Study On Self-adaptive Cutting Control Of Shearer Based On CPS

The shearer is the core equipment of the fully mechanized mining face,and the self-adaptive cutting control of shearer is the basis for realizing its intelligence,and it is also an important measure to ensure the optimal comprehensive performance of shearer and the safe and reliable operation of the unit.In order to solve the problem of self-adaptive cutting control of shearer under complex coal seam conditions,the coal seam of 4602 working face in Yangcun Coal Mine of Yanzhou Coal Industry Group and the spiral drum of MG2×55/250-BWD shearer were taken as engineering objects.Relying on the Natural Science Foundation Project"Research on the Power Transfer Law and Structural Evolution Theory of High-efficiency Cutting Roller Coal Falling with Gangue Coal and Rock(51674134)",based on the concept of Cyber-Physical Systems(CPS),The virtual prototyping technology was used as a research method to analyze the feasibility of the coal and rock cutting state recognition model system based on Simulink and explore the optimal control strategy of the shearer self-adaptive cutting state,and the coal and rock cutting state recognition model system and the optimal control strategy were considered as the CPS information system,based on the similarity theory,the shearer self-adaptive cutting control comprehensive test bench was built as the CPS physical system.The information system and the physical system are integrated to realize the closed-loop self-adaptive control of the former to the latter.The researches carried out and the results obtained are as follows:The intrinsic parameters of coal and rock were obtained according to physical tests,and by means of Plackett-Burman test,steepest climbing test and Box-Behnken test,based on coal-rock accumulation simulation test and uniaxial compression simulation test,the contact mechanical parameters of coal and rock particles and the uniaxial compression simulation test were realized.For the calibration of bonding mechanical parameters,the calibration errors are all within the range of 5%.By means of virtual prototype technology,the discrete coal wall model of complex coal seam was established based on EDEM,the rigid flexible coupling dynamic simulation model of shearer was established based on Recur Dyn,the two-way coupling system model of cutting coal and rock of EDEM-Recur Dyn rigid flexible coupling shearer is constructed based on(DEM-MFBD)interface,the reliability of the model was verified by comparison with physical tests,and the height adjustment hydraulic system model was established based on AMEsim,and the two-way coupling simulation model was integrated with EDEM-Recur Dyn to simulate the physical system of shearer cutting coal and rock.A multi-objective optimization model was constructed to optimize the comprehensive performance of the shearer,and the traction speed and drum rotation speed of the shearer under typical working conditions were optimized.The vibration acceleration in the direction of the cutting resistance of the drum was used as the identification characteristic signal,and the SVD noise reduction,CWT,Fancy PCA and Alexnet network transfer learning were combined to realize the recognition of coal and rock cutting state,and the recognition accuracy rate can reach95.09%.The DDPG was used to build a shearer self-adaptive control model,and the superiority of the DDPG algorithm was verified from four aspects:system tracking characteristics,anti-interference,environmental adaptability and control performance of different control algorithms.Based on Simulink,coal and rock cutting state identification model,self-adaptive height adjustment model,and shearer traction speed-drum speed(v_q-n)coordinated speed regulation electronic control system was established,and based on interface technology,shearer self-adaptivecuttingmachine-electric-hydraulic-controlmulti-domain(EDEM-Recur Dyn-AMEsim-Simulink)co-simulation model was established,and simulating and analyzing the performance of adjustment sequence of kinematic parameters between working conditions and different control methods in the shearer cutting process.The optimal control strategy of the shearer self-adaptive cutting under different working conditions was determined:the joint control of v_q-n coordinated speed regulation and drum height self-adaptive adjustment is selected,and the simultaneous control is selected under the working condition that the f value of coal and rock mass decreases.The traction speed takes priority over the drum speed control strategy is selected under the condition that the f value of the coal and rock mass increases.Based on similarity theory,the comprehensive test bench for shearer self-adaptive cutting control was built,Based on OPC technology to establish communication between PLC and Lab VIEW,and complete the construction of the CPS physical system;based on the mixed programming of Lab VIEW and MATLAB,the deep integration of physical system and information system using simulink to establish coal and rock cutting state identification and self-adaptive cutting optimal control strategy of shearer was realized,based on similar test bench experiments,it was verified that it can accurately identify the change of cutting conditions and adjust it according to the optimal control strategy.The obtained shearer traction speed,drum speed and hydraulic cylinder piston displacement’s maximum errors of the similar inversion results and the simulation results are 7.83%、6.46%and 8.17%respectively,which verifies the feasibility of the established coal and rock cutting state identification system and the searched optimal control strategy for the self-adaptive cutting of the shearer,indicating that the information system can realize precise control of the physical system.Combining virtual prototype technology,multi-objective optimization theory,intelligent algorithm of deep reinforcement learning,similarity theory and physical model experiment to study the problem of self-adaptive cutting control of shearers can effectively improve the adaptability of shearers to complex coal seams,and explore a new method for promoting intelligent coal mining.It is an advanced and effective way with strong theoretical significance and engineering application value.This paper has 116 figures,45 tables,224 references.