Search Results

A flat neural network is designed for the on-line state prediction of engine. To reduce the computational cost of weight matrix, a fast recursive algorithm is derived according to the pseudoinverse formula of a partition matrix. Furthermore, the forgetting factor approach is introduced to improve predictive accuracy and robustness of the model. The experiment results indicate that the improved neural network is of good accuracy and strong robustness in prediction, and can apply for the on-line prediction of nonlinear multi input multi output systems like vehicle engines.

For a semi-active suspension design, an important subject is to determine the control law which can achieve good performance both in ride and handling performance. Because of its superiority in non-linear control systems and capability of learning on-line, the fuzzy neural networks (FNNs) control scheme is proposed in this paper for a semi-active suspension system with dynamic absorber. The quarter vehicle model is described by a nonlinear system with three DOF subject to irregular excitation from a road surface and FNNs control scheme is employed. The on-line learning of FNNs to optimize fuzzy inference system is presented. Four kinds of methods, including passive suspension respectively with and without dynamic absorber, semi-active suspension respectively using fuzzy control and FNNs control, are investigated by computer simulation and comparison is made.