Research On Recognition System Of Coal-rock Cutting State Based On CPS

The shearer is the key equipment for fully mechanized mining face,and the recognition of the cutting state of coal and rock is the key technology that affects the intelligent mining of the shearer in complex coal seams.In order to realize the real-time perception and accurate recognition of coal and rock cutting state information,combined with virtual prototype technology,a coal and rock cutting state recognition method based on the CPS concept is proposed.The acquisition,transmission,processing,and identification of coal and rock cutting state information are different.Multi-domain integration of structured data.By analyzing the physical and mechanical characteristics of coal and rock mass in the 4602 working face of Yangcun Mine of Yankuang Group,EDEM software was used to establish coal wall models with different occurrence conditions.According to the main structural parameters of the spiral drum of the MG2x55/250-BW shearer,the Pro/E software is used to parametrically model the spiral drum and import it into EDEM to construct a coupling model of the spiral drum cutting coal and rock.Through EDEM The virtual prototype experiment simulates the information acquisition of the shearer's physical system,simulates the coupling model of the drum cutting coal wall,and obtains the three-dimensional force data of the coal and rock cutting state under different working conditions.In order to ensure the stable and accurate transmission of coal and rock cutting state information to the information processing system,Ptolemy ?software is used to design an information transmission system based on wireless sensor networks,and the overall energy consumption model of the wireless sensor network is constructed,and simulated by the Ptolemy ? system The verification results show that the data packet loss rate of the constructed wireless sensor network transmission signal is up to 1.71%,the data transmission is relatively stable,and the information about the cutting state of coal and rock can be reliably transmitted.The Coif4 wavelet packet is used to decompose the signal transmitted by the information transmission system with a three-layer wavelet packet.Through the analysis of the frequency domain image of the time domain and DFT conversion of each subband signal,it is found that the coal and rock cutting state signals have certain differences.The sub-band signal energy value is used as the characteristic parameter of the coal and rock cutting state recognition signal,and the neural network input vector is constructed,combined with the PSO-BP neural network,the coal and rock cutting state recognition information processing model is established,and the BP neural network model is selected for comparison and analysis,The results show that:the PSO-BP network model predicts the maximum error of 2.794%,which is higher in accuracy and better than the BP network model,and can make accurate judgments on the cutting state of coal and rock.Finally,relying on the MATLAB/GUI platform,build the coal and rock cutting state recognition system software to realize the real-time data acquisition,online perception and recognition of the coal and rock cutting state recognition information,and provide a theoretical basis for the shearer's adaptive height control control.