CN-122014243-A - Multi-mode feature fusion working face cutting-through position returning control system and method

CN122014243ACN 122014243 ACN122014243 ACN 122014243ACN-122014243-A

Abstract

The invention discloses a multi-mode feature fusion working face cut-through position returning control system and a multi-mode feature fusion working face cut-through position returning control method, wherein the system comprises a double-light fusion camera, a coal cutter control mechanism, an AI analysis server, a data analysis server and a dust sensor; the method comprises the following steps of S1 resetting an encoder when the reset sensor corresponds to the reset magnet, meanwhile transmitting the numerical value of the encoder to the data acquisition module, S2 transmitting the dust concentration value to the AI analysis server by the dust sensor, S3 transmitting the photographed original image to the AI analysis server by the double-light fusion camera to identify the relation between the shearer drum and the cutting-through position, S4 transmitting the identification result to the data acquisition module by the AI analysis server, S5 running a process driving engine by the data acquisition module according to the numerical value of the encoder and the identification result, controlling the working state of the shearer by the process driving engine, and S6 carrying out the height adjustment and the tool returning operation of the shearer drum after the shearer is stopped.

Inventors

LI HONGWEI
LI GUOWEI
BIAN WEI
ZHANG GUOQING
FAN JUNHAO
CHEN HONGLIANG
LI YUNFEI

Assignees

郑州恒达智控科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260114

Claims (10)

