车用汽油机过渡工况空燃比的神经网络多步预测控制策略

Air Fuel Ratio Control of Gasoline Engine Based on Neural Network Multi-Step Predictive Model During Transient Conditions

车用汽油机空燃比存在传输延迟,直接用于反馈控制影响空燃比的控制精度.为此,提出了一种基于神经网络的空燃比多步预测控制策略,首先建立了基于BP神经网络的空燃比多步预测模型,利用空燃比预测模型预报空燃比的未来值,利用预测值与期望值的误差及误差变化率,采用模糊控制器对空燃比实施多步预测控制.对HL495发动机两种典型过渡工况实验数据进行仿真,结果表明,该方法能将过渡工况空燃比控制在理论空燃比的±3%以内.

For air fuel ratio signal of a gasoline engine, there exist transmission delay, which affects the control accuracy of air fuel ratio using directive air fuel ratio sensor signals. A multi-step predictive control method based on neural network was provided in this paper. A multi-step predictive model of air fuel ratio based on back propagation neural network was first set up, and a fuzzy controller was then set up using the error of predictive values and expected values and its derivative. The simulation was accomplished using the experimental data of HL495 gasoline engine, and the results show that using multi-step predictive control method, the air fuel ratio error could be contrallod within 3 % under the transient conditions.