Comprehensive and Integrated Mine Ventilation Consultation Model – CIMVCM

Tunnelling and Underground Space Technology 45 (2015) 166–180

Contents lists available at ScienceDirect

Tunnelling and Underground Space Technology journal homepage: www.elsevier.com/locate/tust

Comprehensive and Integrated Mine Ventilation Consultation Model – CIMVCM Jianwei Cheng a,⇑, Yan Wu b, Haiming Xu b, Jin Liu b, Yekang Yang b, Huangjun Deng b, Yi Wang b a b

Key Laboratory of Gas and Fire Control for Coal Mines College of Safety Engineering, China University of Mining and Technology, Xuzhou, Jiangsu 221116, China College of Computer Science and Engineering, China University of Mining and Technology, Xuzhou, Jiangsu 221116, China

a r t i c l e

i n f o

Article history: Received 22 April 2014 Received in revised form 2 September 2014 Accepted 11 September 2014

Keywords: Underground mine ventilation Optimization Mining environmental control Risk identification and mitigation Computer integrated software

a b s t r a c t Since the beginning of mining, the problems induced by ventilation in underground coal mining practices have prompted great needs of reliable consulting tools to assist mining engineers, government agencies or researchers’ investigating and analyzing the mine ventilation system, or making any ventilation managements. Due to the development of mining method and modern mining (e.g. longwall mining) employed in recent years, the mentioned needs become more and more critical for mine operators. In this paper, a PC based computer mine ventilation consultation software model developed by the authors is introduced. This model is called CIMVCM (Comprehensive and Integrated Mine Ventilation Consultation Model). The computer model is a comprehensive, reliable and user-friendly one. The nature of ‘‘comprehensive’’ means the model can be used to consult the system’s rational reliability allocation, can assist engineers to select the best ventilation plan from multiple candidate ones, can quantitatively rate a system’s potential risk and safety degree and also can check the system’s reliability based on field observations. All of those works are very important to mine operators understanding the system and then taking proper measures to control the potential risk. In addition to that, usages of developed computer program are also very easy. Users can operate the program without possessing an in-depth knowledge on computer and mathematical theory. Until today, CIMVCM has been successfully employed in numerous cases of designing and assessing various mine ventilation systems during the past decades. The reliability of mathematical models used in CIMVCM has also been proven by such applications. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction The importance of mine ventilation has not just newly been recognized. As early as 40,000 B.C. in Palaeolithic times, when miners faced oxygen deficiency, toxic gases, harmful dust, etc., they had to develop some methods to course the air through shafts or tunnels to provide fresh air to the underground (Hartman et al., 1997). Until today, delivering a flow of air to the underground workings of a mine of sufficient volume to dilute and remove noxious gases (typically NOx, SO2, methane, CO2 and CO) (Wikipedia, 2013) is still the first and primary objective for an underground coal mine. However, under the background of modern mining methods extensively applied nowadays, the challenges that a mine ventilation engineer faces are not just limited in delivering fresh air from surface any more. As the powerful and mechanized mining equipment employed today, both the underground mining intensity and the coal production scale are expanded by multiple times, which ⇑ Corresponding author. Tel.: +86 516 8359 0598. E-mail address: [email protected] (J. Cheng). http://dx.doi.org/10.1016/j.tust.2014.09.004 0886-7798/Ó 2014 Elsevier Ltd. All rights reserved.

brings great changes in the mine ventilation network designing, topological characterises, controls, etc. During the time of such changes happened, although the mining system becomes much simpler than old time, the system’s reliability is placed more emphasis than ever before. Numerous demands or considerations coming from various aspects, such as geology, regulations, local environment, economic factors, and disaster management, are now needed to balance by engineers. The tasks that a ventilation engineer currently taken are no longer just performing simple ventilation surveys and routinely recording ventilation parameters any more, but instead of a number of complex calculations, comparisons, interdisciplinary decisions. Generally speaking, an eligible mine ventilation system of a modern coal mine should have the following characteristics: (a) Simple and practical; (b) Safety and reliable ventilation facilities (Chang, 1987); (c) Steady underground air flow; (d) Low ventilating resistance and reasonable distribution (Xenergy, 1997, 1998); (e) Strong capabilities to prevent disasters (Zhou, 2009). Concentrating on mentioned problems in the mine ventilation engineering, many researchers started numerous researches

167

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

including planning, optimization, environmental control, automation, monitoring, etc. to find solutions. For the mine ventilation systematic reliability design. Chen et al. (2003) applied multiple fuzzy synthesis estimation methods to assess the safety and reliability of mine ventilation system. In addition, a simulation program SIMURES was developed to model the reliability and maintainability analysis of an iron ore mine (Kumar and Huang, 1993). The ventilation design strategy has also been improved a lot in recent years. Engineers now have considered more factors

Fig. 1. Structure of CIMVCM program package.

