| Coal is the basic energy source of China,and to protect coal mining is to protect national energy security.With the increase in the depth of coal mining,various natural disasters in the mining operation area have become more serious,and the level of intelligence and robotization of mining equipment has put forward higher requirements.In this context,six national ministries and commissions in 2020 to study and formulate the "guidance on accelerating the development of intelligent coal mining",requiring the strengthening of the basic theory of intelligent coal mining and research and development of key basic technologies to ensure that coal resources to achieve "safe,efficient,green,intelligent" mining.At present,the focus of coal intelligence has shifted from the comprehensive mining face to the comprehensive excavation face,and the research and development of intelligent digging equipment,especially intelligent digging machines,has become the most urgent demand of coal production enterprises.In order to break through a series of key technical problems in the intelligence of cantilever roadheader and achieve rapid and intelligent underground coal mine excavation,this thesis focuses on the key basic scientific problem of roadheader intelligence,"intelligent cut-off control method of cantilever roadheader based on multi-sensing information".Through the collection and analysis of multi-sensor information in the underground and the use of relevant intelligent control algorithms and methods,the multi-sensor information-based cantilever type roadheader cutter head speed staging control method,cutter arm swing speed control method and cutter head speed and cutter arm swing speed joint control method are proposed,which solves the problems of incomprehensiveness,unreliability and low intelligence under the control of single sensor information.It also set up the intelligent cutting system of the coal gangue of the roadheader,carried out the simulation experiment of the intelligent cutting of the roadheader,verified the feasibility and superiority of the relevant control strategies and control methods,and realized the accurate intelligent cutting of the roadheader adapting to the change of coal and rock hardness.The specific research work and innovations are as follows.(1)Analysis of cut-off conditions on Boom-type roadheader and the collection of multi-sensor information downhole.By analyzing the working conditions in the cutting process of the roadheader and the working principles of the cutting head and cutting arm,the key multi-sensing information affecting the cutting state of the roadheader is determined,mainly including the current sensing information of the cutting motor,the pressure sensing information of the cutting cylinder and the vibration acceleration sensing information of the cutting arm.the relationship between the hardness state of the coal rock and the relevant multi-sensing information in the cutting process of the roadheader is obtained.Experiments were designed to collect multi-sensor information downhole from a roadheader.The data set of multi-sensor information downhole was obtained by adding multiple sensor records to the roadheader body,and the downhole data was analysed to provide a data basis for intelligent and accurate cutting of the roadheader.(2)Multi-sensor information-based staging control method for cutter head speed of roadheader.Based on the downhole measurement data and working condition analysis,a multi-sensor head speed staging control strategy is proposed.based on the cutter head speed staging control strategy,a cutter head speed staging identification controller is designed using the back propagation(BP)neural network method.The BP controller was optimised using the Improved Particle Swarm Algorithm(IPSO)to address the shortcomings of BP neural networks that tend to fall into local minima and slow convergence in practical applications;the performance and output accuracy of the optimised controller were greatly improved,with an average reduction in training time of 44% and an average reduction in the number of iterations of 15%.Based on Matlab software,a simulation model of the cut-off head speed is built.The simulation analysis results prove that the cut-off head speed staging control method has high control accuracy and response speed.At the same time,experiments are designed to verify the effectiveness of the cut-off head speed staging control method.(3)Multi-sensor information-based cantilever roadheader cutter arm swing speed control method research.By analysing the relationship between multiple sensing information and the variation of cut-off load,a strategy and control scheme for determining the variation of cut-off arm swing speed of roadheader is proposed.The control scheme is divided into two modules: cut-off load identification module and cut-off arm swing speed control module.Based on the proposed cutter arm swing speed control strategy and scheme,the cutter load recognition controller is designed using an improved particle swarm algorithm optimised neural network(IPSO-BP).At the same time,a fuzzy PID controller optimised with an improved simulated annealing particle swarm algorithm(ISAPSO)was used to design the swing speed controller.Based on Simulink software,a simulation control system was established to compare and analyse the multi-sensor information-based cutting arm swing speed control method of the roadheader.The simulation results show that the method can achieve intelligent adjustment of the cutter arm swing speed to the change of coal rock hardness under different cutting conditions,and has the highest control accuracy and the fastest response speed compared with the fuzzy PID control method with single sensor information and the traditional PID control method.(4)Research on the joint control method of cutter head speed and cutter arm swing speed of roadheader based on multi-sensor information.To address the gap in the joint control of cutter head speed and cutter arm swing speed of cantilever roadheader,a joint control strategy for cutter head speed and cutter arm swing speed was proposed based on a comprehensive analysis of the working conditions of the joint cutter.A joint cutting control strategy and scheme is proposed.Based on the proposed joint cut-off control strategy,an LSTM deep-learning neural network controller is designed to identify the load level of underground coal and rock,and output the load classification signal and its change rate.At the same time,a fuzzy inference controller is designed to output the cutter head speed control signal and the cutter arm swing speed control signal by designing suitable fuzzy rules to achieve the joint intelligent control of the cutter head speed and cutter arm swing speed of the roadheader.Simulation analysis was carried out based on Simulink software to establish a simulation control system.The simulation results show that the joint control method can achieve the joint intelligent regulation of the cutting head speed and cutting arm swing speed under conventional and complex working conditions,and the whole control process has fast response time,small overshoot and high robustness.(5)Intelligent gangue cutting control system for cantilevered roadheader construction and experimental verification.In order to further verify and optimise the intelligent cutting control method for roadheader,the intelligent cutting system for cantilever type roadheader was designed and built by comprehensively analysing the intelligent cutting problems of roadheader under various working conditions;the principle of the system was analysed in detail,the composition of the hardware control platform of the system was introduced,and the adaptive cutting control strategy of the cutting arm swing speed,the hard mass point identification control strategy and the bogging prevention control strategy were proposed for general and special working conditions respectively.Using the EBZ135 roadheader as an example,ground validation experiments were conducted to verify the reliability and feasibility of the proposed control system and strategy.The research in this thesis provides the theoretical and technical support for the realisation of fast and intelligent digging by roadheading robots and provides a theoretical and practical basis for further optimisation and engineering applications thereafter. |
PDF Full Text Request