The Simulation And Research Based On Neural Network For Controller Of Hydraulic Position Servo System In Precise Straightening Machine

The precise straightening machine is a advanced manufacturing equipment. It is widely used in machinery manufacturing industry such as automobile, construction machinery , military machinery and so on. Hydraulic position servo system of the precise straightening machine is the executive body and is the key unit to decide the processing precision of the workpiece. It is a typical nonlinear time-varying system. The complex system has widespread changes in the parameters,outside interference and so on. Its control requirement is high and difficult. Its performance has a great influence on the performance of precise straightening machine. For this situation, the work of analyse hydraulic position servo system and search a optimization algorithm to achieve a better control effect.In this paper, the transfer function of hydraulic cylinder is deduced .I consider the two load conditions of no-load pressure and the pressure head press down the workpiece. Establish the corresponding mathematical model and analyse the performance in both cases. Aim at the inadequancies of BP algorithms. Two kinds of improved BP algorithm which are the overall learning rate adaptive algorithm by addition of inertia and adaptive learning rate algorithm is elicited. The simulation test behave the overall learning rate adaptive algorithm by addition of inertia is faster. I design the corresponding controller by combine this improved algorithm and PID control. And apply on the particular type precise straightening machine. Try to select the suitable parameters and make simulation by MATLAB simulation software. The simulation results show good control effects such as a strong robustness, a small over and a steady movement. The methods and conclusions of this paper have the theoretical and practical significance for hydraulic servo system of the precise straightening machine.