in mining practices in addition to VOD (Ventilation on Demand) (Rocque and Sletmoen, 2002; Isaksson et al., 2009), such as hazards reduction (Rawlins, 2006; Brake, 2009; Panigrahi et al., 2009), and face ventilation layout (Zhang et al., 2009). To upgrade or optimize a mine ventilation system, some researchers also carried out a lot of work and obtained well outcomes (Myasnikov and Patrushev, 1981; Prosser et al., 2002; Roman et al., 2002; Loring and Nelson, 2006; Wallace and Sletmoen, 2009). For evaluating the system, back to 1960s or 1970s, only one or two single criteria was popular used to evaluate a system if it can be ventilated easily or not, such criteria including equivalent orifice (Murgue, 1883), total air quantity to working faces, ventilation efficiency, etc. However, Because of the complexity that a modern coal mine ventilation system has, in order to consider all influence factors, comprehensive evaluation models have been gradually developed and also accepted by researchers. Zhou and Wang (2002) tried to make the selection procedure into a mathematical process using twelve quantitative indices to evaluate candidate plans. Fault tree analysis method was also used for evaluating the underground mine escape way (Goodman, 1988). However, although a great of research works has been carried out so far, such new findings and developments in recent years are not systematically summarized. Hence, this research work has contributed to an improved ability to control and operate the mine ventilation system for engineers when they face some complicated and challenge problems in practices. In order to aid their working and improvements on the mine ventilation system, a PC-based computer package CIMVCM (Comprehensive and

Fig. 2. CIMVCM main screen.

Overall reliability of a mine ventilation system

Goal level

Criteria level

Technique ability

Alternatives Mine ventilation power level

Complexity

Mine ventilation network structure and pattern

Ventilation facilities

Important

Economic

Mine atmosphere monitoring

Disaster prevention facilities

Fig. 3. Schematic for analytical structure using the AHP method.

Task

Ventilation managements

168

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

Integrated Mine Ventilation Consultation Model) has been developed. The following sections start with briefly introducing the basic mathematical models which make up the backbones of the computer program package. Key mathematical equations to derive the solutions are going to be discussed and provided. Then, the structure and usages of CIMVCM are described based on the presented mathematical models. Finally, applications of program are also listed.

2. Description of mathematical models and CIMVCM program package Underground mining operations require fresh air for workers to breathe and equipment working. As a mine ventilation engineer, although his routine works are to control or manage the mine ventilation facilities, such as mine main fan, booster fan, various ventilation doors or windows, to safely provide enough fresh air quantity into the underground working areas, some challenges may show up when the following problems arise:     

Designing a new ventilation system. Upgrading the existing mining system. Rating the performance of a mine ventilation system. Evaluating the reliability of a mine ventilation system. Encountering the underground accidents.

Challenges that ventilation engineers are now facing are serious than any time before in history. Therefore, besides the traditional tools, engineers are required to use computer-aided techniques to analyze and/or solve any potential ventilation problems. In this section, a series of following subsections would briefly introduce various mathematical methods/models to derive solutions for some typical problems in ventilation practices. All these can provide helps for mine ventilation engineers to accurately and scientifically make ventilation engineering decisions. As an integrated solution for aforesaid mine ventilation engineering consolation works, a feasible and user friendly computer program which can cover all the previous research efforts should be developed. It is required that such computer program be capable of assisting ventilation engineers to make decisions. In detail, users can easily input data while the computer can automatically perform data preparations, information processing and transformation to final outputs with schematic or tabular views which are very helpful and useful for users to conduct the secondary analyses or to prepare proper management strategies.

Hence, the CIMVCM (Comprehensive and Integrated Mine Ventilation Consultation Model) program is designed based on the developed mathematical models. There are five tasks are accomplished by five modular programs contained in the program package, namely RELAOC, PLANSLE, COMEVA, RELASM and WARNING, respectively. Functionally speaking, the program package includes two designing programs, two evaluation programs and one prediction program. Fig. 1 shows the CIMVCM program structure and Fig. 2 is the opening screen and main menu of the program package. The capabilities of each of the modular programs are presented after descripting the mathematical models in the following sections. 2.1. Reliability allocation model During the time of modern mining extensively applied, underground mines have been expanded larger and larger than before. The longwall face with an ultra-length (>400 m) and the fast advance mining rate (>20 m/day) become more and more common nowadays which contribute to increasing mine’s coal annual production a lot. Due to such changes in mines, the ventilation layout today is totally different. The number of underground roadways needed in the modern coal mines has been greatly reduced comparing to the old mines. Hence, development of an underground coal mine is no longer a very heavy work anymore and, from the viewpoint of the Graph theory, the structure of ventilation system is much ‘‘simpler’’. However, it should be noted that although the system is simplified, its reliability is much more emphasized on than ever before. The reasons are obviously. A simpler system often yields to a poor redundant design, which may decrease the system’s anti-disaster ability. In reality, failures of a mine ventilation system do occasionally happen. Most of such failures can result in potential risk for miners. Investigation reports of mine fatalities in some coal mines show that defects in the mine ventilation system still exist in underground mines and are considered as the root of accidents. Hence, the prevention of a mine ventilation system failure is, no doubt, a challenge for mining engineers. However, how to design a successful and reliable mine ventilation system is a complicated task. Various influence factors (such as natural factors: geological conditions, gas content, and coal spontaneous combustion propensity; technique factors: ventilation capacity; underground air leakage rate, and transportation/delivery capacity; or others) have become more and more important to affect any ventilation managements carried out. Technically speaking, a good method to design a system must be well thought out the relationships between these factors, including difficulties, techniques and

Table 1 Inputs and outputs in reliability allocation model. Model inputs Name Mine ventilation power Description

