基于改进FNN的危险化学品运输事故智能预测

An Intelligent Prediction Method of Dangerous Chemicals Transportation Accident Based on Improved Fuzzy Neural Network

为提高危险化学品运输事故预测水平,提出一种改进的模糊神经网络(FNN)模型。实现对危险化学品运输事故起数的智能预测。首先分析危险化学品运输的危险源因素集,确定危险源因素集包含实值型和经验型2类数据。然后设计一种数据融合模型,该模型通过模糊综合评价来精简FNN结构,在此基础上给出改进的危险化学品运输事故的智能预测算法。最后给出改进的危险化学品运输事故的智能预测算法,并以我国2005—2010年期间每个月发生的危险化学品运输事故起数为数据基础进行计算。结果表明,改进模型的预测精度和各种误差均明显好于普通预测模型,预测结果能够反映危险化学品运输事故的实际情况。

To improve the prediction level of dangerous chemicals transportation accidents,an improved FNN model was built.Intelligent prediction of the number of transportation accidents of dangerous chemicals was realized.Firstly,by analyzing hazard factors sets of transportation of dangerous chemicals,it was proposed that the hazard factor set includes two different kinds of variables： accurate numeral variable and fuzzy language variable.Secondly,a kind of data fusion model was designed to simplify the structure of FNN with a fuzzy comprehensive evaluation method.Based on above result,an improved intelligent prediction algorithm was presented to predict the number of transportation accidents of dangerous chemicals.Finally,calculation was made on the basis of dangerous chemicals transportation accident data from 2005 to 2010.The results show that,compared with common prediction model,the improved model has better predicting efficiency and accuracy for both kinds of errors,which can reflect the actual situation of dangerous chemicals transportation accident.