Abstract:Aiming at the proportional solenoid valves widely used in industry, the system stability and control quality of the negative impact, which caused by hysteresis curve that existed in the valves, were studied. By using of mechanism modeling, curve parameter identification based on Particle Swarm Optimization (PSO) algorithm modeling and neural network curve fitting of three methods, modeling was done for hysteresis curve, and three modeling results were gotten by using Matlab software. The modeling results obtained by the three methods were analyzed and compared with the practical measured data. The results show that to determine curve parameter identification based on PSO algorithm modeling do best to represent the actual hysteresis curve. This method is simple and practical. The model can be used for the feed forward control of the proportional solenoid valve, for the purpose of decrease or eliminate flow hysteresis of the valve.