| Coal is an important basic energy in our country, and occupies an important strategic position in the national economy. Coal mine safety production is directly related to economic construction in long-term sustainable development and the implementation of the energy strategy,and plays a decisive role in building a moderately prosperous society and modernization construction. Therefore,we must attach great importance to coal mine safety production,to ensure the coal industry sustained, stable and healthy.In recent years,China has developed a series of policies and measures to strengthen mine safety management efforts. As in2010,total of2045small coal mines closed for they didn’t have the conditions for safe production in China. Under the guidance of national policy, although the total amount of coal mine accidents decreased, one million tons mortality rate dropped from5.07in2001to0.564in2011,but it is far cry compared with the international advanced countries, such as the U.S.2004-2006one million tons coal mortality was only0.027,0.021and0.045. China is one of countries which have the highest frequency of mine casualties and the largest number of casualties around the world. In addition, the rapid development of network communicaton technology, computing technology and other technologies put forward higher requirements on the safety management of coal mining enterprises,and coal mining enterprises need to take full use of modern technology to speed up the construction of information technology to enhance the safety management level, thus to build a "high-information, automated, high-efficiency"digital mine. IT and higy-tech advances in digital technology for the construction of coal mines to provide a theoretical basis and technical support. Through this research, it helps to identify the cause of coal mine accident hazards, improves the level of safety production and management.In order to analyze the deep-seated reasons of coal mine safety accidents,grasp the main factors affecting safety in production,evaluate the standards of coal mine safety production, and promote the improve of coal mine safety production situation. This thesis combines the characteristics of coal mine safety production,and uses modern and advanced scientific theroretical methods for coal mine safety production activities for dynamic monitoring and early warning,to build a warning index system of coal mine safety and warning model in real-time monitoring and correction,so as to provide theoretical support and technical support for preventing and reducing coal mine accidents. The main conclusions of this thesis include:First,based on the define of security incidents related,and propose causal correction accident,accident consequences randomness,irreparable consequences of accidents and human controllability of four characteristics; based on the existing coal mine accident causation theory, analyses the five categories of coal mine accident factors:management factors, human factors, equipment factors,environmental factors and information factors,we propose an accident causing mechanism for a controllable incidents and compliance of coal mine safety production management practice,that is, since the human mine production systems,equipment and the environment caused by defects in an unsafe condition, with the defect in information and management, resulting in coal mine accident,after emergency rescue, formed the eventual loss of the accident.Second, for the complex geological conditions of coal mine situation, this thesis puts forward that coal industry system has a complex,dynamic variability,change fuzziness and randomness,and emergency rescue and arduous handing task of disaster incident three characteristics,and explains the significant value of coal mine safety assessment based on the principle of "safety is important,prevcaution goes first"; Moreover,through analysis of the situation for the present study found that mine safety indicator system and preferably lack of uniform standards and methods,according to most scholars own experience,indicators follow certain principles to cause subjectivity is too strong, can not reflect the objective existence prone to bias. AHP, Delphi and fuzzy evaluation method are also difficult to avoid the drawbacks; using principal component analysis,the coefficient of variation method,neural networks and other statistical methods should be established on the basis of a large sample volume, large capacity and in most cases the sample can not be obtained. This thesis analyzes the advantages and disadvantages of each method, then proposes method based on information entropy index screening methods, the use of this method for coal mine safety assessment index system to optimize the initial selection.Third, this thesis explains the concept and advantages of fuzzy neural network,and puts forward the principle of fuzzy neural networks, specifically including input layer, fuzzy layer, hidden layer and output layer and defuzzification layer these five levels; Divide the qualitative indicators according to the corresponding provisions, and using the Delphi method to quanify; and empirical analysis,fuzzy neural network model can be seen in the coal mine safety assessment advantages,combined fuzzy theory and neural network technology, fuzzy neural network model both with learning,communication and adaptive capacity,but also has fuzzy thinking, you can make a more objective evaluation results resistance, a sample of known only by the study,it can get expert thinking, then the trained network can be directly used to simulate a sample to be evaluated, reducing human factors of evaluation; same time, using the appropriate network training, the weights obtained, you can know which factors influenced the degree of mine safety, what factors are important to determine the coal mine safety management focus.Fourth, this thesis defines the meaning of coal mine accident warning; gives a BP neural network and genetic algorithm mine safety waring assessement methods,BP neural network has good self-learning,adaptive, nonlinear parallel processing and computing power,so in intelligent control,nonlinear optimization, signal processing,and achieved a wide range of applications. Genetic algorithm is based on the principle of natural selection and genetics random search optimization method, has a strong global optimization capability. So combine the two, with the latter on the former weights and thresholds tuning,you can make the network has a faster convergence rate, so that not only played a neural network generalization mapping ability,but also avoid falling into local pole small problems,and through genetic algorithm optimization neural network weights and thresholds established coal mine accident warning model; combining the actual situation on the Yima Coal Group Stock CO.,LTD apply the accident early warning model, and found that GA-BP and BP neural network has better prediction accuracy and faster convergence,obtained genetic algorithm to optimize BP neural network is feasible and effective.Fifth, this thesis from early waring analysis and pre-control measures to achieve coal mine accidents early warning management,and on this basis,form the workflow of various functions of management to security warning center. Warning analysis includes monitorin, identification,diagnosis and warning assessment four aspects; pre-control measures includes organizational readiness, daily monitoring and emergency management three aspects. Various functional management departments of coal mining enterprises report regularly to the Security Warning Center operations