Method Of Monitoring And Dynamic Planning For Three Machines In A Fully Mechanized Coal-mining Face Under VR Environment

With the continues advance of 'Internet Plus' and 'Made in China 2025' strategy,VR(Virtual reality)technology is being deeply integrated with intelligent and green coal mining technology.'Precision mining' and 'transparent mining' have been put on the agenda,and then the intelligent monitoring and control technology which covers the full time and the all space in a fully mechanized coal-mining face has become the focus of research and development.The current mainstream monitoring existed some defects of the deviation of data and video,the limited load,a character of difficulty to integration and a poor display effect.In order to improve the efficiency and accuracy of remote operation and obtain the accurate information of real-time operation of fully mechanized mining units,three machines equipment,which included the shearer,the scraper conveyor and the hydraulic supports,were taking as the research objects and a virtual image which can reflect the real-time working condition for three machines was established in the VR environment.Then a virtual monitoring and planning method and its prototype system were proposed.This system can conduct VR monitoring and planning at the form of a 3D panorama type display with a high reliability,a high efficiency,a strong timeliness and a clear and delicate picture.Firstly,the condition monitoring and virtual simulation methods of single machine in underground working conditions were studied.Secondly,a series of condition monitoring and simulation methods of three machines under VR environment were studied.It included the virtual collaborative simulation method of three machines in the flat floor condition,the attitude behaviors of coupling characteristics of the shearer and the scraper conveyor in feeding curved sections and under complex conditions and the monitoring method for swarm hydraulic supports memory attitude.Thirdly,the overall design of the condition monitoring system for three machines was studied.It included the VR monitoring method,the VR data real-time synchronization method based on LAN and the multi-software real-time coupling strategy.Then,the dynamic VR planning method for three machines was studied.It included the construction of cooperative mathematical model for three machines and agent model which referred the coupling factors.Finally,a platform for the physical and prototype test of three machines for the condition monitoring and planning was established to conduct the experiments.The main innovations in this paper were as follows:(1)A memory cutting method of virtual shearer was invented.The key technologies and methods of virtual collaborative simulation of three machines in actual underground conditions made a breakthrough.It included the technologies of the virtual walking of shearer,the virtual bending of scraper conveyor,the seamless linkage of hydraulic support parts,the perception among equipment and the virtual 3D model construction of roof and floor model.In the simulation process,it also achieved the controllability of virtual model,interactivity and analysis of the simulation data.(2)A solving method of coupling attitude between the shearer and the scraper conveyor under the two conditions of the uneven floor and the S-shaped bending section respectively was invented.This method can get the real-time coupling relationship between the shearer body pitch angle(or the shearer yaw angle)and the real-time shape of the scraper conveyor.The VR programming software and the information fusion of multi-sensor were used to infer and verify the changing trend of the position and attitude of the shearer.This method can dynamically monitor the operation conditions of the shearer and the scraper conveyor,and provide a theoretical basis for the three-dimensional positioning of the three machines in a fully mechanized coal-mining face.(3)A monitoring solution for a fully mechanized coal-mining face based on collaborative network technology was invented.Using this method,the online coupling of multi-software and the real-time fusion of the monitoring data and the virtual pictures can be realized under 'VR+LAN‘ environment.The phenomena of a big pressure and a carton picture in server host caused by the huge information amount of multi-sensor nodes were solved.This method can also analyze,excavate and predict the historical data and provide a reliable distributed computing solution for panoramic VR monitoring of a fully mechanized coal-mining face.(4)A VR collaborative planning method for the three machines was also proposed based on a multi-agent system(MAS).Using it,a collaborative mathematical model was established that included the effects of a dynamic environment.The communication modes,coordination,and redundant sensing process among multiple agents were also investigated in detail.The online planning and optimal matching of key parameters of three machines were realized.The proposed FMUnitySim offered an effective means and theoretical basis for the rapid planning and safe automatic production of a fully mechanized coal-mining face.