1. A multi-mode feature fusion working face cutting-through position tool return control system is characterized by comprising a double-light fusion camera, a coal mining machine control mechanism, an AI analysis server, a data analysis server and a dust sensor, wherein the double-light fusion camera is positioned at the end position of a working face, the double-light fusion camera is fixed on a hydraulic support and faces the coal mining machine cutting-through position, the signal output end of the double-light fusion camera is connected with the signal input end of the AI analysis server, the signal end of the AI analysis server and the signal end of the coal mining machine control mechanism are both in bidirectional connection with the signal end of the data analysis server, the dust sensor is positioned at the end position of the working face and is fixed on the hydraulic support, and the signal output end of the dust sensor is connected with the signal input end of the data analysis server.
2. The multi-mode feature fusion working face cutting-through position returning control system of claim 1, wherein the AI analysis server comprises an image enhancement module for enhancing an original image shot by the double-light fusion camera, and a video recognition module for judging the relation between a drum of the coal mining machine and cutting-through positions according to the enhanced image, a signal input end of the image enhancement module is in communication connection with a signal output end of the double-light fusion camera, a signal output end of the image enhancement module is connected with a signal input end of the video recognition module, a signal output end of the frequency recognition module is connected with a signal input end of the data analysis server through MQTT broadcasting, and a signal output end of the data analysis server is connected with a signal input end of the image enhancement module.
3. The multi-mode feature fusion working face cut-through position back cutter control system according to claim 2, wherein the data analysis server comprises a data acquisition module for carrying out data acquisition and analysis and a process driving engine for carrying out driving operation of the coal cutter, a signal input end of the data acquisition module is connected with a signal output end of the frequency identification module, a signal output end of the dust sensor and a signal output end of the coal cutter control mechanism, a signal output end of the data acquisition module is connected with a signal input end of the image enhancement module and a signal input end of the process driving engine, and a signal output end of the process driving engine is connected with a signal input end of the coal cutter control mechanism.
4. The multi-mode feature fusion working face cut-through position back cutter control system according to claim 3, wherein the coal cutter control mechanism comprises a coal cutter controller for controlling the action of the coal cutter, an encoder for determining the displacement distance of the coal cutter, two reset magnets and a reset sensor, wherein the two reset magnets are arranged on a cable groove at the end position of the working face, the reset sensor is fixed on the coal cutter and corresponds to the position of the reset magnets, the signal output end of the reset sensor is connected with the signal input end of the coal cutter controller, the signal output end of the coal cutter controller is connected with the signal input end of the encoder, the signal output end of the encoder is connected with the signal input end of the data acquisition module, and the signal output end of the process driving engine is connected with the signal input end of the coal cutter controller.
5. The multi-mode feature fusion working face cut-through position return knife control method is implemented by the multi-mode feature fusion working face cut-through position return knife control system according to claim 4 and is characterized by comprising the following steps: S1, when the coal mining machine moves to the state that the reset sensor corresponds to the reset magnet, the reset sensor generates a signal and sends the signal to the coal mining machine controller, the coal mining machine controller resets the encoder, and meanwhile, the encoder transmits the numerical value of the encoder to the data acquisition module; S2, the dust sensor transmits the dust concentration value to the data acquisition module and transmits the dust concentration value to the AI analysis server through the data acquisition module; s3, transmitting the photographed original image to an AI analysis server by the double-light fusion camera, and carrying out image processing and recognition of the relation between the drum of the coal mining machine and the cutting-through position by the AI analysis server according to the dust concentration value; s4, the AI analysis server transmits the recognition result of the relation between the cutting position of the drum of the coal mining machine and the cutting position to the data acquisition module; s5, the data acquisition module operates a process driving engine according to the numerical value of the encoder and the recognition result of the relation between the drum and the cutting-through position of the coal cutter, and the process driving engine sends an instruction to the coal cutter controller and controls the working state of the coal cutter through the coal cutter controller; s6, carrying out height adjustment and cutter returning operation on the roller of the coal mining machine after the coal mining machine is stopped.
6. The method for controlling the cutting-through position of the working face with multi-mode feature fusion as set forth in claim 5, wherein the step S3 comprises the following steps: S31, the double-light fusion camera transmits the shot image to an image enhancement module, and the image enhancement module performs image processing operation on the image shot by the double-light fusion camera and transmits the processed image to a video identification module; S32, the video recognition module judges and recognizes the relation between the drum and the cutting through position of the coal cutter and sends recognition results to the data acquisition module.
7. The method for controlling the cutting-through position of the working face with multi-modal feature fusion as set forth in claim 6, wherein in the step S31, the image processing operation performed by the image enhancement module on the original image captured by the dual-light fusion camera specifically includes the following steps: s311, for the original image Performing large-scale low-pass filtering to obtain a guide image G (x, y), wherein the calculation formula is as follows: ; Wherein, the In order to be an anisotropic diffusion operator, Is the global average of the illumination components, W, H is the width and height of the original image, and N is the total number of pixels; s312, carrying out normalization processing on the guide image according to the dust concentration value to obtain a dust index, wherein the calculation formula is as follows: , ; s313, obtaining a window radius r and a regular term according to the dust index The calculation formula is as follows; ; ; S314, set [ , [7,20] Is set to [ , Is ] Obtaining a denoised image that varies with dust concentration ] ; S315, denoising the image Inputting the sample into an atmospheric scattering model, wherein the calculation formula is as follows: ; Wherein, the Is a haze-free image, t (x, y) is transmissivity, and A is an atmospheric light component; ; ; Wherein, the In order for the transmission rate to be high, Is a regulating factor related to dust index d; S316, obtaining a fog-free image after comprehensive calculation, wherein the calculation formula is as follows: 。
8. The method for controlling the cutting-through position of the working face with multi-modal feature fusion as set forth in claim 7, wherein the step S32 includes the steps of: S321, input haze-free image Scaled to a uniform size value 640 x 640 and normalized to [0,1]; S322, detecting and outputting target bounding box of coal cutter drum through YOLOv-Det model , wherein, For the center coordinates of the shearer drum target bounding box, For the width of the shearer drum target bounding box, The height of a target boundary frame of the drum of the coal mining machine; s323, identifying boundary points of the coal wall, outputting a plurality of boundary point sets of the coal wall, removing isolated points according to the horizontal position X, and setting a confidence filtering threshold value Removing boundary points of the low-credibility coal wall to obtain a screened boundary point set ; S324, boundary point set Fitting to obtain a linear equation ax+by+c=0 describing the boundary line of the coal wall; s325, calculating the normal distance from the center point of the drum of the coal mining machine to the boundary line of the coal wall, wherein the calculation formula is as follows; ; When (when) If the cutting speed is less than or equal to 0, judging that the center of the coal cutter drum reaches or exceeds the boundary line of the coal wall, and judging that the coal cutter drum reaches the cutting position; when (when) And when the coal cutter drum is judged to reach the cutting-through position, the video identification module sends the identification judgment result to the data acquisition module.
9. The method for controlling the cutting-through position of the working face with multi-mode feature fusion according to claim 8, wherein the step S5 is characterized in that the process driving engine controls the working state of the coal mining machine through the coal mining machine controller, and the method specifically comprises the following steps: S51, when the data acquisition module judges that the length of the coal cutter from the cutting-through position is twice as long as the width of the bracket through the numerical value of the encoder, the process driving engine controls the running speed of the coal cutter to be reduced to 5m/min through the coal cutter controller; S52, when the data acquisition module judges that the length of the coal cutter from the cut-through position is the same as the width of the bracket through the numerical value of the encoder, the process driving engine controls the running speed of the coal cutter to be reduced to 3m/min through the coal cutter controller, and carries out voice alarm; S53, the data acquisition module receives the identification judgment result of the video identification module as follows And the value of the encoder is less than or equal to 0, and the process driving engine controls the coal mining machine to stop through the coal mining machine controller after the value of the encoder is reset.
10. The method for controlling the cutting-through position of the working face with multi-modal feature fusion as set forth in claim 9, wherein in step S5, an error range is set for the process driving engine and the process driving engine is set And when the value of the encoder is less than or equal to 0, comparing the value of the encoder with the value of the cutting-through position in the encoder, and adjusting the value of the cutting-through position in the encoder according to the comparison result.

