Research On Key Technologies Of Centralized Control In Fully Mechanized Mining Face

Introduced the research background of my country's urgent need for intelligent coal mining in the high-quality development stage,explored the research significance of the safety and high efficiency of the centralized control system of the fully mechanized working face,analyzed its working principle,investigated and analyzed the current situation of it at home and abroad,put forward the main research contents of the design of the centralized control system of the fully mechanized working face,the positioning of the shearer,the prediction of the shearer cutting trajectory,and the adaptive operation of the shearer cutting drum.Based on actual working conditions,the function design of the control system in the comprehensive collection was carried out,and the system architecture design was carried out to achieve the goal of parallel real-time operation for the ground and underground.The control system of the coal mining machine,the control system of the hydraulic support and the key system of the conveyor control were carried out.Through the integration of system software and hardware,the control function of comprehensive collection was realized,which provided a basic platform for intelligent operation.In order to strongly support the intelligent operation of fully mechanized mining equipment,a precise positioning scheme based on multi-sensor fusion was proposed.Using the layout characteristics of fully mechanized mining face,the infrared sensor was used to encode the discrete position of the hydraulic support to provide rough positioning for the shearer.Rotary encoder and inertial navigation were used to measure the running speed and attitude of the shearer respectively.Through the Kalman filter algorithm,combined with the shearer motion equation,the coal position was estimated to realize the precise positioning of the shearer.In order to accurately predict the changing law of the cutting trajectory of the shearer,a deep learning method was adopted to collect the standardized operation data of the excellent shearer driver.After the data was cleaned and processed,it was input into the deep neural network model based on LSTM and Res Net for training.Extracted the characteristics of time dimension and space location dimension,predicted the cutting trajectory for a period of time in the future based on historical trajectory information,and guided the standardized mining of coal seams.Based on the prediction results of the cutting trajectory,the self-adaptive operation technology research of the shearer drum was carried out,and the drum action was decomposed.For the drum rotation action,the load-sensitive idea was adopted to realize the self-adaptive speed regulation control of the drum.For the drum height adjustment action,realized the tracking control of the roller trajectory based on the LQR algorithm.From an end-to-end perspective,an adaptive operation control method for the cutting drum was proposed based on deep learning,which provided a new way for the research on the adaptive operation of the drum.According to the shearer fusion positioning,cutting trajectory prediction,and drum adaptive operation technology methods,experimental research was carried out,an experimental system platform was built,and various technical methods were tested and verified.The results showed that the shearer positioning position is accurate,and The overall prediction trend of the cutting trajectory is correct,and the trajectory tracking control is accurate.The research method is practical and effective,and provided a reference for the subsequent in-depth study of the intelligent operation of fully mechanized mining.There are 46 figures,5 tables and 83 references in this thesis.