基于支持向量机的机载吊舱故障诊断优化算法

提升机载吊舱的后勤保障能力，适应吊舱测试中多型号、多故障类型和测试环境动态变化的测试要求，是打赢现代化战争的重要保障。支持向量机（SVM）算法适用于小样本、高维度、非线性分类问题，SVM相关参数是影响算法性能的重要因素。基于K-CV算法和粒子群算法两种改进的SVM模型可以实现SVM参数优化，K-CV算法可以交叉验证优化模型参数，粒子群算法可以对SVM参数进行动态寻优，建立多核SVM吊舱故障诊断模型。两种算法都可以提高吊舱故障诊断模型的准确率，提高模型的学习能力和泛化能力，有效对吊舱的故障进行定量和定位诊断。

Research on Airborne Pod Fault Diagnosis Algorithm Based on the Improved SVM

It is an important guarantee for winning the modernization war to upgrade the logistic support capability of airborne pods and to meet the testing requirements of multi-model, multi-fault types and dynamic changes of test environment. Support vector machine (SVM) algorithm is suitable for small samples, high-dimensional, nonlinear classification problems. SVM-related parameters are important factors that affect the performance of the algorithm. The improved SVM algorithm by K-CV and PSO based on the traditional SVM algorithm is used to validate the parameters of the model. The K-CV algorithm is used to cross-validate optimization model parameters .The PSO algorithm is used to dynamically optimize the SVM parameters and a multi-core SVM pod fault diagnosis model is established. Both algorithms can improve the accuracy of the fault diagnosis model, then, improve the learning ability and generalization ability. The optimized SVM fault diagnosis model can effectively quantify and locate the pod fault.