基于灰关联分析的GA-BP神经网络在汽车油耗估算中的应用  

Vehicle Fuel Consumption Forecast Models of Optimized Bp Neural Network Based on Grey Relation and Genetic Algorithm

作　　者：程晓娟[1] 韩庆兰[2] 全春光[3] 

机构地区：[1]湖南商学院会计学院,湖南长沙410205 [2]中南大学商学院,湖南长沙410083 [3]长沙学院经济管理系,湖南长沙410022

出　　处：《数学的实践与认识》2016年第8期43-51,共9页Mathematics in Practice and Theory

基　　金：国家自然科学基金(71172101);湖南省教育厅重点资助项目(14A017);湖南省社科基金(湘哲社领[2011]12);长沙市科技计划项目(k1407037-41)

摘　　要：从设计参数特征入手分析影响汽车油耗的因素,利用灰关联分析方法,解析了各设计参数对汽车油耗的影响程度,选择其中灰关联度较大的设计参数作为输入数据,综合工况油耗作为输出数据,构建6-5-1层结构的BP神经网络预测模型,并利用遗传算法获得优化后的BP神经网络的权值和阈值,然后训练BP神经网络得到最优值,最后以国内市场340款汽车作为研究样本,进行有效性验证.研究结果表明,模型利用灰关联分析获得影响汽车油耗的主要因素,简化了网络结构;与优化前的BP神经网络相比,具有更高的预测精度和可靠性.Based on the design parameter characteristics,the factors influencing the fuel consumption of automobile are analyzed.By using the method of grey relation analysis,the impact of various design parameters on vehicle fuel consumption is presented.Then choosing those design parameters with lager grey correlation degree as input data and fuel consumption of comprehensive operating condition as the output data,the prediction model of BP neural network with 6-5-1 framework are established.The modified genetic algorithm is used to optimize the weights and thresholds of BP neural network,and then BP neural network is trained to search for the optimal solution.The availability of this prediction method is proved by the empirical study with 340 cars of the domestic market as the research sample.The research shows that he main factors affecting the vehicle fuel consumption are obtained by the grey relational analysis,and the structure of network is simplified.Compared with the forecasting results of BP neural network,the computer simulations have shown that the precision and reliability of this prediction model are better than those of BP prediction model.

关 键 词：灰关联分析 BP神经网络 遗传算法 汽车油耗 预测 

分 类 号：U461.8[机械工程—车辆工程] U471.23[交通运输工程—载运工具运用工程] TP183[交通运输工程—道路与铁道工程]