中文摘要：

提出了基于混沌径向基（RBF）神经网络的汽油机瞬态工况油膜参数辨识方法。利用混沌优化算法确定隐含层高斯函数径向基中心和输出层连接权值，使其达到全局最优，有效地提高RBF神经网络的收敛速度；同时，利用混沌算法训练RBF神经网络，使目标函数取全局最小值或逼近全局最小值，有效地提高辨识模型的辨识精度，并与BP神经网络模型及最小二乘法辨识进行了分析和比较。仿真结果表明：混沌RBF神经网络模型收敛速度快，具有更强的非线性辨识能力，能够有效地提高油膜动态参数的辨识精度，进而得出不同工况下的油膜参数动态特征。

英文摘要：

A transient condition oil film parameter distinguish method for gasoline engine based on Chaos-RBF was put forward. The Chaos algorithm was used to determine and optimaze the implied Gaussian radial basis function center and the out put layer connection weights for acceleration of the RBF neural network convergence rate. While taking advantage of Chaos-RBF neural network training algorithm, the objective function took a global minimum or close to the global minimum value, effectively improving the model identification accuracy. The recognition ability was analyzed and compared with BP neural network model and least square method. It is showed the Chaos-RBF neural network model has stronger nonlinear identification capability, this model can improve identification accuracy of dynamic oil film parameter effectively and achieve further dynamic characteristics of the oil film parameter in different conditions.