By

Values

Power to move mine air circulation in underground condition of

Network structure and pattern Underground roadways’ connection Technique ability

Facilities

Mine atmosphere monitoring

Disaster prevention facilities

Ventilation Managements

Mainly referring to ventilation controls Complexity

A system to provide real-time environmental information within underground mines

Special facilities to control or limit mine disaster (Fire or explosion) expansion

Management Strategies for mine ventilation system

Importance

Economic

Task

Choose one from 1, 3, 5, 7 and 9 (numbers are used to represent each of grades. Due to different condition meanings, different descriptions are made for evaluation) Choose a value of 0.1–1 as interval of 0.1; Representing system’s fuzzy degree reliability A value from 0 to 1; Reflecting the system’s reliability

Cut set a Objective global Model outputs A reliability allocations table, which provide assigned reliable allocations under various conditions of decision-making, different subsystems and fuzzy degrees. In an actual design, the above table can be consulted to determine proper reliable values and improve the design quality based on the mine’s reality

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

169

Fig. 4. Screenshots of RELAOC.

economic costs, etc. and the system’s overall reliability requirements. Mathematically speaking, the reliability allocation can be expressed in Eq. (1)

8  < RS ðR1 ; R2 ;    Ri ;    Rn Þ P RS :~ g S g S ðR1 ; R2 ;    Ri ;    Rn Þ < ~

ð1Þ

where Rs is the reliability index of a mine ventilation system; RS is the expected reliability value of a mine ventilation system. ~ g s is the constraint condition of a mine ventilation system design.

~ g s is the maximum constraint condition of a mine ventilation system design. Ri is the reliability index of the ‘‘i’’ th subsystem. The allocation model must well consider the indeterminate problems in both the decision-making process and the system itself, and then be able to achieve the optimum reliability allocation. Hence, a model scientifically allocating the reliability practice is proposed into the mine ventilation systems design process. In the technical standpoint, the model is consisted of two stages. First, based on previous research findings, the hierarchical structure of a

170

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

Fig. 4 (continued)

Mine head

Mine equivalent orifice

Technology

Ratio of air quantity Network reasonable coefficient Ventilation efficiency

Fan power The Optimal Index System

Economic

Fan efficiency

Second, the proposed reliability allocation model using the fuzzy mathematics calculation is applied to complete and optimize the reliability allocation works. By establishing the fuzzy comprehensive assessment matrix and introducing the concept of ‘‘entropy weight’’, h, in the informatics science (to quantify the uncertainty involved in predicting the value of a random variable), once the total reliability value for a mine system is set as Rs , the reliability allocation for each subsystem can be calculated by the following system of equations:

8 > > > > R ¼ > < 1

Rs =

6 Y hi

!1=6  h1

i¼1

!1=6 > 6 > Y > hk > > : Rk ¼ h1 R1 ¼ hk  Rs = hi

k ¼ ð2; 3; . . . ; 6Þ

ð2Þ

i¼1

Power cost

Stability of fan

Safety

Disaster prevention ability

Ventilation management

Fig. 5. The optimal selection index system.

mine ventilation system is identified by the analytic hierarchy process (AHP) method. Basically speaking, there are three different levels in the structure, which are the goal level, the criteria level and the alternatives level, respectively. The goal level is the total reliability of a mine ventilation system. The criteria level consists of subsystems in a ventilation system (six subsystems are classified here). The alternatives level mainly refers to the influence factors that can affect the mine system’s design. Fig. 3 shows the analytical structure.

The solutions that are derived by the developed model are more accurate and scientific. Therefore, the allocation procedure of this model would ensure that the designed mine ventilation system is the more reasonable and has higher engineering quality. In addition, this model can also help the designers to continue the reliability allocation within the low level subsystem. To use this model, Table 1 lists the data which should be collected as inputs as well as outputs. Descriptions on these data are also shown. RELAOC – This program combining the analytic hierarchy process (AHP) with the fuzzy calculation theory can carry out the reliability allocation for each mine ventilation subunit. Once the descriptions of each influence factor with respect to a subunit and the weighting assessment vector are defined, the program can give out the reliability allocation results under different combinations of k and a. The cut set a can reflect the system’s fuzzy degree. A great value of a means the system has small fuzziness and clear concepts and boundaries. The fuzziness that a decisionmaker has during the process of designing a mine ventilation system can be reflected by the cut set k. The bigger the k is, the greater

171

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180 Table 2 Inputs and outputs in optimal candidate plan selection model. Index name Model inputs Mine head Mine equivalent orifice Ratio of air quantity Network reasonable coefficient Ventilation efficiency Fan power Fan efficiency Power cost Stability of fan Disaster prevention ability Ventilation management

Unit

Notes

Pa m2 (an simple index to assess ventilation resistance) – (Rating score by experts) – (Rating score by experts) % Kw % $/ton – (Rating score by experts) – (Rating score by experts) – (Rating score by experts)

All parameter values for each plan must be collected

Model outputs The resulting sequence of relative degree rk, the best plan can be selected

Fig. 6. Screenshots of PLANSLE.

