CN-122020360-A - Comprehensive evaluation method for all-condition control effect of surrounding rock of coal mine spraying support material

CN122020360ACN 122020360 ACN122020360 ACN 122020360ACN-122020360-A

Abstract

The invention relates to the technical field of coal mine roadway support, and provides a comprehensive evaluation method for the all-working condition of the control effect of surrounding rock of a coal mine spraying support material, which comprises the steps of determining a data characteristic vector according to the test data of spraying materials to be evaluated under all working conditions; the method comprises the steps of inputting data characteristic vectors into a full-working-condition comprehensive evaluation model, outputting full-working-condition control effect probability distribution vectors by the full-working-condition comprehensive evaluation model, determining full-working-condition control effect grades according to the full-working-condition control effect probability distribution vectors, achieving multi-dimensionality such as comprehensive static load, dynamic load, repeated impact, loading and unloading circulation, dynamic and static load collaboration and the like, and guiding engineering practice intuitively.

Inventors

GAO FUQIANG
XI CHAODONG
LIU WENJU
YUAN GUIYANG
LU ZHIGUO
LI PENGJIE
LOU JINFU
YANG LEI
CAO SHUWEN
WANG XIAOQING
LI JIANZHONG
WANG WANJIE

Assignees

中煤科工开采研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20251229

Claims (10)

1. The comprehensive all-condition evaluation method for the control effect of the surrounding rock of the coal mine spraying support material is characterized by comprising the following steps of: Acquiring test data of the spraying material to be evaluated under each working condition; determining a data characteristic vector according to the to-be-evaluated spray material test data under each working condition, wherein the data characteristic vector comprises test scores of each working condition; The data feature vector is input into a full-working-condition comprehensive evaluation model, and the full-working-condition comprehensive evaluation model outputs a full-working-condition control effect probability distribution vector, wherein the full-working-condition comprehensive evaluation model is a model which is obtained by taking a sample data feature vector determined according to sample spraying material test data and a control effect grade label corresponding to each working condition in the sample data feature vector as input and through machine learning training and is used for comprehensively evaluating the full-working-condition control effect of surrounding rock of a coal mine spraying support material; And determining the full-working-condition control effect level according to the full-working-condition control effect probability distribution vector.
2. The method for comprehensively evaluating the control effect of the surrounding rock of the coal mine spraying support material according to claim 1 is characterized by further comprising the steps of obtaining a comprehensive evaluation model of all working conditions, comprising: Sample spraying material test data under all working conditions are obtained, sample data feature vectors are determined according to the sample spraying material test data under all working conditions, and control effect grade labels corresponding to all working conditions in the sample data feature vectors are determined; generating a virtual sample data feature vector according to the sample data feature vector, and determining a control effect grade label corresponding to each working condition in the virtual sample data feature vector; Model training is carried out according to the sample data feature vector, the virtual sample data feature vector and the control effect grade labels corresponding to all working conditions, and a full-working-condition comprehensive evaluation model is constructed; and constraining and updating model parameters based on a preset composite loss function, wherein the composite loss function comprises a classification error loss function and a physical consistency constraint loss function.
3. The method for comprehensively evaluating the control effect of surrounding rock of the coal mine spraying support material according to claim 2, wherein the generating the virtual sample data feature vector according to the sample data feature vector comprises the following steps: carrying out standardization processing on the sample data feature vector to obtain a standardized sample data feature vector; mapping the normalized sample data feature vector into Gaussian distribution parameters of a potential space with preset dimensionality; Based on a re-parameterized extraction mode, sampling data in a potential space, introducing standard normal distributed noise into the sampled data, and generating potential vectors; and constructing a decoder network, mapping the potential vectors into an original feature space, and generating virtual sample data feature vectors.
4. The method for comprehensively evaluating the control effect of surrounding rock of the coal mine spraying support material according to claim 3, wherein the Gaussian distribution parameters comprise a mean vector and a logarithmic variance vector.
5. The method for comprehensively evaluating the control effect of surrounding rock of a coal mine spraying support material according to claim 4, wherein the step of introducing standard normal distribution noise into the sampled data to generate potential vectors comprises the steps of: According to the mean value vector, the logarithmic variance vector and the randomly introduced standard overall distribution noise, the potential vector is determined by adopting the following calculation formula; Wherein, the As a potential vector of the vector, As a mean value vector of the data set, As the variance vector of the variance, Noise is normally distributed for a standard.
6. The comprehensive evaluation method for the control effect of the surrounding rock of the coal mine spraying support material according to claim 1, wherein the working conditions comprise repeated impact, loading and unloading circulation, dynamic and static load coordination, static load basic mechanics and dynamic load impact resistance.
7. The method for comprehensively evaluating the control effect of surrounding rock of the coal mine spraying support material according to claim 3, wherein the step of carrying out standardization processing on the sample data feature vector to obtain the standardized sample data feature vector comprises the following steps: Carrying out standardization processing on the sample data feature vector, and obtaining the standardized sample data feature vector by adopting the following calculation formula; Wherein, the For the normalized sample data feature vector, For the test score for the j-th operating condition, Is the average value in all samples under the j-th working condition, And j epsilon {1, 2, 3, 4, 5}, which is the standard deviation in all samples under the j-th working condition.
8. The utility model provides a colliery spraying support material country rock control effect's full operating mode comprehensive evaluation device which characterized in that includes: the acquisition module is used for acquiring the test data of the spraying material to be evaluated under each working condition; the determining module is used for determining a data characteristic vector according to the to-be-evaluated spray material test data under each working condition, wherein the data characteristic vector comprises test scores of each working condition; The processing module is used for inputting the data feature vector into a full-working-condition comprehensive evaluation model, and outputting a full-working-condition control effect probability distribution vector by the full-working-condition comprehensive evaluation model, wherein the full-working-condition comprehensive evaluation model is a model which is obtained by taking a sample data feature vector determined according to sample spraying material test data and a control effect grade label corresponding to each working condition in the sample data feature vector as input and through machine learning training and is used for comprehensively evaluating the full-working-condition control effect of surrounding rock of a coal mine spraying support material; And the evaluation module is used for determining the full-working-condition control effect level according to the full-working-condition control effect probability distribution vector.
9. An electronic device comprising a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor when executing the program implements the method for comprehensive evaluation of the control effect of surrounding rock of a coal mine spraying support material according to any one of claims 1 to 7.
10. A non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program when executed by a processor implements an all-condition comprehensive evaluation method of the control effect of a coal mine spray support material surrounding rock according to any one of claims 1 to 7.

