Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory > Artificial Neural Networks and Computing

The Research of Comprehensive Evaluation Model of Coal Mine Safety Based on Neural Network

Author ChenWei
Tutor LiChuanGui;ChaiJianShe
School Capital University of Economics
Course Safety Technology and Engineering
Keywords Coal Mine Safety Influencing factors Index system Neural Networks BP algorithm MATLAB neural network toolbox Evaluation Model
CLC TP183
Type Master's thesis
Year 2010
Downloads 273
Quotes 5
Download Dissertation

The paper first analyzes the status quo of China 's coal mine safety , and discusses the need to carry out a comprehensive evaluation of coal mine safety , and then on the basis of accident causation theory , combined with the complex and diverse characteristics of the coal mine production system with fault tree analysis method , man - machine - environment the three common methods of analysis and analytic hierarchy process analysis model , build a coal mine safety indicators as tools for analysis of coal mine safety evaluation . Coal mine safety evaluation index system analysis model through the coal mine safety indicators , longitudinally divided into three levels of the target layer, middle layer , and factors layer , horizontal division of the factors is the quality of personnel , production equipment factors , the production of environmental factors and safety management factors The four aspects of a total of 34 indicators . Neural network structure , algorithm , the local minimum problems of BP neural network learning algorithm using additional momentum method improvements , improved BP neural network to establish a comprehensive evaluation model for coal mine safety , determine the model three ( 34-14-5 ) structure using neural network toolbox in MATLAB programming the computer to achieve the establishment of the neural network model for comprehensive evaluation of coal mine safety . 20 instances of domestic representative sample of neural network training using additional momentum method based on adaptive adjustment step method , the problem of slow convergence of the presence of the error function in the training process , improve , and finally through the the evaluation model subordinate of Geng Village, Changcun mine of Yima Coal Industry Group Company , the results are consistent with the actual situation shows that the model successfully established a complex nonlinear relationship between the influencing factors of coal mine safety situation has a certain practical value. With the sample library and the gradual development of the model evaluation results will be more precise, its value will be higher and higher .

Related Dissertations
More Dissertations