forecasting models for wind energy forecasting: a case study in China. Renew Sustain Energy Rev 2015;44:271–88. [20] Zeng M, Duan J, Wang L, Zhang Y, Xue S. Orderly grid connection of renewable energy generation in China: management mode, existing problems and solutions. Renew Sustain Energy Rev 2015;41:14–28. [21] He Y, Pang Y, Zhang J, Xia T, Zhang T. Feed-in tariff mechanisms for large-scale wind power in China. Renew Sustain Energy Rev 2015;51:9–17. [22] Yuan J, Na C, Xu Y, Zhao C. Wind turbine manufacturing in China: a review. Renew Sustain Energy Rev 2015;51:1235–44. [23] CWEA . Wind power brownouts preliminary statistics. Beijing: China Environmental Science Press; 2014. p. 2015. [24] The National Energy Administration. 〈http://www.nea.gov.cn/〉. [25] The China Wind Energy Association. 〈http://www.cwea.org.cn/〉. [26] Johnston L, Díaz-González F, Gomis-Bellmunt O, et al. Methodology for the economic optimisation of energy storage systems for frequency support in wind power plants. Appl Energy 2015;137:660–9. [27] Sun S, Liu F, Xue S, et al. Review on wind power development in China: current situation and improvement strategies to realize future development. Renew Sustain Energy Rev 2015;45:589–99. [28] China’s Energy development strategic action plan (2014–2020). The State Council, Beijing, 20140.11.19.
the evaluation model. In addition, we will consider the impact of social and environmental factors on the operation of the wind farm. In future research, we will extend the operation of wind farm to a wider field, in order to more comprehensive and effective evaluation of the operation of wind farm. Acknowledgements This study forms part of a research project supported by the Ministry of Education Innovation Team Project of China (No. IRT1285), the National Natural Science Foundation of China (Nos. 51666017, 51606163), the major research projects of Xinjiang Uygur Autonomous Region (No. 201230115-3), the Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2016D01C062), and the Xinjiang University Doctor Innovation Project (No. XJUBSCX-201223), the Scientific Research Fund Project of Xinjiang Institute of Engineering (No. 2015xgy291712). The authors would like to express their gratitude for the support of these funding authorities. References [1] Fan J, Wang Q, Sun W. The failure of China's energy development strategy 2050 and its impact on carbon emissions. Renew Sustain Energy Rev 2015;49:1160–70. [2] Fan X, Wang W, Shi R, Li F. Analysis and countermeasures of wind power curtailment in China. Renew Sustain Energy Rev 2015;52:1429–36. [3] Petroleum B. BP Statistical Review of World Energy 2015. London; 2015. [4] Fan X, Wang W, Shi R, Li F. Review of developments and insights into an index system of wind power utilization level. Renew Sustain Energy Rev 2015;48:463–71. [5] GWEC. Global Wind Statistics 2014, 2015.2.10, Brussels, Belgium. [6] Li Junfeng, Cai Fengbo, Qiao Liming, Gao Hu, Wang Qixue, Tan Wenqian, et al. China wind power review and outlook. Beijing: China Environmental Science Press; 2014. [7] Zhao X, Ren L. Focus on the development of offshore wind power in China: has the golden period come?. Renew Energy 2015;81:644–57. [8] Xue B, Ma Z, Geng Y, Heck P, Ren W, Tobias M. A life cycle co-benefits assessment
471
Contents lists available at ScienceDirect
Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser
crossmark
Comprehensive evaluation index system for wind power utilization levels in wind farms in China ⁎
Rui-jing Shia,b, Xiao-chao Fanb, , Ying Hea a b
Xinjiang Institute of Engineering, No. 1350, Ayding Lake Road, Toutun River District, Urumqi 830000, PR China College of Electrical Engineering, Xinjiang University, No. 1230, Yanan Road, Tianshan District, Urumqi 830000, PR China
A R T I C L E I N F O
A BS T RAC T
Keywords: Wind power farm in China Wind power curtailment Comprehensive evaluation index system Wind power utilization level Improved analytic hierarchy process (IAHP) Fuzzy comprehensive evaluation method (FCEM)
China's wind power installed capacity is the largest in the world, but the utilization of wind power equipment is not very good, far behind USA. In this paper, the development of China's wind power is reviewed, and the present wind power curtailment restricts the sound development of China's wind power industry. The characteristics of China's wind farm are summered. With the insatiability and intermittence, wind power is not welcome to China's electric grid, and large-scale wind power construction does not match with the existing power grid, therefore, wind power curtailment is serious and the level of wind power utilization is very low. To solve the wind power curtailment rationing problem, in this paper, combined with the characteristics of China's wind farm operation, the wind power utilization level evaluation index system has been built, reflecting the wind resource characteristics, wind power equipment type, wind power output, wind power curtailment, grid technology, operation management and so on. Taking Hami wind farm in Xinjiang province as an example, wind power utilization level is evaluated comprehensively, combined the improved analytic hierarchy process (IAHP) analysis and fuzzy comprehensive evaluation method (FCEM). The results show that the establishment of wind power utilization level comprehensive evaluation index system is helpful to find the main factors which effect the level of wind power utilization and improve the wind power field operation, which can provide reference for the planning and design of wind farm, and the results have certain value on theoretical significance and engineering application.
