神经网络的电传动装甲车辆发动机建模与仿真

Modeling and Simulation Research on Engine for Electric Drive Vehicle Based on Neural Network

发动机是电传动装甲车辆的核心部件,其动态特性具有很强的非线性,通过采用遗传算法优化的BP神经网络学习发动机台架实验数据,得到发动机稳态模型,在此基础上结合控制理论知识建立了发动机动态模型,通过仿真检测了模型的精度,证明了模型的可行性。

As the core parts of Electic Drive Vehicle,engine＇s dynamic characteristics has high non-liner characteristic. This article researches through using tile BP neural network which is optimized by genetic algorithm to learn a great of experimental data established the steady-stay model of engine, and then combined with control theory established dynamic model. Then tested model errors that informs the feasibility of the model by computer-simulation.