the uncertainty is made in the decision-making process. Ventilation engineers can follow any ‘‘guidelines’’ when designing a system based on their design strategies. It should be noted that whatever the combination of k and a is, the goal of system’s overall reliability can be maintained. Fig. 4 is the screenshots. 2.2. Optimal candidate plan selection model Large variations in geological conditions and in the development and production requirements necessitate the need to consider multiple factors in the selection of mine ventilation system plans. At that time, ventilation engineers should not only consider current ventilation needs required by the underground production,

but also plan the system’s flexibility for the future. In order to give full considerations to these factors, a well and synthetic index system including all aspects of ventilation system must be developed. This indexing system chooses sub-indexes from the economical, technical and safety aspects as shown in Fig. 5. The Set Pair Analysis (SPA) theory which was developed by Zhao (2000) is a kind of system analysis method and its main concept is to combine both the certain and uncertain phenomena into a system of certainty and uncertainty. Then the quantitative description and transition relationship could be represented by a mathematical model for finding reasonable results. The most important procedure when using SPA is that to calculate and generate the degree of fuzzy correlation as:

172

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

Ratio of mine air quantity Underground highest temperature Qualification and quantity of mine air

Methane concentration Methane ultralimit frequency Air Qualify in working faces Frequency of working face or development head in series Number of instability diagonal branch

Rationality of network Number of independent mesh Number of diagonal branch Rationality of air quantity self-adjustment Underground ventilation facilities

Number of facilities per thousand meters Quality of ventilation facilities

The Evaluation Index System

Quality of disaster prevention facilities Quality of air –flow reversing system

Disaster prevention ability

Failure rate of Safety monitoring system Network self-adjusting ability Capability of professional

Management

Perfect rules and regulations Mine head Stability of mine fan Mine fan

Flexibility of air-flow reversing system Mine fan capability Mine fan efficiency

Fig. 7. The evaluation index system.

Fig. 8. The TWW function.

lðsk ; uÞ ¼ ak þ bk  i þ ck  j

ð3Þ

where the coefficients in the equation are the identical (ak), opposite (bk) and differential (ck) subordinate degrees with considering different weighting factor (W) for each index as:

ak ¼

n X wr  akr

ð4Þ

r¼1

ck ¼

n X wr  ckr

ð5Þ

r¼1

bk ¼

n X wr  bkr r¼1

ð6Þ

The meaning of akr or ckr in the SPA analysis represents how close the candidate plan Sk is to the best or worst plan. It is a piece of information with certainty. So the relative nearness degree rk is k defined as: r k ¼ a aþc . Thus, we can compute the value of rk for a k k set of candidate plans. If max fr k g ¼ rki , the plan Sk can be consid16i6m ered to be the best one. For each candidate, featured parameters must be collected in order to proceed the optimal plan selection work. Table 2 lists all needed parameters. PLANSLE – In the planning stage for a new coal mine or for a change or upgrade to an old mine, multiple viable ventilation plans could be proposed to cope with varying geological conditions, production rates, mining laws and regulations, etc. Each of the plans may put emphasis on one or a number of different considerations. Therefore, this program is used to assist to select the best ventilation plan. Users need to collect all technical parameters listed in Fig. 2 for each candidate plan and input them into the program. Based on the mathematical calculations, the program can generate the order of plans based on the relative nearness degree rk. The big advantage of this method is that some severe influences caused by anthropogenic factors or single index can be avoided in the final result. Therefore, the selection is the more rational and economical among the proposed viable plans. Fig. 6 is the screenshots.

173

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180 Table 3 Recommended values for different ak (Cheng et al., 2010). First level

Second level

Qualification and quantity of mine air

Ratio of mine air quantity

Units

Better class

Medium class

Worse class

°C

Methane concentration

%

1:2 6 x21 < 1:3 26 6 x22 < 30 0:5 6 x23 < 0:75

1:3 6 x31 < 1:5

Underground highest temperature

1:1 6 x11 < 1:2 24 6 x12 < 26

0:1 6

Rationality of network

Underground ventilation facilities

Disaster prevention ability

Mine fan

Mine fan

Time

1 6 x14 < 2

2 6 x24 < 3

3 6 x34 < 4

Score

90 6 x15 < 95

80 6 x25 < 90

70 6 x35 < 80

Frequency of working face or development head in series Number of instability diagonal branch