1. Introduction The wind power industry in China has been rapidly developing in the past 10 years under the favorable new energy policy support of the country, which offers unique advantages [1,2]. With the considerable development of the population and economy, China now faces a serious energy crisis and severe environmental pollution [3,4]. According to the Global Wind Energy Association statistics, the new wind power installed capacity of China in 2014 was 23,196 MW, which represents a 44.2% increase relative to the recorded value in 2013 [5]. Despite this large installed capacity, the country's wind power utilization rate remains low. The year 2010 was an important turning point in China's wind power industry. In 2010, China's total installed capacity of wind power surpassed that of the USA, thereby ranking China first in the world. However, the phenomenon of wind power curtailment appeared. The accumulative power of wind turbines in the country reached 44.7 GW, but 31% of these wind turbines could not be interconnected. The total amount of wind power curtailment reached 3.94×109 kW h at the ⁎
curtailment rate of 10%. Since then, China's wind power industry has been struggling between distress and glory [2,4,6]. The wind power curtailment of China from 2010 to 2014 is shown in Fig. 1. The most serious case of wind power curtailment occurred in 2012. From 2010 to 2014, the total wind power curtailment reached 20.82×109 kW h and the curtailment rate was 17.12%. Consequently, the direct economic losses reached 11.4 billion yuan [2,5,6]. China's wind resources are concentrated in large scales and are far away from the load center. Wind power curtailment is mainly concentrated in North, Northeast, and Northwest China, the southeast coast, and other remote areas. Most of these areas are located at the end of the power grid [2,7,8]. Thus, the power grid is relatively weak there. The wind power construction is faster than the growth of the local power accommodation capacity, and the wind power integration scale exceeds the power export capacity of the power grid [7–9]. The actual output of wind farms in China has yet to match expectations because of certain factors, such as wind resources, performance of wind turbine equipment, management of wind farm operations, system disposal capacity, one-sided pursuit of installed
Corresponding author. E-mail addresses: [email protected], [email protected] (X.-c. Fan).
http://dx.doi.org/10.1016/j.rser.2016.11.168 Received 10 September 2015; Received in revised form 11 May 2016; Accepted 12 November 2016 1364-0321/ © 2016 Elsevier Ltd. All rights reserved.
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
power industry. 2. Wind power in China China is a large country with excellent wind energy production that reaches approximately 3.226 billion kW. With technological advancement, the wind energy reserves of the country amount to 1 billion kW, which is close to that of the USA. Hence, China is regarded as one of the five major wind power producers in the world [14–16]. The distribution of the annual duration of wind speed above 3 m/s in China is shown in Fig. 2 [1,2]. Wind energy resources depend on wind energy density and the annual cumulative hours of wind energy. The distribution of the average wind power density in areas with a height of 70 m is shown in Fig. 3 [17]. Wind energy resources are greatly influenced by the terrain. In China, wind energy is mainly distributed in the following areas [18– 20]:
Fig. 1. The wind power curtailment of China from 2010 to 2014 [2,4,5]. (Source: CWEA, GWEC. CWP: Curtailed Wind Power, PCWP:Proportion of Curtailed Wind Power, Economic loss: ￥109 yuan).
capacity and the efficiency ignored [10–12]. The method of multiple indexes is used to analyze the utilization level of wind power, because it is closely related with many factors. In the literature [2,13], three levels of wind power utilization index system is constructed, and the system includes basic indicators, the development of scale indicators and the use of efficiency indicators. Among them, the annual wind power generation accounted for the proportion of power consumption and wind power installed capacity accounted for the proportion of total net installed capacity highlight the scale of the development of wind power in various regions of our country, and the annual utilization hours index is a comprehensive embodiment of accommodating wind power in the power system. For the evaluation methods, domestic and foreign scholars have proposed dozens, and several methods have been widely used, such as principal component analysis method (PCAM), Delphi method, the grey comprehensive evaluation method (GCEM), artificial neural network method (ANNM), analytic hierarchy process (AHP) and fuzzy comprehensive evaluation method (FCEM), and so on, the advantages and disadvantages of these methods as shown in Table 1. Therefore, in this paper, the wind farm operation comprehensive evaluation model is established, combined the improved analytic hierarchy process (IAHP) and fuzzy comprehensive evaluation method (FCEM), the status of the wind energy farm operation is comprehensively evaluated. The results can facilitate the operation and planning of wind farms and promote the scientific development of the wind
(1) Southeast coast and its islands, which serve as the largest wind energy resources; (2) In ner Mongolia and Gansu in the north, which are major wind energy resources; (3) Heilongjiang and Eastern Jilin and the Liaodong Peninsula coast, which also provide a considerable amount of wind energy; (4) The Qinghai Tibet Plateau and the three northern regions of the northern coastal area, which serve as a large wind energy sources; (5) Yunnan, Guizhou, Sichuan, Gansu, Southern Shaanxi, Henan, Western Hunan, the mountainous areas of Fujian, Guangdong, Guangxi, and the Tarim Basin, which feature the smallest wind area. With China's recent focus on the development of nine 10 million kW class wind power bases, nine wind power bases have been built in Hami in Xinjiang, Jiuquan in Gansu, and in other areas, as shown in Fig. 4. The development of these wind power bases follows the wind resource distribution in China and the layout involving the “building of a large base for a large power grid.” The nine large wind farms are expected to reach an installed capacity of more than 79 GW in late 2015; this value should account for more than 75% of the total wind power of the country [21,22]. At present, China's wind power industry is a global leader in the field of wind power development. Since 2010, the total wind power installed capacity of the country has been ranked first [2]. By the end of 2012, the wind power of China had reached 13.5 TWh, thereby making wind power the third largest type of power supplied by the country, with thermal power and hydropower topping the list [3,4]. By the end of 2014, the cumulative wind power installed capacity had reached nearly 114.6 million kW, which accounted for 7% of the total power installed capacity during that period. Moreover, the grid connected capacity reached nearly 1 million kW, which corresponded to the operation of nearly 7 million wind power units, i.e., more than 1500 wind farms. Fig. 5 shows the wind power development in China from 2001 to 2014 [2–4]. According to the China Wind Energy Association statistics [23], the most pressing problem of the Chinese wind power industry is wind power curtailment. The details of the wind power curtailment of China in 2014 are illustrated in Fig. 1. During the said period, the total wind power curtailment reached 70 billion kWh, which corresponded to a direct economic loss of nearly 40 billion yuan over the past five years. As a result of the rejection of wind power turbines for use in the power grid, the poor power supply structure, and the limited regulation capacity of the power system, the output of wind power turbines is restricted, and wind power equipment is even shut down. These conditions increase the severity of the rationing of disposable wind power. Table 2 presents the wind power curtailment in major Chinese provinces in 2014 [24]. The rate of wind power curtailment reached
Table 1 Comparison of several common comprehensive evaluation methods. Evaluation methods
Advantages
Disadvantages
PCAM
Effective reduction of the number of original variables by using dimension reduction ideas, to achieve the effect of fast convergence speed. Give full play to the role of experts, benefit by mutual discussion，high accuracy This method is applicable to the problem of accurate and objective index. This method can deal with nonlinear and non local large scale complex systems, and has strong adaptability and fault tolerance. Level of clarity and ease of analysis
When the principal component factor is positive and negative, the comprehensive evaluation is not clear. The process is complex, and the time is long.
