基于GA-SVM的舰船装备临修经费需求预测模型  

A Demand Forecasting Model for Ship Equipment Temporary Repair Funds Based on The GA-SVM

作　　者：黄栋[1] 费良 

机构地区：[1]海军工程大学装备经济管理系,湖北武汉430033

出　　处：《数学的实践与认识》2017年第13期91-97,共7页Mathematics in Practice and Theory

基　　金：国防科研重点研究项目资助

摘　　要：针对舰船装备临修经费需求预测得不到满意解的问题,运用遗传算法将SVM相应的参数进行优化,建立了基于GA-SVM的舰船装备临修经费预测模型.通过将GA-SVM模型与BP神经网络模型的预测结果进行对比分析,结果表明:GASVM的预测效果更优异,对舰船装备临修经费需求预测有更好的参考意义.Ship equipment temporary repair funds demand forecasting is not satisfied with the solution of the problem, using genetic algorithm to the corresponding SVM parameters optimization, based on GA is established - the SVM prediction model of ship equipment temporary repair funds. By using GA - SVM model and BP neural network model of prediction accuracy were analyzed, the results show that the GA - SVM prediction effect is more outstanding, temporary repair equipment of ship funds demand forecasting has a better reference.

关 键 词：临修 GA—SVM 经费预测 

分 类 号：O212.1[理学—概率论与数理统计] TP18[理学—数学]