| The intelligent control method of the key equipment of working face,namely the shearer,the hydraulic support and the scraper conveyor,is the only way to realize the construction of intelligent unattended working face and achieve the goal of building an intelligent unattended working face and realize the intelligent control of fully mechanized mining equipment.Among them,the electro-hydraulic control system of hydraulic support is the bridge and link that connects the shearer and the scraper conveyor,and its intelligent control method is crucial to realize the automatic operation of fully mechanized mining face.However,the existing intelligent perception technology,intelligent control method and intelligent correction strategy can not meet the needs of automatic operation of fully mechanized mining face,especially the precise positioning of shearer,the monitoring of hydraulic support posture and scraper conveyor straightness correction method,which directly restrict the development of automatic production of unattended working face.Therefore,this paper addresses the above bottleneck problems and focuses on the corresponding intelligent perception and intelligent control methods,which has great theoretical value and practical significance for improving the intelligent level of the coal mine unattended working face,improving production efficiency and improving mining safety.This paper takes the "three machines" of unattended working faces as the research object,with the goal of solving the core problem of intelligent control of key equipment in working faces,improving the accuracy of intelligent perception and control.Through the establishment of inertial measurement unit(IMU)sensor network,analysis error source and characteristics of the filter,the design and optimization of coal accurate positioning method,hydraulic support position accurate perception and control method,scraper conveyor linear awareness correction method and control method of verification test platform were system researched.The specific research contents are as follows:Shearer positioning is susceptible to calculation error and cumulative error.According to this situation,this paper establishes an error propagation model during the positioning process of the strapdown inertial navigation system(SINS),realizes the calculation and prediction of the positioning error of the coal mining machine,and designs the algorithm namely Dual SINS Integrated Positioning Method base on Non-Holonomic Constraints Kalman Filter(NHC-KF)and Adaptive Kalman Filter(AKF).The algorithm suppresses the initial solution algorithm error and cumulative positioning error of the system,ensuring that the system always achieves high positioning accuracy during long-term operation.Simulation research shows that the method proposed in this paper can reduce the positioning error of the SINS by more than 50%,with a maximum reduction of 85%;After long-term movement,multiple orbit changes,and disturbances,the positioning result of the system maintains an error of less than 0.08 m compared to the actual trajectory,showing good trajectory tracking ability and positioning accuracy.The position perception accuracy of hydraulic support is susceptible to environmental noise and random noise interference,which affects its automatic control and affects the production efficiency and safety of the mine.To solve this problem,this paper integrates the principles of direct measurement of hydraulic support posture and indirect analysis of kinematics,designs a model for the perception of hydraulic support posture and height,analyzes the error sources and error characteristics based on the model,and then proposes the Unscented Kalman Filter(UKF)based on the Improved Gradient Descent(IGD)algorithm for accurate perception of the support attitude,and Kalman Filter based on Improved Gradient Descent(IGD-KF)for accurate perception of the support height.These algorithms can effectively suppress and correct attitude angle and support height calculation errors,and improve perception accuracy.Through simulation,it has been proven that the above method can achieve accurate perception of the support posture and support height of hydraulic supports.The maximum error of pitch angle perception is 1.53 °,and the maximum error of height perception is 6.86 mm,which is more than 15% lower than the commonly used methods."Three straight and one flat" is the prerequisite to ensure the automated production of unattended working face.However,the straightness of the scraper conveyor is controlled by many interrelated factors,which seriously affects the detection accuracy of straightness and the control accuracy of the central slot nudging.To this end,this paper studies the structural composition and interconnection relationship of the scraper conveyor,establishes the scraper conveyor alignment trajectory and straightness perception model.The error sources in the process of scraper conveyor straightness perception are analyzed,and then the Maximum Correntropy Criterion Kalman Filter(MCKF)algorithm is proposed and the method of scraper conveyor straightness perception and correction is formed.On the one hand,it improves the accuracy of position perception,and on the other hand,it corrects the displacement,reducing the impact of displacement error on the straightness correction and trajectory tracking of the scraper conveyor.The simulation results show that the maximum perception error of the scraper conveyor coordinate is 31 mm,and the deviation between the accumulated travel and the ideal travel is 0.36 m.This achieves a dual effect of precise perception of the scraper conveyor arrangement trajectory and precise control of the displacement.In order to apply the control method proposed in this article and verify its effectiveness,a electro-hydraulic control system of hydraulic support and a three-machine test platform for the working face were developed based on the requirements of coal mining process.A power circuit with high power conversion efficiency and low output ripple has been designed to meet the high power consumption requirements of intelligent bracket controllers.To meet the high communication bandwidth requirements of the Internet of Things and big data in intelligent systems,an RS485/CAN bus dual channel communication system circuit has been designed,achieving high-speed and reliable communication with high error tolerance and low latency.Based on the above electro-hydraulic control system,this paper designs and builds a simulation test platform for the operation of the "three machines" in a fully mechanized mining face.On the basis of analyzing the mechanical mechanisms and motion principles of key equipment,the full hydraulic drive of the simulated hydraulic support,the complete restoration of the connection relationship between the middle slot of the scraper machine,and the bidirectional walking of the variable frequency electric traction of the coal mining machine were achieved.It maximized the simulation of actual constraint relationships and operational conditions between equipment,meeting the requirements for verifying the effectiveness of control methods.By utilizing the above hydraulic support electro-hydraulic control system and the "three machines" operation simulation test platform of the fully mechanized mining face,this article has completed the effectiveness verification of the proposed intelligent control method.The experimental results show that the hydraulic support electro-hydraulic control system controls the intelligent operation of the three machine simulation platform,achieving a maximum error of 0.08 m for continuous positioning of the coal mining machine and an average error as low as 0.04m;The maximum error in pitch angle perception of hydraulic support posture is 1.13°,and the maximum error in static perception of support height is 6.7mm;The maximum error in trajectory perception of the scraper conveyor is 50 mm,the correction amount of the triple movement trajectory is 129 mm,and the straightness error is 36.5mm.The perception and control accuracy of corresponding physical quantities have been greatly improved,improving the level of intelligent control. |
PDF Full Text Request