Delphi
GCEM
ANNM
AHP
FCEM
This method can solve the problem of fuzziness and uncertainty.
Only judge the pros and cons, do not reflect the absolute level. Requires a large number of samples and marginal conditions
Evaluation of the object of the factors can not be too much, generally not more than 9 If the indicators do not have mutual independence, it is difficult to solve the information related issues
462
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Fig. 2. Distribution of annual duration of wind speed above 3 m/s in China [1,2]. (Source: CWEA, China Meteorological Station).
the least fluctuation. (2) The construction of large-scale wind farms does not match existing power systems, and such condition leads to serious wind power curtailment. Wind farm construction is faster than grid planning for thermal power and hydropower. In some instances, wind power and other power sources are inadequate, and peak shaving is weak. When the acceptance ability of a power grid is limited, the dispatching department remotely controls the wind turbine output. In such a case, a large amount of wind power is curtailed. Examples of wind power turbine output in a full load condition and in a curtailment case are shown in Figs. 9 and 10, respectively. A high wind power curtailment clearly equates to low wind power generating capacity and significantly low wind power utilization rate.
more than 20% in Jilin, Heilongjiang, and Inner Mongolia (IM). Wind power curtailment was also serious in Xinjiang, Liaoning, and Hebei, with the rate exceeding 16%. 3. Characteristics of an operating wind farm in China According to the statistics of the National Energy Administration and China Wind Energy Association (2010–2014) [24,25], the characteristics of an operating wind farm in China are as follows: (1) The randomness of the wind power output is strong, and intermittence is obvious. The curve of the daily power output of a wind farm in Hami in Xinjiang is shown in Fig. 6. The power output of the wind farm is obviously unstable, and the intermittence and volatility are evident. The power output of a wind farm is influenced by wind speed and the type of wind power unit. Under different wind conditions, the power output of wind turbines of the same type may vary, as shown in Fig. 7. A strong wind speed equates to a large power output. Under the same wind conditions, wind power capacity of different type wind turbine may also vary, as shown in Fig. 8. At 0.5 s, the power output of all wind turbines fluctuates. Specifically, the power output of direct-drive wind turbines shows the most obvious fluctuation, that of doubly fed wind turbines fluctuates frequently, and that of asynchronous wind turbines is relatively stable. After 0.5 s, the power output of direct-drive wind turbines becomes stable and reaches its peak, that of doubly fed wind turbines shows slight fluctuations, and that of asynchronous wind turbines shows
In summary, a large-scale wind power-connected grid influences the secure operation of a power system because of the randomness and volatility of wind energy. Thus, wind farms in China should be beneficial and sustainable to ensure the healthy development of the wind power industry. Along with the establishment of friendly wind farms, provincial power grids must also be operated to improve the level of wind power utilization, reduce the influence of wind farm operations on power systems, and promote a positive interaction between grid networks and wind farms.
4. Index system for wind power utilization level In order to reasonably evaluate the development of wind power in china and fully grasp the level of wind power utilization, a diversified 463
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Fig. 3. Distribution of the average wind power density in China's land 70 m height [16]. (Source: CWEA, China Meteorological Station).
wind power utilization level index comparison system is constructed in the literature [2]. The future wind power utilization level is forecasted and analyzed by using the overall optimization model of power system. The development level of China's wind power should be comprehensive study from multiple aspects. We should focus on the coordinated development of wind power and power system level and improve the overall power system planning and operation level, which is the key to realize wind power scientific and efficient development. In the literature [2], the overall utilization level of wind power industry in China is studied, and this paper focuses on the utilization level of a certain wind farm. In addition to the object of study, the research methods are also different, in literature [2] the future wind power utilization level in China is forecasted and analyzed by using the overall optimization model of power system, however, in this paper the wind farm operation comprehensive evaluation model is established by using the method IAHP and FCEM. Wind power penetration is used in a global scale to indicate the level of wind power farm utilization. The level of wind power farm utilization is related not only to the size of wind resources and the installed capacity but also to the power system. In the present work, we consider the characteristics of wind power farm operations and determine the factors that affect the level of wind power utilization. These factors include wind resource characteristics, type of wind turbine, wind power productivity, equipment operation, wind power curtailment, and friendly wind farm conditions [26].
P=
π ρDW2 v 3CP. 8
(1)
The electric energy is written as
W = PT =
π ρDW2 v 3CP T , 8
(2)
where ρ is the air density, v is the input wind speed, DW is the diameter of the wind wheel, CP is the power factor of the wind turbine, and T denotes the effective wind speed hours. As shown, the factors that affect conversion of energy from wind electricity in wind resources are the wind speed, air density, and effective wind speed hours. 4.2. Type of wind power turbine At present, wind farms in China come in the form of squirrel cage induction generators, doubly fed induction generators, and direct-drive permanent magnet wind generators [9]. Given that wind turbines have several types, their features tend to differ. Specifically, their abilities to capture wind energy, the quality of their power output, and the conditions that stabilize the power grid after the occurrence of a fault are not identical. Their effects also differ even with the use of the same model produced by different manufacturers. The types and characteristics of wind turbines are presented in Table 3.
