Variable Sampling Period Time Delay Compensation Strategy Based on RBF Neural Network

According to time-varying and uncertain time delays, a new variable sampling period approach is presented to mitigate the effect of time delay in this paper. Firstly, a RBF neural network with the best approximation and the global optimal performance is adopted to predict the time delay. Secondly, the time delay occurred at current sampling step is taken as the sampling period to establish the networked control system model. Then, a method in combination with optimal control and classical pole placement is used to reduce the amount of calculation and improve the system's precision and real-time performance. Finally, simulation results show that the method has a good effect on time delay compensation.