The Application And Research Of Fuzzy Neural Network PID Control In The Switch Of Pinch Roll System

Hydraulic control system often contains both position control and the forcecontrol, if they switch directly, these will produce an instantaneous fluctuation, whichwill also affect the precision and stability of the system.In the production of strip steel, the role of the pinch roll is to bend the head ofstrip steel, guide the strip into the winder, and then set up tension between the drum.The pinch roll is drove by hydraulic system which belongs to the electro-hydraulicservo control. This system contains position control, force control and the switchbetween them. Its performance affects the situation of the production line and thequality of product directly. The purpose of this paper is to design a controller whichcan make the pinch roll operate stably, accurately and fast during the position controland force control system, and then realize the no jitter switch.The real hydraulic servo control system of rolling mill has many complexworking conditions.There are many uncertainties interferences of external load, and the parametersof the controlled object often change a lot with time. What’s more, the hydraulicsystem itself has the dead zone, hysteresis, obviously nonlinear, uncertainty, etc. Soit’s too hard to establish a mathematic model of the system precisely, which make theclassic PID control difficult to meet the requirements of some system. This papercombines the neural network which has the ability of self-learning, massively parallelprocessing with the fuzzy control which is easy to understand and robust, using thefuzzy neural network PID control method to adjust the PID parameters dynamicallyin real-time base on the actual situation.This paper analyzes the pinch roll and establishes the dynamic model ofelectro-hydraulic servo system in the second chapter. The third chapter introduces theprinciple and design process of fuzzy neural network PID controller in detail, buildsthe system model in simulink, and then makes a contrast of simulation under differentconditions. The simulation results show that the new controller has a good improve inthe system performance. The fourth chapter is the experiments, which is to simulatethe working situation of pinch roll, and verify the performance of all controllers inactual operation. At last, the experiments get the same conclusion with thesimulations.Though the improved PID control, it realizes the smooth switch betweenposition system and force system, and meets the requirements of stable, accurate andfast. It verifies the correctness of the theory through a lot of experiments.