4.1. Wind resource characteristics
4.3. Operation of wind power equipment
The power output of a wind turbine is written as
With the development of the manufacturing industry dedicated to 464
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Fig. 4. Nine wind power bases in China.
wind power equipment and the improvement of wind power technology, the cost of wind power has been reduced, but the quality of equipment during operations remains problematic. The rate of wind power equipment utilization and the repair duration in cases of equipment failure are used to measure the quality of a wind turbine. The rate of wind turbine utilization can be expressed as
η=
T−A × 100%. T−B
(3)
In Eq. (3), A is the downtime for equipment failure or routine maintenance, B is the non-equipment-related downtime, and T denotes the statistical hours.
Fig. 5. The development of China's wind power from 2001 to 2014. (NIC: New Installed Capacity, CIC: Cumulative Installed Capacity, GR NIC: Growth NIC %, GR CIC: Growth CIC %).
465
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
Table 2 Curtailed wind power of China's key provinces in 2014. (CIC: Cumulative Installed Capacity, NGC: New Grid Capacity, CGC: Cumulative Grid Capacity, GE: Grid Electricity, CWP: Curtailed Wind Power, PCWP: Proportion of Curtailed Wind Power, PCWP=CWP/(CWP+GE)). Source: The National Energy Administration (http://www.nea.gov.cn/) [24]. Rank
Province
CIC/GW
NGC/MW
CGC/MW
GE/TWh
CWP/TWh
PCWP/%
1 2 3 4 5 6 7 8 9 10
Jilin Heilongjiang IM Xinjiang Liaoning Hebei Gansu Shandong Shanxi Ningxia
4.65 5.53 22.31 9.67 7.11 9.87 10.73 8.26 5.86 6.14
689.2 1336.3 2309.8 248.9 1091.5 1729.8 2541.3 1282.8 1921.0 2501.3
4256.6 5109.7 20185.6 9012.0 6953.8 9056.6 9768.7 8005.1 5673.6 5893.2
9.87 15.57 83.22 25.32 20.02 39.22 27 26.17 17.15 15.13
5.37 5.14 16.9 5.88 4.24 7.89 1.77 0.7 0.33 0.16
25.24 24.82 23.03 18.85 17.48 16.75 6.15 2.61 1.89 1.06
Fig. 9. Wind power turbine output in a full load condition. Fig. 6. The curve of the daily power output of a wind farm in Hami in Xinjiang.
Fig. 10. Wind power turbine output in a curtailment case. Table 3 Comparison of the types and characteristics of wind turbines.
Fig. 7. The same type of wind turbine output power in different wind conditions.
Characteristic
Asynchronous type
Doubly fed type
Direct-drive type
Generator type
Excitation
Excitation
Fault frequency System reliability Influence on power grid Recovery power generation after fault Reactive absorption Low-voltage ride through
High Low Serious Difficult
Higher Higher Serious Easy
Permanent magnet Low High Minimal Easy
Yes Weak
No Strong
No Stronger
Fig. 8. Under the same wind conditions, output of different type wind turbines.
4.4. Wind power productivity
where Ei is the power output of each wind turbine at the outlet and n is the number of wind turbines in a wind farm.
4.4.1. Wind power generation capacity The wind power generation capacity refers to the sum of the power output per wind power unit during typhoon periods, i.e.,
4.4.2. Wind power installed capacity The wind power installed capacity is written as n
P=
n
E=
∑ Ei, i =1
∑ Pi, i =1
(4)
(5)
where Pi is the capacity of each wind turbine and n is the number of 466
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
power utilization, an evaluation index system for wind power utilization levels is established, as shown in Fig. 11. The proposed system includes comprehensive factors, including the production of wind power and the interaction between wind power and power grids, which affect the level of wind power utilization in the process.
wind turbines in a wind farm. 4.5. Loss of wind power Power plants experience a certain amount of loss in the power generation and transmission process. Wind power loss mainly involves the internal consumption of wind power (electric rate of comprehensive field and loss resulting from equipment failure and maintenance) and the loss resulting from limited power networks. In the case of wind power curtailment, a wind farm is capable of sending electricity but fails to do so because of the limitation in the power grid transmission channel and other factors, including safety issues. These factors do not include loss of electricity as a result of equipment failure. The calculation method for wind power generation in a wind farm is released by an electrical supervisor when the output of the wind farm is limited; the model machine method is used to calculate wind power.
5. Comprehensive evaluation method 5.1. Improved analytic hierarchy process In the improved analytic hierarchy process, a judgment matrix that uses three scales instead of the traditional nine scales is constructed with a high degree of convergence speed and consistency [27]. The steps are as follows: 1. To construct a hierarchy, divide the evaluation object into a target layer, a criterion layer, and an index layer. 2. Calculate the weight of the criterion layer as follows:
4.6. Friendly wind farm conditions (1) Establish the comparison matrix according to the relative importance of two indicators:
To improve the positive interaction between wind farms and power grids and the safe operation level of power networks, provincial power grid operations and wind farm management must meet the following requirements:
⎡0 ⎢ a1 A = ⎢⎢ a2 ⎢⋮ ⎣ an
(1) Basic wind farm management The basic management of a wind farm includes data management, personnel management, and regulatory systems. (2) Grid technical conditions According to the “anti-accident measures for wind power grid operations” and the “technology requirements for wind farm access power systems,” wind farms must be equipped with a dynamic reactive power compensation device and an on-line power quality monitoring device. Moreover, wind farms must be checked for their low-voltage ride through capability, accuracy of the relay protection device, and equipped fault recording and high-precision forecasting devices. (3) Operations management
a1 a11 a21 ⋮ an1
a2 a12 a22 ⋮ a n2
⋯ ⋯ ⋯ ⋯ ⋯
an ⎤ a1n ⎥ a2n ⎥ , ⎥ ⋮ ⎥ ann ⎦
(6)
where a1,…,an are the indexes of the criterion layer; aij can be expressed as follows:
⎧ 2 ai is more important than a j; ⎪ aij = ⎨1 ai and a j are equally important; . ⎪ 0 a is more important than a . j i ⎩
(7)
(2) Establish the structure judgment matrix B, i.e.