Description

Multi-mode feature fusion working face cutting-through position returning control system and method Technical Field The invention relates to the field of coal mining, in particular to a multi-mode feature fusion working face cut-through position return control system and method. Background The intelligent and normalized fully mechanized mining face is a development trend in the field of coal mining. Under the condition that the fully mechanized coal mining face is few and unmanned, sensing equipment gesture through video and physical sensors becomes one of important guarantees of safe and efficient production of coal mines, and particularly, the coal cutter is capable of recognizing and automatically calibrating cutting-through positions in the coal cutting planning process, and whether the cutting-through positions accurately affect whether the triangular coal on the face can realize full-process automatic coal cutting or not is greatly influenced. In the automatic coal cutting process of triangular coal, the coal cutter advances to a cutting through position, and under the condition of no manual intervention, the coal cutter needs to accurately stop to the cutting through position, the heights of the left roller and the right roller are adjusted, and then the cutter is returned. At present, the means for ensuring accurate cutting-through positions mainly comprise an encoder, the cutting-through positions are influenced by factors such as upward movement, downward movement and the like of a working face, each cutter is required to be manually calibrated, accumulated errors exist in the encoder, periodic manual calibration is required, the operation process is complex, the automation of the whole process of triangular coal is difficult, and the intelligent and normalized pushing of the working face are influenced. Disclosure of Invention The invention aims to solve the problems and provide a working face cutting-through position returning control system and method for improving cutting-through position recognition precision, reducing manual intervention and guaranteeing multi-mode feature fusion of triangular coal automatic and stable operation. In order to achieve the above object, the technical scheme of the present invention is as follows: The utility model provides a multi-mode feature fusion's working face cuts through position and returns sword control system, the system includes two light integration cameras, coal-winning machine control mechanism, AI analysis server, data analysis server, dust sensor, two light integration cameras are located working face end positions, two light integration cameras are fixed on the hydraulic support and towards coal-winning machine cuts through the position, two light integration cameras's signal output part is connected with AI analysis server's signal input part, AI analysis server's signal terminal, coal-winning machine control mechanism's signal terminal is all with data analysis server's signal terminal two-way connection, dust sensor is located working face end positions and fixes on the hydraulic support, dust sensor's signal output part is connected with data analysis server's signal input part. Further, the AI analysis server comprises an image enhancement module for enhancing an original image shot by the double-light fusion camera, and a video recognition module for judging the relation between the drum of the coal mining machine and the cutting-through position according to the enhanced image, wherein a signal input end of the image enhancement module is in communication connection with a signal output end of the double-light fusion camera, a signal output end of the image enhancement module is connected with a signal input end of the video recognition module, a signal output end of the frequency recognition module is connected with a signal input end of the data analysis server through MQTT broadcasting, and a signal output end of the data analysis server is connected with a signal input end of the image enhancement module. Further, the data analysis server comprises a data acquisition module for carrying out data acquisition and analysis and a process driving engine for carrying out driving operation of the coal mining machine, wherein a signal input end of the data acquisition module is connected with a signal output end of the frequency identification module, a signal output end of the dust sensor and a signal output end of the coal mining machine control mechanism, a signal output end of the data acquisition module is connected with a signal input end of the image enhancement module and a signal input end of the process driving engine, and a signal output end of the process driving engine is connected with a signal input end of the coal mining machine control mechanism. Further, the coal cutter control mechanism comprises a coal cutter controller for controlling the action of the coal cutter, an encoder for determining the displacement distance of