一种多参数模糊神经网络控制系统调度算法

摘要

In view of the data transmission of multi-loop network control system and based on both fuzzy con-trol and neural network,a fuzzy neural network scheduling algorithm was proposed;at first,having the self-learning ability of the fuzzy neural network employed to train both error and error rate of the control loop,to re-cord the fuzzy control’s rules as well as to determine the network requirements of each control loop;secondly, combining with the network demand parameters and the network emergency degree parameters,having the dy-namic weight algorithm used to adjust the priority of control loop online;finally,having the simulation model of multiloop control system established for the dynamic weight which determines the scheduling priority.The simulation results show that this scheduling algorithm has good dynamic performance and it can ensure stability of the system at the same network bandwidth utilization rate.%针对多回路网络控制系统的数据传输问题，提出了一种多参数模糊神经网络调度算法。首先，运用模糊神经网络的自学习能力对回路误差及误差变化率参数进行训练，并记忆模糊控制规则，确定各控制回路的网络需求度。其次，结合网络需求度参数和网络紧急度参数，采用动态权重算法确定各回路优先级，实现各回路优先级的在线调整。最后，建立了模糊神经网络调度优先级动态权重可调的多回路控制系统的仿真模型。仿真结果表明：在相同的网络带宽占用率下，该调度算法具有良好的动态控制性能，并保证了系统的稳定性。