⎡0 ⎢ ⎢ a1 B = ⎢ a2 ⎢⋮ ⎢ ⎣ an
Operations management mainly involves thorough examinations and repairs. It is best utilized when dealing with accidents and ensuring safe operations. According to the analysis of the factors affecting the level of wind
a1 b11 b 21 ⋮ bn1
a2 b12 b 22 ⋮ bn2
Fig. 11. Index system for wind power utilization level.
467
⋯ ⋯ ⋯ ⋯ ⋯
an ⎤ ⎥ b1n ⎥ b 2n ⎥ ⋮ ⎥ ⎥ bnn ⎦
(8)
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
⎧ ci − cj + 1 ci ≥ cj ⎪ ⎪ cmin −1 ⎨ bij = ⎡ c − c ⎤ ⎪ ⎢ j i + 1⎥ ci < cj ⎪ ⎣ cmin ⎦ ⎩
Table 4 The definition of the corresponding relation.
n
ci =
∑ aij j =1
(9)
cmin = min{c1, c2, …, cn}.
Grade
Excellent
Good
Moderate
Qualified
Poor
Symbol Quantized value
q1 0.9
q2 0.7
q3 0.5
q4 0.3
q5 0.1
(3) Calculate the weight, and determine the consistency. To calculate the maximum eigenvalue of matrix B, γmax, and the corresponding feature vector C, which is the weight vector, γmax is integrated into the consistency index I=(γmax−n)/(n−1). If I is less than 0.1, then the judgment matrix is in conformity with the requirements. Otherwise, the comparison matrix needs to be recalculated, and the consistency must be checked until the requirements are met. (4) Calculate the weight of each index. The weight vector formed from the judgment matrix is obtained by m experts by repeating processes ①–③, i.e.,
C (k ) = {c1k , c2k , …, cnk }, (k = 1, 2 … m)
E=
1 m
(10)
m
∑ C (k ),
(k = 1, 2 … m). (11)
k =1
In the formula, m is the number of participants, C(k) is the weight vector formed by the k judgment matrix, and E is the weight vector of expectations. Each index weight vector P can be obtained by normalizing weight vector E.
Fig. 12. UHV transmission projects in Hami.
5.2. Comprehensive fuzzy evaluation method
In this way, the level of wind power utilization is converted to the value of W, and wk is the degree in which the wind power utilization level belongs to the k level in the evaluation.
5.2.1. Determining the evaluation level According to the national standard and the feasibility report of the wind farm, the wind power utilization level of a wind farm is divided into five evaluation grades: D={excellent, good, moderate, qualified, poor}, and Dk∈[0,1].
5.2.4. Determining the level of the evaluation object according to the evaluation grade The evaluation level is a non-number; thus, it should be quantified. The definition of the corresponding relation is shown in Table 4. Let Q be the evaluation value of the wind power utilization level. The weighted average method is used to determine the wind power utilization level Q, i.e.,
5.2.2. Building the single factor evaluation matrix According to Fig. 11, the evaluation index system for wind power utilization level is given as A={A1, A2, A3, A4, A5, A6}. Let the i evaluation of a single factor be Ri=(ri1, ri2, ri3, ri4, ri5). In the formula, rik is the degree in which the i factor belongs to the k level in the evaluation.
⎡ R1 ⎤ ⎢ R ⎥ ⎡ r11 ⎢ 2 ⎥ ⎢ r21 ⎢ R ⎥ ⎢r R = ⎢ 3 ⎥ = ⎢ r31 41 R4 ⎢ ⎥ ⎢ r51 R ⎢ 5 ⎢ ⎥ ⎣ r61 ⎣ R6 ⎦
r12 r22 r32 r42 r52 r62
r13 r23 r33 r43 r53 r63
r14 r24 r34 r44 r54 r64
r15 ⎤ r25 ⎥ r35 ⎥ r45 ⎥ r55 ⎥⎥ r65 ⎦
Q = (w10 , w20 , w30 , w40 , w50 )⋅( q1 q2 q3 q4 q5 )T
Thus, the comprehensive evaluation of wind power utilization levels is transformed into a quantitative analysis. 6. Case application The Xinjiang grid connected to the northwest 750 kV channel began its operations on November 3, 2010. The Xinjiang power grid was officially connected to the national grid during this period [28]. Hami, the output port of electric power, is the east gate of Xinjiang. The electric power output under high pressure is shown in Fig. 12. The A line is the AC project of Golmud–Hami at 750 kV. The B line is ± 800 kV between Sourthern Hami and Zhengzhou. The C line is ± 1,100 kV between Northern Hami and Chongqing. Built outside Hami–Zhengzhou, the Hami–Chongqing UHV DC transmission project, and the Hami–Sand–Golmud 750 kV second channel, Hami is the first to house a strong and smart power grid and is thus a fire, wind, and light cluster power energy base. Take for example an operating wind farm in Hami. The main parameters according to the evaluation index system of 2014 obtained from the Hami Statistics Bureau are shown in Table 5.
