Study on Thrust Hydraulic System of TBM Based on AMESim a MATLAB/Simulink Interface

Abstract

The choice of fuzzy rules and membership function of fuzzy self-adjusting PID controller of tunneling boring machine (TBM)’s thrust hydraulic system exists the defect of subjectivity to a great extent. The genetic algorithm (GA) was adopted to optimize fuzzy rules and parameters of membership function and an improved PID controller based on GA was designed. In order to conduct the co-simulation, the mechanical model was built in AMESim and control model was designed in MATLAB/Simulink and co-simulation interface was used to carry out the co-simulation. The simulation results show that the improved PID controller can eliminate the overshoot and improved the response and steady state performance. Introduction With the rapid development of social productivity, TBM is becoming more and more popular in long linear tunnel construction. As the heart of TBM, the thrust hydraulic system is not only to realize the movement forward of TBM’s thrust cylinder, but also to achieve TBM’s postural adjustment and fallback alone when the segment is installed [1].So far, according to the study on thrust hydraulic system of TBM, predecessors have already done a lot of interesting research work and achieved fruitful results. Yang et al [2] applied pressure and velocity compound control strategy based on fuzzy self-adjusting PID algorithm to accomplish speed-control of TBM’s thrust hydraulic system. Liu et al [3] designed an adaptive PID controller for thrust speed of TBM’s thrust hydraulic system based on BP neural networks. Shi et al [4] designed the PID controller based on the single neural to achieve the adaptive control of shield’s thrust speed. Hou et al [5] used the algorithm of RBF neural network to control the speed of hydraulic thrust system of telescopic shield TBM. This paper applied GA to optimize the parameters of membership function and the rules of fuzzy PID controller. The simulation results show that the improved controller has eliminated the overshoot phenomenon and enhanced the robustness of hydraulic thrust system, which has made TBM’s thrust system a good performance. Introduction of TBM’s Thrust Hydraulic System The principle of TBM’s thrust hydraulic system is shown in Fig.1. When the system is in normal working condition, two four-way electromagnetic reversing valve 1 is closed and hydraulic fluid enters the big cavity of thrust cylinder by the way of proportional flow control valve, the right of three four-way electromagnetic directional valve 1, hydraulic-lock 5 and balanced valve 6 to drive disc cutters to cut soil. When TBM is installing the segments, proportional flow control valve is shorted and the fluid enters the rod cavity by the way of two four-way directional valves 1, the left of three four-way directional valve 4 and balanced valve 6 to accomplish the movement of rapid fallback. In this process, balanced valve 6 can ensure the stability of fallback operation. When the system is holding the pressure, for example, when the operation is to change tools, with hydraulic 4th International Conference on Mechatronics, Materials, Chemistry and Computer Engineering (ICMMCCE 2015) © 2015. The authors Published by Atlantis Press 2809 lock 5 and Y in a function of the three four-way directional valve 4, the system forms a locking-circuit and can avoid the leakage of fluid when the excavation is stop. 1.two four-way electromagnetic reversing valve 2.proportional flow control valve 3.proportional pressure relief valve 4.three four-way electromagnetic directional valve 5.hydraulic-lock 6.balanced valve 7.cylinder Fig.1 The principle of thrust hydraulic system Optimum Design of Fuzzy Self-adjusting PID Controller Based on GA The Design of Fuzzy Self-turning PID Controller. Inputs of fuzzy self-turning PID controller are error E and its change rate EC and p K 、 i K and d K are intermediate variables. Fuzzy outputs are the adjustment values of p K ∆ 、 i K ∆ and d K ∆ .The basic domain of inputs and outputs is [-3 3] and the fuzzy domain is{-3, -2,-1,0,1,2,3}. The adaptive parameters of every control cycle are calculated according to the following formula: