CN-122020114-A - Coal cutter cutting curve generation method and system, electronic equipment and storage medium

CN122020114ACN 122020114 ACN122020114 ACN 122020114ACN-122020114-A

Abstract

The invention provides a coal cutter cutting curve generation method, a system, electronic equipment and a storage medium, and relates to the technical field of coal cutters. The method comprises the steps of obtaining operation data of at least two nearest coal cutter operations, generating corresponding behavior pattern nodes according to the operation data of each coal cutter operation, obtaining working condition migration data between adjacent coal cutter operations in the at least two nearest coal cutter operations, generating corresponding behavior pattern edges according to the working condition migration data, obtaining behavior patterns of the at least two nearest coal cutter operations based on the behavior pattern nodes and the behavior pattern edges, inputting the behavior patterns into a cutting curve generation model, and obtaining a cutting curve of the next coal cutter operation output by the cutting curve generation model, wherein the cutting curve generation model is obtained by training a sample cutting curve of the next coal cutter operation based on the operation data of at least two sample coal cutter operations, and can improve accuracy of the generated cutting curve of the coal cutter.

Inventors

LI MINGZHONG
LIU QING
YAO YUPENG
LI ZHONGZHONG
XIONG WU
LIU JUNFENG
WU XIAOBAO

Assignees

北京天玛智控科技股份有限公司
北京煤科天玛自动化科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (10)

1. The method for generating the cutting curve of the coal mining machine is characterized by comprising the following steps of: Acquiring operation data of the operation of the at least two latest coal mining machines so as to generate corresponding behavior pattern nodes according to the operation data of the operation of each coal mining machine; acquiring working condition migration data between adjacent coal cutter operations in the operation of the at least two nearest coal cutters so as to generate corresponding behavior pattern edges according to the working condition migration data; acquiring a behavior spectrum of the operation of the nearest at least two-cutter coal mining machine based on the behavior spectrum node and the behavior spectrum edge; Inputting the behavior patterns to a cutting curve generation model to obtain a cutting curve of the next coal cutter operation output by the cutting curve generation model, wherein the cutting curve generation model is obtained by training the sample behavior patterns based on operation data of at least two sample coal cutter operations and the sample cutting curve of the next coal cutter operation.
2. The method for generating a cutting curve of a coal cutter according to claim 1, wherein the step of inputting the behavior pattern into a cutting curve generating model to obtain a cutting curve of a next cutter coal cutter operation output from the cutting curve generating model comprises the steps of: based on a spatial feature extraction layer of the cutting curve generation model, performing spatial semantic feature extraction processing on the behavior spectrum to obtain a coal mining machine operation space embedded vector sequence of the behavior spectrum; based on a time feature extraction layer of the cutting curve generation model, performing time semantic feature extraction processing on the coal cutter operation embedded vector sequence of the behavior spectrum to obtain the coal cutter operation space-time embedded vector sequence of the behavior spectrum; And mapping the time-space embedded vector sequence of the coal mining machine operation of the behavior map based on the full-connection layer of the cutting curve generation model to obtain the cutting curve of the next coal mining machine operation.
3. The method for generating a cutting curve of a coal mining machine according to claim 2, wherein, based on a spatial feature extraction layer of the cutting curve generation model, the behavior pattern is subjected to spatial semantic feature extraction processing, and before the sequence of embedding vectors in the operation space of the coal mining machine of the behavior pattern is obtained, the method further comprises: Acquiring time interval items of the behavior pattern edges so as to calculate a time attenuation factor according to the time interval items; And setting the attention weight of a spatial feature extraction layer of the cutting curve generating model based on the time attenuation factor.
4. The shearer cutting curve generation method of claim 2, wherein the spatial feature extraction layer of the cutting curve generation model comprises at least two graph embedding layers; based on the spatial feature extraction layer of the cutting curve generation model, performing spatial semantic feature extraction processing on the behavior spectrum, and before obtaining the coal mining machine operation space embedded vector sequence of the behavior spectrum, the method further comprises: Determining the static characteristics of each cutter of coal cutter operation according to the parameters of the coal cutter and the coal mining environment; And setting a bias vector of each graph embedding layer of the spatial feature extraction layer of the cutting curve generation model based on the static features, or setting a conditional embedding vector of an output layer of the spatial feature extraction layer of the cutting curve generation model based on the static features.
5. The method for generating a cutting curve of a coal mining machine according to claim 2, wherein the time feature extraction layer based on the cutting curve generation model performs time semantic feature extraction processing on the coal mining machine operation embedded vector sequence of the behavior pattern to obtain the coal mining machine operation space-time embedded vector sequence of the behavior pattern, and the method comprises the following steps: determining static characteristics of each cutter of coal cutter operation according to coal cutter parameters and coal mining environment, and determining an initialization hiding state of a time characteristic extraction layer of the cutting curve generation model based on the static characteristics; Arranging the coal cutter operation embedded vector sequence of the behavior pattern according to time sequence to obtain the time sequence embedded vector sequence of the behavior pattern; and determining the hidden state of each behavior spectrum node in the time sequence embedded vector sequence based on the initialized hidden state of the time characteristic extraction layer of the cutting curve generation model so as to obtain the coal mining machine operation time-space embedded vector sequence of the behavior spectrum.
6. The shearer cutting curve generation method of claim 1, wherein the number of operational data of the at least two-cutter sample shearer operations is greater than the number of operational data of the at least two-cutter shearer operations.
7. The shearer cutting curve generation method of claim 1, wherein the loss function of the cutting curve generation model is determined based on a deviation of a predicted cutting curve from the sample cutting curve, a gradient penalty of operational data of the sample shearer operation, and a deviation of the predicted cutting curve from a coal cliffside boundary, which are output by an original model of the cutting curve generation model; Wherein the coal-rock boundary is obtained after operation of the at least two-cutter sample shearer.
8. A shearer cutting curve generation system, comprising: The behavior pattern node acquisition module is used for acquiring the operation data of the operation of the at least two latest coal mining machines so as to generate corresponding behavior pattern nodes according to the operation data of the operation of each coal mining machine; the behavior pattern edge acquisition module is used for acquiring working condition migration data between adjacent coal cutter operations in the operation of the at least two nearest coal cutters so as to generate corresponding behavior pattern edges according to the working condition migration data; The behavior spectrum acquisition module is used for acquiring the behavior spectrum of the operation of the nearest at least two-cutter coal mining machine based on the behavior spectrum nodes and the behavior spectrum edges; The cutting curve generation module is used for inputting the behavior pattern into a cutting curve generation model to obtain a cutting curve of the next coal cutter operation output by the cutting curve generation model, wherein the cutting curve generation model is obtained by training the sample cutting curve of the next coal cutter operation based on the operation data of at least two sample coal cutter operations.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements the shearer cutting curve generation method of any one of claims 1 to 7 when the computer program is executed by the processor.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the shearer cutting curve generation method of any one of claims 1 to 7.