(12)
5.2.3. Determining the fuzzy evaluation set W, the comprehensive fuzzy evaluation set for wind power utilization levels, represents the product of the normalized weight vector P and the single factor evaluation matrix R, i.e.,
W = ( w1 w2 w3 w4 w5 w6 ) = P⋅R ⎡ r11 r12 r13 ⎢ r21 r22 r23 ⎢r r r p p p p p p = ( 1 2 3 4 5 6 )⋅⎢ r31 r32 r33 ⎢ r41 r42 r43 ⎢ 51 52 53 ⎣ r61 r62 r63
(14)
r14 r24 r34 r44 r54 r64
r15 ⎤ r25 ⎥ r35 ⎥ r45 ⎥ r55 ⎥⎥ r65 ⎦
(13)
468
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
A21 1 A22 3 A23 5
Table 5 Main parameters obtained from Hami wind farm in 2014. Type
Direct-drive type
Unit
Cumulative Year
Wind resource
Average wind speed Effective wind speed hours Average air density
m/s h kg/m3
8.21 8,306.22 1.08
Wind power productivity
Generating capacity Installed capacity
MkWh MW
152.43 50.5
Loss of wind power
Farm consumption Outage loss rate Wind power curtailment
% % %
0.74 11.1 5.1
Equipment operation
Wind turbine utilization Repair time for cases of equipment failure
% h
94.45 3,718.12
A3
A2 A3 A4 A5 A6
5
3
1/ 5 1/ 1 1/ 1/ 5 3 9 1/ 3 1 1/ 3 7 5 9 7 1 1/ 1/ 1/ 1/ 7 3 5 11 3 7 5 1/ 3
7
1/ 6.40 3 3 1/ 7 5 1/ 5 11 3 1 1/ 9 9 1
0.04
2
0.75 0.25
I=0 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A4.
A4
0.14
I=0.02 & $2lt;0.1 Uniform convergence
A31 A32 γmax Weight P I=(γmax−n)/(n−1)
A31 1 3 A32 1/3 1
A41 A42 γmax Weight P I=(γmax−n)/(n−1)
3 A41 1 A42 1/3 1
2
0.75 0.25
I=0 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A5.
A5
A51 A52 A53 γmax Weight P I=(γmax−n)/(n−1)
A51 1 A52 3 A53 5
1/3 1/5 3.04 1 1/3 3 1
0.1 0.26 0.64
I=0.02 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A6.
A6
A61 A62 A63 γmax Weight P I=(γmax−n)/(n−1)
A61 1 A62 5 A63 3
A1 A2 A3 A4 A5 A6 γmax Weight P I=(γmax−n)/ (n−1)
A1 1
0.1 0.26 0.64
The judgment matrix and weight based on A3.
(1) Determining the evaluation factor set As shown in Fig. 11, the evaluation factor set of the first subtarget A is A={A1, A2, A3, A4, A5, A6}. A1={A11, A12, A13}, A2={A21, A22, A23}, A3={A31, A32}, A4={A41, A42}, A5={A51, A52, A53}, and A6={A61, A62, A63}. The evaluation factor sets of the third subtarget A are A61={A611, A612, A613}, A62={A621, A622, A623, A624}, and A63={A631, A632, A633}. (2) Determining the weight set Ten experts were invited to participate in the review of this project. According to the feasibility report and relevant national standards for wind farm operations, the experts provided scores on the basis of the comparison of two indexes that represented the importance of the various factors shown in Fig. 11. The judgment matrix and the weight of each index are obtained using Eqs. (6)– (11). The judgment matrix and weight based on A.
A
1/3 1/5 3.04 1 1/3 3 1
1/5 1/3 3.04 1 3 1/3 1
0.1 0.64 0.26
I=0.02 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A61.
I=0.08 & $2lt;0.1 Uniform convergence
A61
A611 A612 A613 γmax Weight P I=(γmax−n)/(n−1)
A611 1 A612 5 A613 3
0.07 0.47 0.02
1/5 1 1/3
1/3 3 1
3.04
0.1 0.64 0.26
I=0.02 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A62. 0.26 A62
A621 A622 A623 A624 γmax Weight P I=(γmax−n)/ (n−1)
A621 A622 A623 A624
1 3 5 7
The judgment matrix and weight based on A1.
A1
A11 A12 A13 γmax Weight P I=(γmax−n)/(n−1)
3 A11 1 A12 1/3 1 A13 3 5
1/3 3.04 1/5 1
0.26 0.1 0.64
I=0.02 & $2lt;0.1 Uniform convergence
1/5 1/3 1 3
1/7 1/5 1/3 1
4.12
0.06 0.12 0.26 0.56
I=0.04 & $2lt;0.1 Uniform convergence
The judgment matrix and weight based on A63.
The judgment matrix and weight based on A2.
A2
1/3 1 3 5
A63
A631 A632 A633 γmax Weight P I=(γmax−n)/(n−1)
A631 1 A632 3
A21 A22 A23 γmax Weight P I=(γmax−n)/(n−1) 469
1/3 1
1/5 1/3
3.04
0.1 0.26
I=0.02 & $2lt;0.1 Uniform con-
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
A633 5
3
1
0.64
7. Result analysis
vergence
(3) Determining the evaluation set The evaluation set is D={excellent, good, moderate, qualified, poor}, and Dk∈[0,1]. (4) Single-factor evaluation
The results of the first-stage subtarget evaluation are shown Table 6. Table 6 the results of the first-stage subtarget evaluation.
(a) Third-stage subtarget evaluation The fuzzy evaluation matrix of the third-level subtargets is obtained using the evaluation criteria based on actual wind farm configurations, “anti-accident measures for wind power grid operations,” and “technical requirements for wind farms connected to power systems.”