Description

Comprehensive evaluation method for all-condition control effect of surrounding rock of coal mine spraying support material Technical Field The invention relates to the technical field of coal mine roadway support, in particular to a comprehensive evaluation method for the all-condition control effect of surrounding rock of a coal mine spraying support material. Background Coal resources are important guarantee for energy safety in China. Along with the exhaustion of shallow resources, coal mining gradually extends to the deep part, and the geological environment of surrounding rock of the deep roadway is increasingly complex. Surrounding rock is not only subjected to high ground stress static load, but also frequently faces dynamic impact caused by mining activities, fault activation or roof fracture, and cyclic loading and unloading effects caused by alternate mining footage. The complex mechanical environment of static load, dynamic load, fatigue and rheology has extremely high requirements on the performances of roadway support materials. The existing evaluation system for the control effect of the surrounding rock of the spraying support material still has obvious limitations, and is mainly characterized in that 1) the evaluation dimension is single, and the existing laboratory evaluation method is usually aimed at a single working condition. For example, a simple static load test does not reflect the shatter resistance of a material under impact ground pressure. Although there are test methods for single working conditions such as repeated impact, loading and unloading cycle, dynamic and static combination, etc., the test results are often independent. It is difficult to judge the comprehensive applicability of the material by a single index. 2) Lack of quantitative assessment of "all-condition" adaptability-damage to downhole surrounding rock is often the result of coupling of multiple mechanisms. One material may have very high static strength but insufficient toughness to resist impact, and another material may have good toughness but poor fatigue resistance to failure after multiple cyclic loads. Disclosure of Invention Aiming at the problems existing in the prior art, the invention provides a comprehensive evaluation method for the all-working condition of the surrounding rock control effect of a coal mine spraying support material. The invention provides a comprehensive evaluation method for the all-working condition of the surrounding rock control effect of a coal mine spraying support material, which comprises the following steps: Acquiring test data of the spraying material to be evaluated under each working condition; determining a data characteristic vector according to the to-be-evaluated spray material test data under each working condition, wherein the data characteristic vector comprises test scores of each working condition; The data feature vector is input into a full-working-condition comprehensive evaluation model, and the full-working-condition comprehensive evaluation model outputs a full-working-condition control effect probability distribution vector, wherein the full-working-condition comprehensive evaluation model is a model which is obtained by taking a sample data feature vector determined according to sample spraying material test data and a control effect grade label corresponding to each working condition in the sample data feature vector as input and through machine learning training and is used for comprehensively evaluating the full-working-condition control effect of surrounding rock of a coal mine spraying support material; And determining the full-working-condition control effect level according to the full-working-condition control effect probability distribution vector. According to the full-working-condition comprehensive evaluation method for the control effect of the surrounding rock of the coal mine spraying support material, the method further comprises the steps of obtaining a full-working-condition comprehensive evaluation model, wherein the full-working-condition comprehensive evaluation model comprises the following steps: Sample spraying material test data under all working conditions are obtained, sample data feature vectors are determined according to the sample spraying material test data under all working conditions, and control effect grade labels corresponding to all working conditions in the sample data feature vectors are determined; generating a virtual sample data feature vector according to the sample data feature vector, and determining a control effect grade label corresponding to each working condition in the virtual sample data feature vector; Model training is carried out according to the sample data feature vector, the virtual sample data feature vector and the control effect grade labels corresponding to all working conditions, and a full-working-condition comprehensive evaluation model is constructed; and constraining and updating mode