%

x26

20 6 x36 < 25 3 6 x37 < 4

50 6 x28 < 100

100 6 x38 < 150

Number of diagonal branch

10 6 x19 < 20

20 6 x29 < 30

30 6 x39 < 35

Rationality of air quantity self-adjustment

0:5 6 x110 < 1

Number of facilities per thousand metes

1 6 x111 < 3

1 6 x210 < 1:25 3 6 xt11 < 7

7 6 x311 < 10

1:25 6 x310 < 1:5

Quality of ventilation facilities

%

95 6 x112 < 99

90 6 x212 < 95

85 6 x312 < 90

Quality of disaster prevention facilities

Score %

95 6 x113 < 98 1 1 6 x115 < 2

93 6 x213 < 95

Quality of air-flow reversing system Failure rate of Safety monitoring system

2 6 x215 < 3

91 6 x313 < 93 0 3 6 x315 < 4

1 6 x116 < 1:25

1:25 6 x216 < 1:5

1:5 6 x316 < 1:7

Capability of professional

Score

90 6

Perfect rules and regulations

Score

80 6

x117 x118

< 95

85 6

< 90

70 6

x217 x218

< 90

80 6 x317 < 85

< 80

60 6 x318 < 70

Mine head

Pa

1000 6 x119 < 2924

2924 6 x219 < 3538

3538 6 x319 < 4368

Stability of mine fan

Score

80 6 x120 < 95

50 6 x220 < 80

20 6 x320 < 50

Flexibility of air-flow reversing system

Score

90 6 x121 < 95

85 6 x221 < 90

80 6 x321 < 85

Mine fan capability

%

1:2 6 x122 < 1:3 80 6 x123 < 85

1:1 6 x222 < 1:2 70 6 x223 < 80

60 6 x323 < 70

X2

X3

X4

X5

X6

X7

X8

X9

X10

X11

1.05

23

0

0

100

0

0

0

0

0

0

1.70

42

2

5

60

30

5

200

60

3

25

X12

X13

X15

X16

X17

X18

X19

X20

X21

X22

X23

x0j

80

90

0

0

70

30

200

0

70

1

50

x5j

100

100

5

2

100

100

4410

100

100

1.2

90

2.3. Comprehensive evaluation model In order to conduct a comprehensive evaluation for a mine ventilation system, the indexes must roundly and in-depth be selected to fully characterize the system’s characteristics. Fig. 7 shows that indexes are picked from the six aspects. As a mine ventilation system is a typical gray system, the author has developed a new evaluation method combining gray cluster analysis and fuzzy evaluation (Cheng, 2008). In general, gray cluster analysis (Liu and Guo, 1999) is used to build the evaluation matrix while fuzzy theory is applied to obtain the final score. In detail, with the help of the triangular Whitening weight (TWW) function (as shown in Fig. 8), the index j to the kth gray class is obtained and can be expressed as:

x R ½ak1 ; akþ2 

kk ak1 akþ2 x akþ2 kk

< 20

2 6 x27 < 3

X1

> > :

10 6

1 6 x18 < 50

x0j x5j

¼

< 10

1 6 x17 < 2

Symbol

k1

56

x16

Number of independent mesh

Table 4 Recommended values for a0 and as+2 (Cheng et al., 2010).

k f j ðxÞ

< 0:5

Methane ultra limit frequency

Mine fan efficiency

8 0 > > < xa

0:75 6 x33 < 1

Air qualify in working faces

Network self-adjusting ability Management

x13

30 6 x32 < 34

x 2 ½ak1 ; kk  x 2 ½kk ; akþ2 

where j = 1, 2, . . ., m and k = 1, 2, . . ., s.

ð7Þ

1:05 6 x322 < 1:1

If the value of Xij (measurement value) is obtained, the degree k k of membership f j ðxÞ can be calculated. It can be seem that f j ðxÞ is actually a set of fuzzed values with respect to different gray class. However, the integrated cluster coefficient rki calculated P k from rki ¼ m j¼1 f j ðxij Þ  gj , where gj is weighting factor could be used to build a i  k matrix which is so called the fuzzy compree1; C e 2;    ; C e n ÞT . On the other hand, by define ¼ ðC hensive matrix R ing the assessment matrix (C) is C = {Best(C1), medium(C2), e The equation worse(C3)} and the weighting factor matrix A, eR e is used to calculate the finial rating score (see Tables 3 A and 4). COMEVA – For a running mine ventilation system, one of urgent requirements is to evaluate its effectiveness and efficiency in order to help mine operators more understand and control the system. By selecting indexes that are able to reflect all characteristics of a mine ventilation system, this program can evaluate the system with a rating score based on the mathematical model of gray cluster analysis-fuzzy theory. Like PLANSLE, Users need to input the measured field data for each index (Listed in Fig. 7) when an onsite mine investigation has been carried out. Then, the result can be derived. Fig. 9 is the screenshots. 2.4. Reliability assessment model A mining system is a complex system. The interactions between the components can greatly affect the behavior of the entire system. For example, Even if only a small part of the system cannot work properly, severe consequences can happen which may break down the entire system. Generally speaking, to quantify the reliability of a mine ventilation system, there are two categories of approaches: the analytical method and the random simulation

174

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

Fig. 9. Screenshots of COMEVA.

method. However, although the analytical method has a very clear mathematical concept and an expression form to determine the evaluation results, there are some big drawbacks associated with

it. It is only good for handling a small scale system with a number of limited components and fails to consider the random characteristics of each component in the system. By Comparison, the

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

Fig. 10. The basic principle behind Monte Carlo simulation (Wittwer, 2004).

175

random simulation method receives more and more attentions so far. The Monte-Carlo Simulation (MCS) method is going be used for performing the reliability assessment. Basically, the Monte Carlo simulation can be categorized as a sampling test. The aim of the simulation is to obtain the results’ distributions based on various inputs that are randomly generated from probability distributions which closely match actual data we already have to simulate the process of sampling from an actual population. Fig. 10 illustrates an example to show the basic principle behind Monte Carlo simulation. To examine the reliability of a mine ventilation system, a typical mine ventilation network is chosen as the study problem. The very first step of setting up the reliability model is to create the probability function for each of the variables involved in the study problem. For a mine ventilation system, the mine roadway is the basic element in a system and the function of a mine roadway is to deliver enough air quantities to specific sites in the mine. Therefore, the corresponding air quantity probability distribution function of each roadway (variables) should be created. Then, the next step is to set up the failure criteria for a successful ventilation system. Whether the airflow quantity that a mine roadway can transport is sufficient or not that is very important for mine ventilation system. In ventilation engineering, the failure of a mine roadway can

Fig. 11. Screenshots of RELASM.