management diagnostic status reports, and details of the pre-control measures taken and the results obtained. Safety warning center for real-time monitoring of early waring indicators, when monitoring results in a normal state,continue to monitor; when monitoring results indicate the alert of the stage, take appropriate measures through the pre-control measures library to rectify; When monitoring results indicate a state of crisis, establish immediately the Crisis Management Team to implement the rescue work until the production system pull through reachi a normal operating state. The measures taken to deal with crisis countermeasures feedback to the library,and through a long-term accumulation, accident warning management system achieves a strong immunity. In addition, the security warning centers are also responsible for the work of the entire coal mining enterprises organizing, directing and coordinating, and conduct regular checks and summary fo various department.In addition,this thesis gives the effective and failure of two angles to launch the appropriate management systems and mine waring state emergency management,and proposes the reasonable and effective coal mine accidents warning management system operating mode. Mine warning management system should be running around the "man-machine-environment-management-information" to carry out its activities. Warning management of causal factors will produce two results:an effective warning management of production system turned the corner, then the production system can continue to run; Failure warning management should be initiated emergency measures to stop the run into the adjustment phase. In the state of emergency management, through the implementation of timely and correct pre-control measures can promote the production system to return to normal production status, but the consequences of failure may cause significant of serious accidents. Whether enterprises are run by the state of crisis or emergency management into normal production status, the process and results of its occurrence parameters will feedback the database and countermeasure library of warning management system, so it is reasonable to adjust and optimize the process of the next cycle che early warning activities. Coal mine accident warning in accordance with the proposed principles of management theory shows that earl warning management including safety management failures,fluctuations in production management, mjor hazard monitoring, identification, diagnosis,evaluation, pre-control and so on.Sixth,this thesis provides both countermeasures and suggestions to further improve China coal mine production safety management security system based on the formal and informal institutions. Formal system,as the mainstream of social systems,mandatory features, which is an informal system construction provided the conditions. Country and coal mining enterprises conduct coordination control through formal system of coal mine safety production management,there are three areas:improving mine safety production standards,increase investment and strengthen coal mine production safety emergency rescue safeguards. Formal system of laws and regulations can not be separated, to increase its enforcement, create a healthy environment for the development of the coal mine safety production management. Formal institutional arrangements should also be combined with the specific circumstances around to adjust and re-defined, local governments need to make reasonable modifications to the relevant fixed and flexible disposition. Informal system, although not a substitute for the formal system, but it is also an integral part of the system, is a kind of spontaneous,non-mandatory, continuity and universality characteristics of conduct to regulate the industry,complementary harmony with the formal system relationship, it is not hard to use legal means to achieve such, it relies on the code of conduct, moral traditions and other flexible means. This informal institutional arrangemients consists:strengthen mine safety technical training,strengthen publicity and cultural construction of coal mine safety,strengthen exchanges and cooperation with foreign security technologies and strengthen information technology.Formal system provides the basis for the construction of the informal system, and an informal system to make up for the lack of formal institutions. The formal system is mandatory but inseparable from the informal system,the informal system,although insufficient to make up, but lack of documented requirements,can not constrain coal mine safety and production behavior. To promote coal mine safety production management,it is inseparable from the construction of the two systems, and the two systems can only effective unity of purpose. Besides system is not static, only with the times can promote sustainable development of coal mine safety.The main innovation is reflected in the thesis:(1) Propose for the coal mine accident causation mechanism. This thesis based on the drawing on accident causation theory,categorizes the reason formation of coal mine accidents into human factors,equipment factors,environmental factors,management factors and information factors of five aspects, and compare and analysis the role of risk factors to explore the interaction between various factors and accident evolution path proposed mechanism of cola mine accident causation logical framework, and as the theoretical framework of evaluation index system of coal mine safety.(2) Establish a "man-machine-environment-management-information" evaluation index model. From the existing research,the majority of scholars,according to accident causation theory "man-machine-environment" index evaluation system, the lack of security information management system considerations. From our present situation mine information management, strengthen information security management is particularly important and urgent. Therefore, this thesis combines people,machines,environment,management, information and other elements, taking into account the personnel, equipment, environment, management, and information problems and defects that may cause results to establish a more comprehensive assessment index system model, with some the value of innovation. Optimize the coal mine safety evaluation index system by using the information entropy method. The reasonable of indicator system will directly affect the accuracy of the evaluation results,through the comparative study of indicators of screening methods, this thesis choose a more objective information entropy method to optimize the index system.Combine with fuzzy theory and neural network model to construct fuzzy neural network model. The technology combine of fuzzy theory and neural network, the use of their respective advantages of the two theories, while avoiding their defects, presents a new way of thinking.For mine production safety evaluation of complex systems(3) Construct network prediction model of coal mine safety based on genetic algorithm optimization neural. From the existing ways of coal mine safety warning, fuzzy comprehensive evaluation, AHP, gray relational analysis methods are concerned approach. However, these methods are often subjective process, and directly led to increased uncertainty assessment results as well as an increase in ambiguity. This neural network model applied to the construction of coal mine safety warning model, taking into account the advantages of genetic algorithm, so neural network model has a strong learning ability and fast convergence, has a certain innovation. |
PDF Full Text Request