Description

Coal cutter cutting curve generation method and system, electronic equipment and storage medium Technical Field The invention relates to the technical field of coal mining machines, in particular to a coal mining machine cutting curve generation method, a system, electronic equipment and a storage medium. Background The intelligent construction of coal mines is an important direction of development of coal mine industry in China. In the process, the cutting curve of the coal cutter is used as a key control instruction for guiding the automatic cutting of the coal cutter, and the method has important significance for improving the coal mining efficiency and the overall intelligent level of the coal mine. However, the underground environment of the coal mine is complex, the interface identification degree of the coal and rock is low, and the actual demarcation curve of the coal seam and the surrounding rock is difficult to be directly perceived. At present, the conventional method mainly relies on inertial navigation equipment of a coal mining machine to calculate a cutting curve, but the method has the two limitations that firstly, the accuracy requirement on a sensor is high, and secondly, the real-time change of the posture of the coal mining machine in the cutting process cannot be fully considered. These factors lead to inaccurate cutting curve output, limiting the development of planned cutting techniques to a higher level of intelligence. Disclosure of Invention The invention provides a method, a system, electronic equipment and a storage medium for generating a cutting curve of a coal cutter, which are used for solving the problem of dependence on inertial navigation equipment of the coal cutter in the prior art and improving the accuracy of the generated cutting curve of the coal cutter. The invention provides a coal cutter cutting curve generation method, which comprises the following steps: Acquiring operation data of the operation of the at least two latest coal mining machines so as to generate corresponding behavior pattern nodes according to the operation data of the operation of each coal mining machine; acquiring working condition migration data between adjacent coal cutter operations in the operation of the at least two nearest coal cutters so as to generate corresponding behavior pattern edges according to the working condition migration data; acquiring a behavior spectrum of the operation of the nearest at least two-cutter coal mining machine based on the behavior spectrum node and the behavior spectrum edge; Inputting the behavior patterns to a cutting curve generation model to obtain a cutting curve of the next coal cutter operation output by the cutting curve generation model, wherein the cutting curve generation model is obtained by training the sample behavior patterns based on operation data of at least two sample coal cutter operations and the sample cutting curve of the next coal cutter operation. According to the method for generating the cutting curve of the coal cutter, the behavior pattern is input into the cutting curve generating model to obtain the cutting curve of the operation of the next cutter coal cutter output by the cutting curve generating model, and the method comprises the following steps: based on a spatial feature extraction layer of the cutting curve generation model, performing spatial semantic feature extraction processing on the behavior spectrum to obtain a coal mining machine operation space embedded vector sequence of the behavior spectrum; based on a time feature extraction layer of the cutting curve generation model, performing time semantic feature extraction processing on the coal cutter operation embedded vector sequence of the behavior spectrum to obtain the coal cutter operation space-time embedded vector sequence of the behavior spectrum; And mapping the time-space embedded vector sequence of the coal mining machine operation of the behavior map based on the full-connection layer of the cutting curve generation model to obtain the cutting curve of the next coal mining machine operation. According to the method for generating the cutting curve of the coal mining machine, provided by the invention, based on the spatial feature extraction layer of the cutting curve generation model, the behavior spectrum is subjected to spatial semantic feature extraction processing, and before the embedding vector sequence of the operation space of the coal mining machine of the behavior spectrum is obtained, the method further comprises the following steps: Acquiring time interval items of the behavior pattern edges so as to calculate a time attenuation factor according to the time interval items; And setting the attention weight of a spatial feature extraction layer of the cutting curve generating model based on the time attenuation factor. According to the method for generating the cutting curve of the coal mining machine, the spatial feature extraction layer of t