176

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

Fig. 11 (continued)

Fig. 12. Structure of early warning model with integrating RS and SVM.

be defined as the airflow quantity in this roadway is less than the required minimum air quantity or more than the allowable maximum quantity. The constraint group of a mine roadway can be shown as:

Table 5 Allowable entry air velocity. Type of roadway

Ventilation shafts Hosting shafts Smooth lined main airways Main haulage routes Conveyor drifts Working faces Entries Belt entries

Allowable velocity (m/s) Min.

Max.

7 4.3 3.7 2.8 2.3 1.6 1 0.8

20 10 8 6 5 4 3 2.5



Q Min ¼ V Min  S Q Max ¼ V Max  S

ð8Þ

where Q is the airflow quantity, m3/s; S is the entry cross sectional area, m2; Vmin is the minimum air velocity allowed in the regulation, m2/s; Vmax is the maximum air velocity allowed in the regulation, m2/s;

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

177

Fig. 13. Screenshots of WARNING.

The values of Vmin and Vmin are listed as Table 5. Various types of mine roadways have different allowable values. The final step is to carry out the stochastic sampling for involved variables in the reliability model of the studied mine ventilation network. By regularly increasing the numbers of sampling times, relevant descriptive statistics indexes are examined to check whether the simulation will converge. Outputs (Reliability index and other descriptive statistics indexes) are prepared and the simulation is terminated once everything meets requirements (see Table 6). RELASM – This program uses the Monte-Carlo Simulation (MCS) to evaluate the reliability of a mine ventilation system. Com-

paring to the traditional analytical assessment methods, this method considers random characteristics existing in the studied ventilation system. Hence, some severe influences caused by random events happened in a mine ventilation system can be taken into account for the final assessment result. Users are required to input all monitored airflow quantity data for each underground roadway and the corresponding failure criteria during a period of observation time into the program. RELASM is to deduce the reliability assessment result by performing complicated stochastic sampling and statistical modeling. This program can well help mine operators examine the safety of the mine ventilation system. Fig. 11 is the screenshots.

178

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

Fig. 13 (continued)

Table 6 Inputs and outputs in reliability assessment model. Model inputs A sampling set of the airflow quantity in a roadway must be obtained as Q1, Q2, . . ., Qn, during the period of observing time as well as other underground roadways. By using these data, the corresponding air quantity probability distribution functions for each roadway can be created Computer simulation procedure The Oracle Crystal BallÒ, spreadsheet-based software, is used to perform the Monte-Carlo Simulation (MCS) Model outputs Reliability, R

Reflecting the system’s overall Reliability

2.5. System early warning model The goals of protecting miners’s safety and eliminating the healthy risks are important responsibilities which a mine ventila-

tion engineer should take. Hence, they are required to continually monitor and track the ventilation system’s safety level and to implement different ventilation managements in order to prevent any ventilation-related accidents and unnecessary deaths. However,

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180 Table 7 Inputs and outputs in warning model. Attribute name Model inputs Ratio of mine air quantity Methane ultra limit frequency Frequency of working faces or development head in series Number of instability diagonal branch Number of facilities per thousand meters Percentage of qualified ventilation facilities Quality of disaster prevention facilities Failure rate of safety monitoring system Capability of professional Mine head Stability of mine fan Mine fan efficiency

3. Applications of CIMVCM

Unit

Notes

– Times

The number of attribute would be reduced due to the RS calculation. Hence, less parameters are used as inputs for SVM model

%

179

Over the past decade, the authors have been constantly worked on various mine ventilation projects. Based on these invaluable experiences, the CIMVCM program package was developed to meet the needs in the field of mining ventilation practices and researches. Due to its demonstrated usefulness, the program packages have been successfully applied in numerous cases researches. Some of the many successful application cases are listed here:

– – % – (Rating by experts) % Scores Pa Scores %

Model outputs (Early Warning Risk level) Level 1 Reliable Level 2 Normal Level 3 Dangerous Level 4 Failure

the coal mine ventilation is a super complicated system. Lots of influence factors could control or impact the behaviors of system. Hence, a running mine ventilation system may not be kept as constant, and changes timely due to the dynamic production. Stating from the quantitative point, the system shows a fluctuant wave around a certain value. However, if this deviation degree is too large to be controlled by the system, it may indicate that any potential risks existing. An integrated early warning model is proposed to improve the mine ventilation safety due to such above considerations. The model itself is comprised of two sub models. One is called the optimal selection model and the other is. The optimal selection model is with the help of Rough Set theory using the attributes reduction as a data processing tool to largely minimize the dimension and the complexity of the input matrix for the evaluating sub-model. Then, support vector machines (SVM) based the risk evaluation model is going to predict the mine ventilation system’s risk level. Fig. 12 shows the structure of the integrated model. It can be seen that the data are processed within two different steps. Firstly, the best (attributes) indexes which can fully characterize the mine ventilation system are selected from various indexes based on the attribute reduction technique with applying Rough Set theory. It is good for improving the evaluation efficiency in the next step. Then, the pervious step results are used as input parameters used in the risk evaluation model to perform the classification. Finally, the risk evaluation results can be obtained (see Table 7). WARNING – This program can predict the system’s overall potential risk levels, which are characterized ‘‘Reliable’’ (1), ‘‘Normal’’ (2), ‘‘Dangerous’’ (3), and ‘‘Failure’’ (4), by analyzing the ventilation indexes. Currently, twelve indexes are set as default values and fifteen case samples are collected to set up the SVM training matrix. However, the program allows users to change indexes and also updates training matrix for better accurate and scientific prediction results. The program could be used as a useful tool not only to track the status of the mine ventilation system timely but also to assess the upgraded one. Based on the results, mining engineers could take proper measurements to avoid any potential accident risks. Fig. 13 is the screenshot.

