Research Of Predictive Control Based On Mpso-rbf In Gas Supervision System

Gas outburst is considered as the main restrictive factor in coal mine production safety. In viewing of the fact that accident frequently happens in heading face, which could cause damage economic losses and casualties, the supervision system is served as an important measurement in guaranteeing the safety of mine production. Local fans and moving pump underground are the primary monitoring equipment in supervision system. However, the existing control technology is poor and hard to meet the requirements of heading face continuous ventilation and gas safety emission. Therefore, control technology studying has important practical significance.With consideration of the characteristics of the gas monitoring system and the basic framework based on nonlinear model predictive control, a new architecture and algorithm of the coal mine gas monitoring system is proposed in this paper. Prediction model based RBF and particle swarm optimization are adopted firstly in this theory. In order to demonstrate the generalization ability of RBF network model and the speed and capability optimization searching for particle swarm optimizing, a feasibility analysis is conducted and emulation test is processed. Thus, the verified theory is applied to coal mine gas monitoring system. Simulation results show that this method provides a new effective way to reduce the risk of gas outburst and puts an end to the traditional control of gas emissions from the previous way of reducing the accident rate of gas heading face. This method has a high practical value and broad application prospects in the coal mine gas monitoring system.