Based On Improved Ant Colony Algorithm And RBF Neural Network Risk Prediction Of Coal Mine Safety Research

With the rapid development of China’s economy and increasing demand for energy,Coal resources become an important strategic resources in the economic and socialdevelopment of our country. Our country is not only a major coal producer but also amain coal consumer, the security situation in coal industry become very serious.Although the total coal mine accidents have a downward trend in recent years, butmajor workplace accidents still happen, Coal mine safety in poor condition, coaldevelopment is faced with both opportunities and challenges. Although the problem ofcoal mine safety has long cause the attention of the scholars, but for the research startslate, weak foundation for the forecast of coal mine safety risk, this thesis under is thebackground of this study.This paper regards coal mine safety risk as the research object, and on the basis ofthe accident cause theory and man-machine-ring system analysis theory, establishingthe index system of coal mine safety risk prediction. Use the improved ant colonyalgorithm to optimize for RBF neural network parameters. In order to overcome thetraditional RBF neural network slow convergence speed, fell into local minima easilyand the disadvantage of the low precision. The RBF neural network generalizationability is improved, and make it has the characteristics of good output stability and fastconvergence speed. On this basis, setting up the prediction model based on improvedant colony-RBF neural network and using MATLAB software computing tools forprecise calculations, the prediction result of coal mine safety risk more accurate, andproviding more reliable decision basis for decision makers. Paper finally use theempirical analysis to verify the effectiveness of the proposed prediction model.