| Although our country has acquired the capability to research and produce jumbo multi-head roof and rib-bolters, it has a far cry from the advanced world level in productivity, safety reliability, and performance.The solutions was put forward that was to accomplish the jumbo bolter's automatic dividing and lock function by means of intelligent braking system on the basis of in-depth insight about national jumbo bolter's characteristics, and braking system's theoretical models was set up based on it. Besides, each subsystem's mathematical model was set up according to the concrete structure of the braking system, and the model was corrected by modern control method.Combining the advantages of fuzzy control and neural network, the neural-fuzzy reasoning method was brought forward, which acted as control algorithm of braking system's controller. The research was carried out on drilling arm braking system's control method in accordance with system's mathematical model, reasonable initial fuzzy system and training sample were established according to the effect of membership function's structural parameter on the reasoning system and the effect of PID three control parameters on the control system. Based on this, high property neural-fuzzy reasoning system and neural-fuzzy PID controller were designed.According to the established mathematical model and control algorithm, drilling arm's braking system was simulated with the help of MATLAB simulation software, braking system's property was analyzed and tested, and compared the result with the result of fuzzy control brake system. The simulation result indicates that the control system has good adaptability, high location accuracy, and has certain anti-jamming ability, thus it can satisfy function index of jumbo bolter's automatic dividing and lock system when it is used under the mining environment, compared with fuzzy PID controller, neural-fuzzy PID controller has obvious superiority not only on control accuracy, but also on control time. |
PDF Full Text Request