Research On Coal Mine Production Safety Situation Evaluation Based On Hidden Danger Data Fusion

Mine s affected by complex production environment,technical equipment,staff,management level and many other factors.The assessment,prediction and early warning of mine safety situation need to consider all kinds of factors comprehensively.The deployment and application of IOT in coal mine research on the coal safety production situation awareness with the sensor information fusion deeply,but the sensor fusion is not enough to extract feature from the sensor data to meet the requirements which coal mine production safety situation awareness need.Therefore,this thesis studies the current coal safety production situation awareness in the following aspects:First of all,this thesis analyzes the coal mine safety situation and related concepts as well as discusses the necessity of introducing coal mine production safety situation is discusses.Based on the difficulty of using data fusion to understand the situation of coal mine production safety,it puts forward the corresponding solution and discusses the feasibility of the solution.And it builds the understanding model of coal mine production safety situation based on data fusion which adapts to the current level of coal mine safety monitoring.Secondly,the coal mine production safety status assessment method based on hidden events is researched from the safety management perspectives of Safety-I and Safety-II.From the point of view of Safety-I,this thesis constructs the evaluation system of coal mine production safety level based on the theory which contains man,machine,environment and management.After then,It used BP neural network to learn expert experience to realize the evaluation of coal mine production safety level.On the other way,from the perspective of Safety-II,the functional resonance theory is taken as the cause mechanism of accidents.Through the division of coal mine production functions,it constructs the system of coal mine production risk threat assessment based on the functional resonance theory,and use the hierarchical evaluation method to evaluate the coal mine production risk threat.The coal mine production safety situation is explained through two kinds of safety management perspective,which makes the evaluation more comprehensive.Furthermore,several coal mine safety status assessment results are taken as the supporting evidence of coal mine production safety situation,and D-S evidence theory is used to reason these evidences.In the process of reasoning,on the basis of generating basic probability assignment with fuzzy attribute model,the membership function of various evaluation results is determined with safety policy,which reduces the difficulty of constructing fuzzy attribute model and makes the evaluation of coal mine production safety situation more accurate.Finally,the model proposed in this thesis is applied in a practical case by taking the 2803 working face of Chensilou as the research object.The application of the case shows that the understanding of coal mine production safety situation based on hidden danger data is feasible and of practical significance.There are 21 figures,19 tables and 94 references in this thesis.