⎡1 ⎡ 0.7 0.3 0 0 0 ⎤ ⎢ ⎢ ⎥ RA61 = 0.4 0.5 0.1 0 0 RA62 = ⎢1 ⎢⎣ ⎥⎦ ⎢1 0.3 0.6 0.1 0 0 ⎣1 ⎡ 0.2 0.8 0 0 0 ⎤ = ⎢ 0.4 0.5 0.1 0 0 ⎥ ⎢⎣ ⎥ 0.7 0.2 0.1 0 0 ⎦
0 0 0 0
0 0 0 0
0 0 0 0
0⎤ 0⎥ R ⎥ 0 ⎥ A63 0⎦
(15)
(16)
0.7832 0.6536 0.7850 0.6050 0.5720 0.8599
Good Moderate Good Moderate Moderate Good
A6
A11 A12 A13 A41 A42 A61 A62 A63
QA4 = ( 0.1750 0.4250 0.2250 0.1000 0.0750 ) ⋅( 0.9 0.7 0.5 0.3 0.1)T = 0.6050
Weight
Result
Grade
0.26 0.1 0.64 0.75 0.25 0.1 0.64 0.26
0.80 0.76 0.78 0.60 0.62 0.7628 0.90 0.7964
Good Good Good Moderate Moderate Good Excellent Good
On the basis of the evaluation of these indicators, the corresponding conclusions can be drawn: (a) The wind resources are in good condition (0.7832). In 2014, the average wind speed of the evaluated wind farm was 8.21 m/s, which accounted for 86.99% of its annual wind power. The wind farm can thus be considered as a good wind resource. (b) The power production is not high (0.6050). With the selected wind farm being new, the unit operation necessitates a few adjustments, especially in terms of unit fault number. The fan equipment has an average utilization rate of only 94.45%, which is lower than the average utilization rate of a Chinese wind turbine equipment. Moreover, many people have abandoned wind power rationing, which leads to low electrical energy productivity. (c) Wind farms have certain characteristics (0.8599). Wind turbines are large-capacity direct-drive units with full power during A-D-A transformation and grid operation. These devices do not require reactive power compensation, and they feature a low-voltage ride through capability, which greatly improves the stability of wind farm operations under certain friendly conditions. Thus, the evaluation results are in agreement with the actual situation.
The fuzzy comprehensive evaluation matrix of the second-level subtarget RA is obtained with the formula F = P⋅R .
0 0 ⎤ 0.0720 0.0720 ⎥ ⎥ 0.0500 0 ⎥ 0.1000 0.0750 ⎥ 0.2820 0 ⎥ 0 0 ⎥⎦
0.14 0.04 0.07 0.47 0.02 0.26
A4
⎡ 0.2 0.2 0.2 0.2 0.2 ⎤ RA2 = ⎢ 0.2 0.2 0.2 0.2 0.2 ⎥ ⎢⎣ ⎥ 0.3 0.6 0.1 0 0 ⎦ 0.2 0.1 0 0 ⎤ R = ⎡ 0.2 0.4 0.2 0.1 0.1⎤ ⎥ ⎢ ⎥ A4 0.3 0.1 0.2 0 ⎦ ⎣ 0.1 0.5 0.3 0.1 0 ⎦ ⎡ 0.4040 0.5060 0.0900 0 0 ⎤ 0.3 0.1 0 0 ⎤ 0 0 0 0⎥ 0.2 0.1 0.1 0 ⎥ RA6 = ⎢1 ⎥ ⎢⎣ ⎥ 0.2 0.4 0.4 0 ⎦ 0.5720 0.3380 0.0900 0 0 ⎦
0 0.1360 0.1000 0.2250 0.2920 0.0324
Wind resource characteristics A1 Type of wind turbine A2 Equipment operation A3 Wind power productivity A4 Wind power loss A5 Friendly wind farm conditions A6
A1
0.5 0 0 0 ⎤ 0.7 0 0 0 ⎥ ⎥ 0.6 0 0 0 ⎦
0.5840 0.4560 0.2250 0.4250 0.2100 0.1385
Grade
Index
(17)
⎡ 0.4160 ⎢ 0.2640 ⎢ RA = ⎢ 0.6250 ⎢ 0.1750 ⎢ 0.2160 ⎢⎣ 0.8297
Result
Q A1 = ( 0.4160 0.5840 0 0 0 ) ⋅( 0.9 0.7 0.5 0.3 0.1)T = 0.7832 QA6 = ( 0.8297 0.1385 0.0324 0 0 ) ⋅ ( 0.9 0.7 0.5 0.3 0.1)T = 0.8599
(b) Second-stage subtarget evaluation Combined with the given comment set for evaluate indexes A1, A2, A3, A4, A5, and A6, the fuzzy statistical method is used to obtain the following fuzzy evaluation matrices:
⎡ 0.5 RA1 = ⎢ 0.3 ⎢⎣ 0.4 ⎡ 0.7 RA3 = ⎢ ⎣ 0.4 ⎡ 0.6 RA5 = ⎢ 0.6 ⎢⎣ 0
Weight
In the first level, the main factors influencing the wind power utilization level are indexes A6, A4, and A1, which are evaluated. The comprehensive evaluation of each index can be calculated with a fuzzy evaluation vector:
The fuzzy evaluation matrix of the third-level subtargets is obtained with the formula F = P⋅R .
⎡ 0.404 0.506 0.09 0 0 ⎤ RA6 = ⎢ 1 0 0 0 0⎥ ⎢⎣ ⎥ 0.572 0.338 0.09 0 0 ⎦
Index
(18)
(c) First-stage subtarget evaluation The fuzzy evaluation vector of A is obtained with the formula FA = WA⋅RA , i.e.,
FA = ( 0.4167 0.3557 0.1325 0.0590 0.0381).
