基于PSO-SVM的航材消耗预测模型研究
Aerial Material Consumption Prediction Model Based on PSO-SVM
许浩 1田才艳 1毛瑞柯 1辜兴磊 1常川1
作者信息
- 1. 中国民用航空飞行学院,四川 德阳 618307
- 折叠
摘要
航材消耗预测是航材库存精准管理的前提,提高航材消耗量预测精度能显著降低库存管理成本.为解决航材消耗预测中因航材消耗影响因素较多、样本数据量少而造成的预测效果差、精度低等问题,提出一种将粒子群算法及支持向量机相结合的航材消耗预测模型,首先使用粒子群算法寻优支持向量机参数组合,然后结合原始数据优化支持向量机参数组合得到PSO-SVM航材消耗预测模型,结果表明,PSO-SVM模型的预测效果较好,泛化能力较强.
Abstract
Aerial material consumption prediction is a prerequisite for precise management of aerial material inventory,and improving the accuracy of aerial material consumption prediction can significantly reduce inventory management cost.To solve the problems of poor prediction performance and low accuracy caused by multiple influencing factors and small sample data in aerial material consumption prediction,an aerial material consumption prediction model that combines Particle Swarm Optimization and Support Vector Machine is proposed.Firstly,the Particle Swarm Optimization is used to optimize Support Vector Machine parameter combination,and then,combined with the original data,it optimizes Support Vector Machine parameter combination to obtain the PSO-SVM aerial material consumption prediction model.The results indicate that the PSO-SVM model has good predictive performance and strong generalization ability.
关键词
航材消耗/粒子群优化/支持向量机/消耗预测Key words
aerial material consumption/Particle Swarm Optimization/Support Vector Machine/consumption prediction引用本文复制引用
基金项目
中央高校基本科研业务费专项(2023)(QJ-2023-006)
出版年
2024
国家科技期刊平台
NSTL
维普
万方数据