 Reliability allocation designs of subsystems or units in mine ventilation for several mines have been well guided. CIMVCM has been applied to design or adjust the reliability considerations when a mine ventilation system is in initial designing or upgrading (Cheng and Yang, 2013; Cheng et al., 2014).  Optimizing and updating a number of mine ventilation systems with different unique characteristics and production histories (Zhang et al., 2007; Cheng, 2008; Cheng and Yang, 2008a, 2008b; Cheng et al., 2010). Using CIMVCM, the severe influence caused by anthropogenic factors or single index can be avoided in the final result.  Assessing and rating several mine ventilation systems’ performance. By using the results, mine operators ensure or improve the mine ventilation system to achieve more rational and economical state (Yang et al., 2006; Cheng and Yang, 2007; Cheng, 2008, 2013; Sun et al., 2008).  Mitigating accidental risks and protecting miners in mine ventilation system. Early-warning can help mining engineers take proper measurements to avoid any happening of accidents (Cheng and Yang, 2012, 2013). 4. Conclusions Accurate evaluation and prediction of a mine ventilation system can provide very valuable information for mine operators to take proper measurements for improving the mining safety. In order to copy with the complicated problems within the subject of mine ventilation engineering, a PC-based computer package CIMVCM (Comprehensive and Integrated Mine Ventilation Consultation Model) has been developed. This program package provides a versatile, reliable and easy-to-use tool for those engineers who are involved in the ventilation management. Five module programs, namely RELAOC, PLANSLE, COMEVA, RELASM and WARNING, are included in this package and each of them performs a special task. Due to the solid theories used and extensive field validations, the ventilation consultation program package has been proven to be a versatile and reliable Decision Support System (DSS) tool in the field of mine ventilation engineering. However, the following points should be improved for the future research; (a) the accuracy of some models (RELAOC, PLANSLE and COMEVA) is heavily dependent on the ventilation indexes. It can be seen that the current index system may not well fully characterize the ventilation system. Therefore, the optimal selection of ventilation indexes for different mine scales or their unique layout should be further modified and improved; (b) it is suggested the program package can be linked with the in-mine environment monitoring system. Thus, the program can acquit data from the monitoring system directly which can dynamically reflect and check the ventilation system without any time delay. Acknowledgments This work is financially supported by grants from the Fundamental Research Funds for Central Universities (Grant No. 2013QNA01), the National Science Foundation of China (Grant

180

J. Cheng et al. / Tunnelling and Underground Space Technology 45 (2015) 166–180

No. 51304203), the Natural Science Foundation of Jiangsu Province of China for Youths (Grant No. BK20130191) and Specialized Research Fund for the Doctoral Program of Higher Education (Grant No. 20130095120001); the Program for Changjiang Scholars and Innovative Research Team in University (IRT13098) and Priority Academic Program Development of Jiangsu Higher Education Institutions; the authors are grateful for these supports. References Brake, D.J., 2009. The growing use of hazardous primary ventilation systems in hardrock mines. In: Panigrahi, D.C. (Ed.), 9th International Mine Ventilation Congress. Taylor & Grancis Group CRC Press, Dhanban, India, pp. 43–60. Chang, X., 1987. Transient-State Simulation of Mine Ventilation Systems. Michigan Technological Univ, Houghton (USA). Chen, K., Fu, Q., Liu, X., LI, X., 2003. Design and application of estimation software for safety reliability of mine ventilation system. J. China Univ. Mining Technol. Chin. Edition 32, 393–398. Cheng, J., 2008. Study on the Optimizing and Evaluating Coal Mine Ventilation System. China University of Mining & Technology, Xuzhou. Cheng, J., 2013. Assessment of mine ventilation system reliability using random simulation method. Environ. Eng. Manage. J. (accepted for publication). Cheng, J., Yang, S., 2007. A comprehensive evaluation model for ventilation effects of a longwall working face. China Saf. Sci. J. 34, 50–53. Cheng, J., Yang, S., 2008a. Upgrading mine ventilation system at Tangshangou mine. Ind. Saf. Environ. Prot. 34, 50–53. Cheng, J., Yang, S., 2008b. Mine ventilation optimization planning based on fuzzyset pair theory. Coal Technol. 28, 86–89. Cheng, J., Yang, S., 2012. Data mining applications in evaluating mine ventilation system. Saf. Sci. 50, 918–922. Cheng, J., Yang, S., 2013. Applications of Process Safety and Assurance Techniques in Mine Ventilation Engineering. China University of Mining & Technology Press. Cheng, J., Yang, S., Luo, Y., 2010. Mathematical Models for Optimizing and Evaluating Mine Ventilation System. In: Hardcastle, S., McKinnon, D. (Eds.), 13th United States/North American Mine Ventilation Symposium. Laurentian University, Sudbury, Canada, pp. 278–285. Cheng, J., Zhou, F., Yang, S., 2014. A reliability allocation model and Application in Designing a Mine Ventilation System. Iranian J. Sci. Technol. Publ. 38, 61–73. Goodman, G.V., 1988. An assessment of coal mine escapeway reliability using fault tree analysis. Min. Sci. Technol. 7, 205–215. Hartman, H.L., Mutmansky, J.M., Ramani, R.V., Wang, Y.J., 1997. Mine Ventilation and Air Conditioning. Wiley-Interscience. Isaksson, M., Andersson, P., Ekenstedt, F., 2009. Ventilation on demand: systems used by boliden mineral AB, Kristineberg Mines, 1987-2008. In: Panigrahi, D.C. (Ed.), 9th International Mine Ventilation Congress. Taylor & Grancis Group CRC Press, Dhanban, India, pp. 105–112. Kumar, U., Huang, Y., 1993. Reliability analysis of a mine production system-A case study. In: Reliability and Maintainability Symposium, 1993. Proceedings., Annual, IEEE, pp. 167–172. Liu, S.F., Guo, T.B., 1999. Gray System Theory and its Application. Scientific Press, Beijing, China. Loring, D.M., Nelson, B.V., 2006. Transition of the Henderson mine ventilation system to the new lower levels. In: Mutmansky, J.M., Ramani, R.V. (Eds.), 11th

