Decision Model for Workplace Relocation Based on Behavior Analysis

YAO Yupeng; XIONG Wu

doi:10.16545/j.cnki.cmet.2025.01.001

Abstract：A behavior analysis based working face relocation decision model was proposed to address the issues of the inability of preset relocation parameters for hydraulic support controllers to adapt to diverse production environments and high rack loss rates.By analyzing the mechanism of hydraulic support frame shifting,the influencing factors related to the frame shifting operation were obtained,and the parameterized influencing factors were used as inputs for the frame shifting decision model,and the parameters of the frame shifting combined action model were used as outputs.A frame shifting decision model based on radial basis function(RBF)neural network was constructed to address the problem of model overfitting caused by multiple nonlinear inputs and output parameters exceeding the input parameters.The model can achieve fast fitting of multiple nonlinear inputs.The model was trained using artificial relocation data from Inner Mongolia,Shanxi,Xinjiang,and other regions,achieving parameter prediction for relocation under various working conditions.The experimental results showed that compared with the preset parameter method,the model had higher prediction accuracy and could effectively simulate the decision-making behavior of operators in various working conditions.Through on-site deployment testing at the 81202 working face of a certain mine,the relocation parameters generated by the relocation decision model have significantly reduced the rate of machine relocation and frame loss compared to the preset parameter method,which has practical significance for promoting the realization of coal mine production automation.