Recommendation Study On Pressure Relief Scheme For Coal Mine Drilling Based On BP Neural Network

In the process of coal mining,a variety of disasters can occur,and impact pressure is one of the most typical disasters.Because of its great harmfulness and easy to cause huge losses and casualties after its occurrence,how to prevent and control the impact pressure has always been one of the key issues in the research field of coal mining industry.Among the various measures for the prevention and control of impact pressure,drilling pressure relief is widely used because of its simple construction process,high safety,good cost performance and strong adaptability.At present,the drilling pressure relief scheme is mainly given by experts based on experience,which is more subjective,has certain ambiguity,and takes more time and is relatively inefficient.Therefore,the parameter recommendation of drilling pressure relief scheme using BP neural network has important theoretical and practical value.Based on the principle of pressure relief in coal mine drilling and the current research results,this thesis analyzes the parameters and influencing factors of pressure relief scheme in coal mine borehole,constructs a BP neural network model,and finally verifies the effectiveness of the model by example,as follows:Firstly,combined with the principle of borehole pressure relief,the influencing factors of borehole pressure relief are analyzed,and the input layer and output layer of the model are determined by combining the principle of the current drilling pressure relief scheme.The input layer of the model is mainly composed of six indicators:burial depth,coal thickness,inclination angle,uniaxial compression resistance,impact risk and regional type,covering many influencing factors.The output layer of the model is mainly composed of three indicators: drilling diameter,drilling depth and drilling spacing,and determines the three parameters of the drilling pressure relief scheme.Secondly,according to the basic principle of BP neural network,the specific number of layers and nodes of BP neural network structure are determined,and a recommendation model of drilling and pressure relief scheme based on BP neural network is constructed using python language.The geological conditions and borehole pressure relief scheme of multiple coal mines in Shandong Province,Xinjiang Uygur Autonomous Region and Inner Mongolia Autonomous Region were selected to form 29 training cases,randomly divided into training set and test set,and the constructed BP neural network was trained,and the recommended model of drilling pressure relief scheme based on BP neural network was obtained.Finally,the 1131 working face remining roadway of a coal mine in Xinjiang is selected as a research case to verify the effectiveness of the recommended model of borehole pressure relief scheme.By analyzing the similarity analysis and error calculation between the prediction results and the actual application scheme,the similarity is high and the error is less than 20%,which determines that the model has a good effect,which is conducive to the design of coal mine drilling pressure relief scheme,and then effectively guides the formulation of coal mine drilling pressure relief scheme.The thesis has 15 pictures,27 tables and 81 references.