关键词: 灰色神经网络, 混合粒子群优化算法, 瓦斯分布区域, 煤矿机器人

Abstract: To get more accurate prediction of the gas distribution area and to provide basis for the coal mine robot to avoid the dangerous area of gas, this paper presents a new method for prediction of the gas distribution ahead 10m of coal mine robot by using HPSO-GNN. Experimental results show that the average relative error of gas concentration predicted by HPSO-GNN is 4.83%, and the overall prediction accuracy of the HPSO-GNN prediction method is improved by 57.48% compared to the GNN prediction method, the predicted results of gas distribution area are in good agreement with the measured values. The method can accurately predict the gas concentration distribution area, and provides a basis for the coal mine robot to avoid the dangerous area of gas.