Research And Application Of Improved Fuzzy Neural Network PID Algorithm

Traditional PID algorithm is widely used in industrial control because of its simple structure and clear principle.With the development of modern industry,the traditional PID algorithm can not guarantee the accuracy and stability of the control when dealing with the nonlinear system.The fuzzy neural network has the ability to deal with nonlinear characteristics,and it can carry out fuzzy control reasoning and artificial neural network self-learning.Therefore,the combination of fuzzy neural network algorithm and traditional PID algorithm can solve the control problem of nonlinear system very well.Because the initial weights and thresholds of fuzzy neural network algorithm are obtained randomly,and there are too many parameters to be optimized,which leads to the problems of poor real-time control and the control accuracy of fuzzy neural network PID algorithm in the control process.Based on this,in this paper,particle swarm optimization(PSO)is proposed to optimize the initial weights and thresholds of fuzzy neural network.Aiming at the problems of poor global optimization ability and slow convergence speed existing in the conventional particle swarm optimization algorithm,this paper proposes to improve the conventional particle swarm optimization algorithm by activating the stagnant particles and searching in the local area of the global optimal particles.The global optimization ability of the improved algorithm will be significantly enhanced,but the convergence speed of the improved algorithm is still slow.In order to further improve the global optimization ability and convergence speed of the algorithm,this paper proposes the artificial bee colony improved by beetle antennae search.Artificial bee colony improved by beetle antennae search algorithm is based on the traditional artificial bee colony algorithm and the beetle antennae search strategy.By simulating the foraging and searching of the antenna,the left and right antennae are compared with the odor concentration to determine the direction of advance.The introduction of beetle atennae sarch algorithm can make bees get a clear search direction when they are searching for food,enhance the bee's search ability for food sources,enhance the global optimization ability of the algorithm and accelerate the convergence speed.The above two intelligent optimization algorithms are used to improve the fuzzy neural network PID algorithm,and the industrial flash tank is taken as the controlled object.Two improved fuzzy neural network PID algorithm and conventional fuzzy neural network PID algorithm are used to carry out pressure simulation experiments on the flash tank system.The results show that the adjustment time of the fuzzy neural network PID algorithm optimized by artificial bee colony improved by beetle antennae search is further shortened and the overshoot is lower,which makes the pressure system of the flash tank have better control performance.With the help of SMPT-1000 advanced process control experimental equipment and SIMATIC PCS7 of Siemens,the pressure control experiment of flash tank is designed,and the control effect of fuzzy neural network PID algorithm optimized by artificial bee colony improved by beetle antennae search is verified.