车用汽油机过渡工况进气流量预测研究

提出了一种基于混合遗传算法的径向基神经网络(HGARBF)的车用汽油机过渡工况进气流量预测模型。首先设计了一种新的混合遗传算法,利用梯度算法每次迭代得到的结果来改进遗传算法的群体,将遗传算法的最优个体与梯度算法的迭代解相比较,选择其中的最优点作为梯度算法下一步迭代的起始点,运用该混合遗传算法进行径向基神经网络参数的优化,改善径向基神经网络不同初始参数对其性能的影响;然后建立了基于HGARBF网络的过渡工况进气流量的预测模型。仿真结果表明,该预测模型优于经典的进气流量平均值模型,为精确及时测试汽油机进气流量提供了新的方法。

A Prediction on the Induction Air Flow Rate of Gasoline Engine During Transient Condition

For solving the problem that the severe fluctuation of air induction and the lagged response of air flow sensor seriously affect the control accuracy of air/fuel ratio during transient condition,a prediction model for induction air flow rate of gasoline engine is presented based on RBF network with hybrid genetic algorithm.First,a new hybrid genetic algorithm is designed,and the populations of genetic algorithm are improved by using the result of each iteration in gradient algorithm.Then the better one between the best individual in genetic algorithm and the current iteration result of gradient algorithm is chosen as the starting point of next iteration of gradient algorithm.Next,an optimization is conducted on the parameters of RBF neural network by using hybrid genetic algorithm for improving the performance of RBF neural network.Finally a prediction model for induction air flow rate in transient condition is established based on RBF neural network with hybrid genetic algorithm.The simulation results show that the prediction model is superior to the classical mean value model with better accuracy and timeliness.