In summary, the evaluation index system and method proposed in this study are scientific in nature and can thus serve as references in the planning and design of wind farms. The depth of this research is to be improved, in this paper we only select a wind farm to study, which is not able to fully verify the wind power farm operation comprehensive evaluation model rationality. We should select different types of wind farm to verify, in order to improve
Then, the comprehensive evaluation of A is performed as follows:
QA = [ 0.4167 0.3557 0.1325 0.0590 0.0381]⋅( 0.9 0.7 0.5 0.3 0.1)T = 0.7118 The result shows the good grade of the wind power utilization level of the wind farm. 470
Renewable and Sustainable Energy Reviews 69 (2017) 461–471
R.-j. Shi et al.
of wind power in China. Renew Sustain Energy Rev 2015;41:338–46. [9] Li X, Huang Y, Ma S, Li C, Ma X. Analysis on the Influence Factors of Wind Power Accommodation. J Power Energy Eng 2015;3(04):162. [10] Pei W, Chen Y, Sheng K, Deng W, Du Y, Qi Z. Temporal-spatial analysis and improvement measures of Chinese power system for wind power curtailment problem. Renew Sustain Energy Rev 2015;49:148–68. [11] Kane L, Ault GW. Evaluation of wind power curtailment in active network management schemes. IEEE Trans Power Syst 2015;30(2):672–9. [12] He Y, Pang Y, Zhang J, Xia T, Zhang T. Feed-in tariff mechanisms for large-scale wind power in China. Renew Sustain Energy Rev 2015;51:9–17. [13] Bai JH, Zhang Q, Xin SX, Xie GH. Research on Key indexes Comparison of wind. Power Util Lev Electr Power 2012;45(11):7–12. [14] World Wind Energy Association (WWEA) . Small wind world report 2014. Germany: WWEA; 2014. [15] Sun S, Liu F, Xue S, Zeng M, Zeng F. Review on wind power development in China: current situation and improvement strategies to realize future development. Renew Sustain Energy Rev 2015;45:589–99. [16] Guo Y, Ru P, Su J, Anadon LD. Not in my backyard, but not far away from me: local acceptance of wind power in China. Energy 2015;82:722–33. [17] CMA, wind and solar energy resources center. China wind and solar energy resources bulletin 2014. 〈http://cwera.cma.gov.cn〉. [18] Twidell J, Weir T. Renewable energy resources. Routledge; 2015. [19] Xiao L, Wang J, Dong Y, Wu J. Combined forecasting models for wind energy forecasting: a case study in China. Renew Sustain Energy Rev 2015;44:271–88. [20] Zeng M, Duan J, Wang L, Zhang Y, Xue S. Orderly grid connection of renewable energy generation in China: management mode, existing problems and solutions. Renew Sustain Energy Rev 2015;41:14–28. [21] He Y, Pang Y, Zhang J, Xia T, Zhang T. Feed-in tariff mechanisms for large-scale wind power in China. Renew Sustain Energy Rev 2015;51:9–17. [22] Yuan J, Na C, Xu Y, Zhao C. Wind turbine manufacturing in China: a review. Renew Sustain Energy Rev 2015;51:1235–44. [23] CWEA . Wind power brownouts preliminary statistics. Beijing: China Environmental Science Press; 2014. p. 2015. [24] The National Energy Administration. 〈http://www.nea.gov.cn/〉. [25] The China Wind Energy Association. 〈http://www.cwea.org.cn/〉. [26] Johnston L, Díaz-González F, Gomis-Bellmunt O, et al. Methodology for the economic optimisation of energy storage systems for frequency support in wind power plants. Appl Energy 2015;137:660–9. [27] Sun S, Liu F, Xue S, et al. Review on wind power development in China: current situation and improvement strategies to realize future development. Renew Sustain Energy Rev 2015;45:589–99. [28] China’s Energy development strategic action plan (2014–2020). The State Council, Beijing, 20140.11.19.
the evaluation model. In addition, we will consider the impact of social and environmental factors on the operation of the wind farm. In future research, we will extend the operation of wind farm to a wider field, in order to more comprehensive and effective evaluation of the operation of wind farm. Acknowledgements This study forms part of a research project supported by the Ministry of Education Innovation Team Project of China (No. IRT1285), the National Natural Science Foundation of China (Nos. 51666017, 51606163), the major research projects of Xinjiang Uygur Autonomous Region (No. 201230115-3), the Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2016D01C062), and the Xinjiang University Doctor Innovation Project (No. XJUBSCX-201223), the Scientific Research Fund Project of Xinjiang Institute of Engineering (No. 2015xgy291712). The authors would like to express their gratitude for the support of these funding authorities. References [1] Fan J, Wang Q, Sun W. The failure of China's energy development strategy 2050 and its impact on carbon emissions. Renew Sustain Energy Rev 2015;49:1160–70. [2] Fan X, Wang W, Shi R, Li F. Analysis and countermeasures of wind power curtailment in China. Renew Sustain Energy Rev 2015;52:1429–36. [3] Petroleum B. BP Statistical Review of World Energy 2015. London; 2015. [4] Fan X, Wang W, Shi R, Li F. Review of developments and insights into an index system of wind power utilization level. Renew Sustain Energy Rev 2015;48:463–71. [5] GWEC. Global Wind Statistics 2014, 2015.2.10, Brussels, Belgium. [6] Li Junfeng, Cai Fengbo, Qiao Liming, Gao Hu, Wang Qixue, Tan Wenqian, et al. China wind power review and outlook. Beijing: China Environmental Science Press; 2014. [7] Zhao X, Ren L. Focus on the development of offshore wind power in China: has the golden period come?. Renew Energy 2015;81:644–57. [8] Xue B, Ma Z, Geng Y, Heck P, Ren W, Tobias M. A life cycle co-benefits assessment
471