Research Of Fire Prediction And Control System Based On AFSA-FCM Clustering Algorithm

Mine fire is a huge obstacle to progress,Research of Technology for Mine Fire Prediction is necessary.The traditional fire prediction by predict in single parameter can not solve the problem from more and more complex condition.This paper analysis influence from many kinds of factor.Then the article analyze the characteristics of serveral burning stages,and come up with the characteristics of mine fire prediction.This paper analysis control technology used in Mine Fire Prediction,auto control damper can prevent a large number of toxic and harmful gases to other workplace invasion and spread in catastrophic period.First,the paper calculate mine fire sample data by fuzzy c-means clustering,and compare the result with result in sample,get correct rate of fuzzy c-means clustering,and analysis the advantages and disadvantages of the fuzzy c-means clustering using in mine fire predictionBased on the shortcomings of the fuzzy c-means algorithm,this paper use artificial fish school algorithm to optimize it,then predict fire with the fire samples and compare the result with results from bp neural network classification algorithm and fuzzy c-means clustering classification algorithm,get a better correct rate.In the case of a small amount samples,AFSA-FCM has a better correct rate than bp neural network...