基于支持向量机的飞机备件需求预测

Requirement Prediction of Aircraft Spare Parts based on Support Vector Machines

支持向量机是一种机器学习算法,在国外已广泛应用于工程实践领域。首先探讨了支持向量机回归预测模型的学习和预测机制,分析其中三个重要参数对算法的影响规律,得出一套定性的参数选择方法,然后将支持向量机引入到装备综合保障分析之中,构建了飞机备件智能预测模型,并对某型军用飞机备件需求进行了预测和分析,结果表明:基于支持向量机的备件需求预测是有效的、可行的。

Support Vector Machines(SVMs)is an algorithm of machine learning,which was widely used in many fields abroad.Firstly,the algorithms of learning and prediction were discussed in the paper.Secondly,the influence of three important parameters in SVMs was analyzed and a qualitative method was proposed on how to choose the parameters.Then SVMs was used in equipments logistic support analysis,and an aircraft spares prediction model was proposed,by which,we estimated spares provision of an aircraft.Lastly,the simulation results proved the effectiveness of SVMs.