Research Towards Adjustment Control Algorithm Of Shearer’s Memorial Cutting On The Basis Of Extreme Learning Machine

The key technology of the less-manned and unmanned working face of the shearers is to realize the automatic lifting of shearers. The automatic lifting control of shearers has important practical significance on improving the reliability of shearer system, extending the equipment life and enhancing the production efficiency as well as ensuring the production safety. This paper focuses on the research towards adjustment control algorithm of shearer’s memorial cutting on the basis of extreme learning machine, the main work is as follows:Through the analysis of the structure and working principle of shearers, the paper combines with the research about shearer memorial cutting system to further ensure the collection and sampling period of memorial cutting parameters information. It also considers the existing mining technique to set up the artificial demonstration cutter-path planning model and analyzes the advantages and disadvantages of particle swarm optimization and genetic algorithm, using the mixture optimization algorithm which combines the two algorithms to plan the demonstration cutter path as well as analyzing the error of the planning path. The results show that the mixture optimization algorithm can meet the job requirements of shearers. The paper deeply studies the extreme learning machine belonging to the traditional neural networks algorithms which is a new type of single hidden layer feedforward neural networks algorithm, showing that in computing speed, computing precision and generalization ability the extreme learning machine has great advantages. Based on the analysis of the shearer hydraulic servo lifting system, aiming at the nonlinear and delayed characteristics of the shearer lifting system, the paper further designs the PID control strategy based on the extreme learning machine and utilizes the Matlab/Simulink software to trace and control the cutting trajectory and provides the error curve.By analyzing the simulation results and tracking precision, the paper indicates that the PID control of extreme learning machine can be applied to the nonlinear, delayed shearerlifting system and can effectively improve the accuracy of trajectory tracking, which has a far-reaching influence on realizing the unmanned working face of coal mining.