The Study on Control System of New Parallel Powered Support Based on Neural Network

摘要

The current four-chain-mechanism powered support can’t bear the side force parallel with coal mining face and must bear the horizontal load by the internal stress of the mechanism. But parallel mechanism has advantages of simple construction, high stiffness, good dynamic performance and restructure. The authors take the mobile platform of 6-SPS mechanism as the beam of powered support, the ground platform as the base of support and six legs as powered props, add some assistant parts up, which consist of the parallel powered support. This paper presents the control system of the parallel powered support. The authors apply the Error Back Propagation network controller to control the extension of the powered props. The neural network model is composed of two import node layers, one implicit layer and three export node layers. Adopting the data given by MATLAB program to train the BP network, putting the feedback data from the displacement transducers and the pressure transducers into the network. The results show that the neural network controller can control effectively the movement of the parallel powered support, and supply some theoretical direction for its industry application.