United States/North American Mine Ventilation Symposium. Taylor & Francis Group, Penn State University, College Park. USA, pp. 41–45. Murgue, D., 1883. Theories and Practices of Centrifugal Ventilating Machines. London. available from: . Myasnikov, A.A., Patrushev, M.A., 1981. Principles of Coal Mine Ventilation Planning. Amerind Publishing Co. Pvt. Ltd, New Delhi. Panigrahi, D.C., Bhowmid, S.C., Acharya, D., Gupta, R., 2009. Design of ventilation system for reducting the radiation dose level in india uranium mines. In: Panigrahi, D.C. (Ed.), 9th International Mine Ventilation Congress. Taylor & Grancis Group CRC Press, Dhanban, India, pp. 21–22. Prosser, B.S., Stinnette, J.D., Paredes, J., 2002. Ventilation optimization at the La Camorra mine. In: Souza, D. (Ed.), 9th United States/North American Mine Ventilation Symposium. Kingston, Ontario, Canada, pp. 57–64. Rawlins, C.A., 2006. Underground mine ventilation planning, heat loads, and diesel equipment. In: Mutmansky, J.M., Ramani, R.V. (Eds.), 11th United States/North American Mine Ventilation Symposium. Taylor & Francis Group, Penn State University, College Park. USA, pp. 75–81. Rocque, P., Sletmoen, L., 2002. Ventilation design for increased production at the red lake mine. In: Souza, D. (Ed.), 9th United States/North American Mine Ventilation Symposium. Kingston, Ontario, Canada, pp. 65–68. Roman, W., Guthrie, J., Ndhlivu, W., 2002. Upgrading ventilation systems at Konkola mine, Zambia. In: Souza, D. (Ed.), 9th United States/North American Mine Ventilation Symposium. Kingston, Ontario, Canada, pp. 77–84. Sun, Y., Yang, S., Cheng, J., 2008. A dual-assessing Model to Evaluate Mine Ventilation System Using Gray Correlation Method. Coal Mine Saf. 42, 44–47. Wallace, K.G., Sletmoen, L., 2009. Ventilation upgrades at the red lake comples. In: Panigrahi, D.C. (Ed.), 9th International Mine Ventilation Congress. Taylor & Grancis Group CRC Press, Dhanban, India, pp. 83–92. Wikipedia, 2013. Underground Mine Ventilation. Wittwer, J.W., 2004. Monte Carlo Simulation Basics. Xenergy, 1997. United States Industrial Electric Motor System Market Assessment. Interim Report Prepared for Oak Ridge National Laboratory, TN, pp. D1–D5. Xenergy, 1998. United States Industrial Electric Motor Systems Market Opportunities Assessment. Final Report Prepared for Oak Ridge National Laboratory, TN, pp. E9–E11. Yang, S., Ou, X., Cheng, J., 2006. Micro-climate evaluation system in thermal mines. J. China Univ. Min. Technol. 16, 1–4. Zhang, Y., Tien, J.C., Wang, D., 2009. A critique of existing ventilation under downhile coal mining conditions. In: Panigrahi, D.C. (Ed.), 9th International Mine Ventilation Congress. Taylor & Grancis Group CRC Press, Dhanban, India, pp. 93–104. Zhang, Z., Yang, S., Cheng, J., 2007. Application of MATLAB in Mine Ventilation Optimization Simulation. Coal Mine Modernization, 49–51. Zhao, K., 2000. Set Pair Analysis and its Application. Zhejiang Science and Technology Press, Hangzhou, China. Zhou, F.B., Wang, D.M., 2002. Fuzzy optimum selection and analysis for optimization and evaluation of mine ventilation systems. J. China Univ. Min. Technol. 3, 62–69. Zhou, L., 2009. Improvement of the Mine Fire Simulation Program MFIRE. West Virginia University, Morgantown.