摘要
针对神经网络方法在涡轮增压发动机性能预测方面存在的缺陷,提出了一种新的基于最小二乘支持向量机的涡轮增压发动机性能智能预测方法。介绍了最小二乘支持向量机的基本算法,分析了涡轮增压发动机的性能指标,选择发动机转速、压缩比、容积效率、平均指示压力和平均制动压力作为预测模型的输入参数,输出功率、输出扭矩和有效燃油消耗率作为预测模型的输出量,进一步建立了基于最小二乘支持向量机的涡轮增压发动机性能预测模型。仿真实例的预测结果表明,所建立的智能涡轮增压发动机性能预测模型是合理有效的。
An intelligent prediction method based on least square support vector machine（ LS-SVM） is proposed to overcome the deficiency of the neural network method in performance prediction of turbocharged engine. The basic algorithm of LS-SVM was introduced. The performance indexes of turbocharged engine were analyzed.The engine speed, compression ratio, volumetric efficiency, mean indicated pressure and mean braking pressure were selected as the input parameters of the prediction model. The output power, output torque and effective fuel consumption rate were used as the output of the prediction model. The performance prediction model of turbocharged engine based on LS-SVM was established. The simulation results show that the proposed model is reasonable and effective.
作者
严其艳
YAN Qi-yan(Department of Mechanical and Electrical Engineering, Guangdong University of Science ＆ Technology, Dongguan 52308)
出处
《测控技术》
CSCD
2018年第5期33-36,共4页
Measurement & Control Technology
基金
2015年度广东高校省级重大科研项目(2015KQNCX191)
关键词
涡轮增压
发动机性能
支持向量机
最小二乘支持向量机
turbo charged
engine performance
support vector machine
least square support vector machine
