基于灰色关联分析与PLS-LSSVM的航材备件需求预测模型

航材备件是保障航空装备日常训练和作战正常使用的重要影响因素,针对部分航材备件样本数据量少,影响因素多且复杂多变,预测结果与装备系统完好性要求偏差较大等问题.建立基于灰色关联分析(GRA)与偏最小二乘(PLS)及最小二乘向量机(LSSVM)相结合的航材备件预测模型,采集某无人机航材备件数据,通过对统计数据进行灰色关联分析,提取航材备件需求的相关因素作为模型训练样本,确定关键因素,利用偏最小二乘对关键因素特征提取,然后将偏最小二乘特征提取后的数据作为最小二乘向量机输入,进行模型构建及分析.通过实验验证了该方法的可行性与适用性,能够满足无人机航材备件预测的实际需要.

The Prediction Model of Airlines Spare Parts Based on Grey-relational Analysis and PLS-LSSVM

Airlines spare parts are used normally as one of the most important factors which guaranteeing daily training and wartime flight of aeronautic equipment,according to the problems of less sample date,many influencing factors with complex and changeable,and the larger discrepancy of predict results system and materiel integrity requirements,building the prediction model of airlines spare parts based on grey-relational analysis(GRA)and partial least squares(PLS)and least square support vector machine(LSSVM)in view of the statistical dates of airlines spare parts in one drone,through grey-relational analysis of collected data,extracting the related factors as the model training samples of airlines spare parts,determining the key factor sets,making the feature extraction through air-materials demands of the key factor sets using partial least squares,then the dates is known as the least square support vector machine input,the model is built and analyzed.The feasibility and adaptability of this way is verified through experiment,and can meet practical need of air-materials demand prediction.