Multi-parameter NCS Scheduling Based on Fuzzy Neural Network

By studying the NCS(networked control system) data transmission of multiple control loops, an algorithm based on multi-parameter fuzzy neural network (MP-FNN) is proposed in this paper. Firstly, the fuzzy rules given by expert experience, through the ability of self-learning, trained the fuzzy neural network and memorized the fuzzy control rules, and obtained the network demand of each control loop. Then, using the idea of real-time task scheduling, according to the idle time and the criticality of each control task, the urgency parameters of the loop control task are obtained from the defined fuzzy control rules. Finally, combining the urgency parameter and the network demand parameter of the control task, the dynamic weight algorithm is used to obtain the priority of each loop and adjust it online before the next sampling moment. This article uses Matlab's truetime toolbox for the establishment and simulation of NCS models, and anfis toolbox for fuzzy neural network training. Compared with several classic scheduling algorithms, it shows that the algorithm has a better improvement on the loop dynamic control performance.