基于SSA-LSSVM的汽油机点火提前角预测研究  

Study on ignition advance angle prediction of gasoline engine based on SSA-LSSVM

作　　者：谢清华 李岳林[1,2] 张五龙 陈侗 尹钰屹 XIE Qinghua;LI Yuelin;ZHANG Wulong;CHEN Tong;YIN Yuyi(Key Laboratory of Safety Design and Reliability Technology for Engineering Vehicle of Hunan Province,Changsha 410114,Hunan,China;School of Automotive and Mechanical Engineering,Changsha University of Science&Technology,Changsha 410114,Hunan,China)

机构地区：[1]长沙理工大学湖南省工程车辆安全性设计与可靠性技术重点实验室,湖南长沙410114 [2]长沙理工大学汽车与机械工程学院,湖南长沙410114

出　　处：《中国工程机械学报》2024年第5期561-566,共6页Chinese Journal of Construction Machinery

基　　金：国家自然科学基金资助项目(51176014)。

摘　　要：为了提高汽油机点火提前角预测的精确性,从而改善发动机性能,构建了一种基于麻雀搜索算法(SSA)优化最小二乘支持向量机(LSSVM)的点火提前角预测模型。该模型通过SSA对LSSVM的正则化参数c和核函数参数σ进行优化辨识,提高模型自身的鲁棒性与泛化能力,使LSSVM的预测性能和泛化性均都达到最佳。运用汽油机试验数据对模型进行训练和预测,并将仿真结果分别与反向传播(BP)神经网络模型、径向基核函数(RBF)神经网络模型进行横向对比分析。仿真结果表明:SSA-LSSVM模型的预测值与试验值的平均相对误差(MRE)控制在2.5%范围之内,与常规的BP预测模型、RBF预测模型相比,MRE分别下降2.35%和1.56%,表明SSA-LSSVM模型具有更高的预测精度,更加适用于汽油机点火提前角预测。To improve the accuracy of ignition advance angle prediction for gasoline engines and thus improve the engine performance.An ignition advance angle prediction model based on sparrow search algorithm(SSA)optimized least squares support vector machine(LSSVM)is constructed.The model optimizes the identification of the regularization parameter c and the kernel function parameterσof the LSSVM by SSA to improve the robustness and generalization ability of the model itself,so that the prediction performance and generalization performance of the LSSVM are both optimal.The model was trained and predicted using gasoline engine test data,and the simulation results were analyzed in side-by-side comparison with the BP neural network model and RBF neural network model,respectively.The simulation results show that the average relative error between the predicted values and the test values of the SSA-LSSVM model is controlled within 2.5%,which is 2.35%and 1.56%lower than that of the conventional BP prediction model and the RBF prediction model,respectively,indicating that the SSA-LSSVM model has a higher prediction accuracy,and it is more suitable for predicting the ignition advance angle of gasoline engines.

关 键 词：汽油机 点火提前角 麻雀搜索算法 最小二乘支持向量机 优化辨识 

分 类 号：TK417.4[动力工程及工程热物理—动力机械及工程]