基于SSA算法优化SVM的发动机润滑油信息状态评估  

Condition Evaluation of Engine Oil Information Based on Support Vector Machine Optimized by Sparrow Search Algorithm

作　　者：李英顺 张国莹 张杨 贺喆 周通 左洋 LI Yingshun;ZHANG Guoying;ZHANG Yang;HE Zhe;ZHOU Tong;ZUO Yang(School of Information Engineering,Beijing Institute of Petrochemical Technology,Beijing 102617,China;Shenyang Shunyi Technology Company Limited,Shenyang Liaoning 110000,China;The Third Military Representative Office of Shenyang Regional Military Representative Bureau of Army Equipment Department in Shenyang,Shenyang Liaoning 110000,China)

机构地区：[1]北京石油化工学院信息工程学院,北京102617 [2]沈阳顺义科技有限公司,辽宁沈阳110000 [3]陆军装备部驻沈阳地区军事代表局驻沈阳第三军事代表室,辽宁沈阳110000

出　　处：《润滑与密封》2023年第2期129-134,共6页Lubrication Engineering

基　　金：辽宁省“兴辽英才计划”项目(XLYC1903015)。

摘　　要：润滑油信息能够有效反映装甲车辆发动机的健康状态,对车辆发动机状态评估十分重要。以某型装甲车辆发动机为研究对象,提出一种基于麻雀搜索算法(Sparrow Search Algorithm,SSA)优化支持向量机(Support Vector Machine,SVM)的发动机状态评估算法。该算法首先对润滑油原始数据进行去噪及归一化处理,然后使用麻雀搜索算法优化支持向量机的核参数与惩罚参数,最后利用寻优后的参数建立评估模型。实验结果表明,采用麻雀搜索算法优化的支持向量机分类准确率高达96.67%,能够有效对发动机状态进行评估,为装甲车辆发动机的换油以及维修提供依据。Lubricating oil information can effectively reflect the health status of armored vehicle engines,which is very important for vehicle engine status evaluation.Taking an armored vehicle engine as the research object,an engine condition evaluation algorithm based on sparrow search algorithm(SSA)optimization support vector machine(SVM)was proposed.In this algorithm,the original data of lubricating oil were denoised and normalized firstly,then the sparrow search algorithm was adopted to optimize the kernel parameters and penalty parameters of SVM,and finally the optimized parameters were used to establish the evaluation model.The experimental results show that the classification accuracy of SVM optimized by SSA is 96.67%,which can effectively evaluate the engine state,and provide a basis for oil change and maintenance of armored vehicle engine.

关 键 词：状态评估 润滑油信息 麻雀搜索算法 支持向量机 

分 类 号：TE626.3[石油与天然气工程—油气加工工程]