Font Size: a A A

Research Of The Electro-Hydraulic Actuator Load Control Methods Based On BP Neural Network

Posted on:2008-10-26Degree:MasterType:Thesis
Country:ChinaCandidate:H G QiuFull Text:PDF
GTID:2132360212978959Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
In this paper, a control method aiming at electric actuator hydraulic loading system is discussed and designed.Firstly a mathematical model of the loading system is established. Based on single channel loading system's mathematical model, the system's dynamic characteristic was analyzed. The key problem is how to eliminate strong position disturbance force, which is an important criterion to evaluate loading system's quality. In this way, how to design a proper control strategy to reduce strong position disturbance force becomes a key factor.Secondly, this paper elaborates the cause of the strong position disturbance force, analyzes the factors which influence it, and adopts Structure Invariability principle as feedforward compensation to the disturbance to diminished strong position disturbance, and digital simulation based on Matlab is introduced to simulate. Simulation results show that the control strategy was feasible.Theoretically feedforward compensation can eliminate the position disturbance, but the high-order differential of the compensator is hard to realize physically. Besides, non-linearity and parameters uncertainty of electro-hydraulic servo system reduce the real practical effect of that control method.Thirdly, aiming at the disadvantages of traditional control method, an optimal control strategy based on BP neutral network was proposed after the combination of Classic Control Theory and BP neutral network, and the close-loop system is proved to be gradually stable. The results of simulation in which control strategy was applied to the electro-hydraulic servo system indicate that the control strategy is effective. The control system's dynamic performance has been improved markedly compared with the pure classical control.At the end of this paper, the test methods, test results and analysis results of the system performance test are provided. The test results show that the system's performance meets most requirements in the design index.
Keywords/Search Tags:Electro-hydraulic servo loading system, strong position disturbance, feedforward compensation, BP neural network, optimal control
PDF Full Text Request
Related items