摘要
为了降低柴油机测控系统的开发时间和费用,采用dSPACE硬件在环仿真平台,对柴油机测控系统进行了开发。采用BP神经网络改进了柴油机准稳态模型,使用Matlab/Simulink软件建立了TBD620V12型增压柴油机的改进准稳态模型,并检验了新增的爆发压力模型和滑油压力模型的精度。在dSPACE处理器中运行该仿真模型,并通过输入输出接口板及接口电路将测控系统与dSPACE平台连接,实现了测控系统与柴油机模型的闭环仿真。仿真结果与试验数据对比表明:相对误差小于3%,从而可用硬件在环仿真代替实机试验,缩短了测控系统的开发周期,并降低了开发成本。
In order to reduce the development cycle and cost, dSPACE hardware-in-the-loop simulation platform was used in developing the monitoring system of diesel engine. BP neural network was used to improve the quasi-stable model of diesel engine. Matlab/Simulink was used to establish an improved quasi-stable model of TBD620V12 turbocharged diesel engine. The precisions of explosion pressure model and lubricating oil pressure model were inspected. The simulation model was operated in dSPACE operator, and the monitoring system was connected with dSPACE platform through I/O boards and interface circuit. The closed loop simulation of the monitoring system and turbocharged diesel engine model was carried out. The relative errors between simulation results and experimental data were less than 3 %, so experimental verification can be substituted by the hardware-in-the-loop simulation, and the development cost and cycle of the monitoring system can be reduced. 6 figs, 8 refs.
出处
《交通运输工程学报》
EI
CSCD
北大核心
2009年第3期66-70,共5页
Journal of Traffic and Transportation Engineering
基金
国防预研基金项目(40101040103)
关键词
柴油机
测控系统
准稳态模型
硬件在环仿真
BP神经网络
diesel engine
monitoring system
quasi-stable model
hardware-in-the-loop simulation
BP neural network
