Intelligent Design Method and System for Coal-Pillar-Free Self-Forming Roadway Based on Machine Learning

Abstract

The present invention discloses an intelligent design method and system for coal pillar-free self-forming roadway based on machine learning, which relates to the technical field of coal mine mining and includes the following steps: S1, establish an intelligent design platform for the 110 mining method; S2, construct a surrounding rock geomechanics model and monitor the deformation and pressure change of the surrounding rock in the model; S3, judge the stability of the surrounding rock and the sufficient filling of the broken expansion of the rock goaf, and complete the decision-making of support parameters, equipment selection, etc.; S4, establish a digital model of the working face; S5, form an intelligent control of "anchoring-cutting-supporting-protecting-control" integrated adaptive coordination through a robot cluster and analyze it through the digital platform of the working face; S6, optimize and form an optimal decision-making model for intelligent coal mining of the 110 mining method. By adopting the above intelligent design method and system for coal pillar-free self-forming roadway based on machine learning, the present invention reduces labor costs, equipment investment and material consumption, reduces the overall construction cost, and promotes the innovation and development of underground engineering technology in China.

Description

Technical Field

[0001] The present invention relates to the technical field of coal mine mining, and in particular to an intelligent design method and system for non-pillar self-forming roadway based on machine learning. Background Technique

[0002] At present, China's intelligent mining technology is in its initial stage, mainly focusing on the application of the 121 mining method. Although the robot system can complete more construction tasks, improve the production efficiency of the mine, replace traditional manual operations, and reduce the dependence on skilled workers, there are still problems such as the setting of mining area coal pillars and gateway coal pillars in the mine, low coal recovery rate, resource waste, low roadway drivage rate, and tight mining and tunneling connection under the intelligent 121 mining method.

[0003] The non-pillar self-forming roadway technology, namely the roof cutting and roadway forming 110 mining method, is one of the effective technologies to maintain the sustainable development of China's coal resources. It can largely reduce or avoid the above problems. The 110 mining method innovatively utilizes the mine pressure, the swelling characteristics of the roof rock mass, and the roadway space to achieve automatic roadway formation and non-pillar mining. This method not only cancels roadway drivage, reduces safety risks, but also uses the roof cutting and pressure relief technology to cut off the pressure transmission between the roadway and the goaf, improving the roadway stability. Compared with the traditional 121 mining method, the 110 mining method eliminates the stress concentration caused by coal pillars, weakens the periodic weighting of the working face, is beneficial to preventing and controlling disasters such as rock bursts and rock bursts, and ensures the safe production of the mine.

[0004] To achieve the full intelligence of mine mining, it is necessary to deeply study the above problems. Only by integrating the concept of intelligent mining with the 110 mining method can various problems in the process of mine mining be truly solved. At present, there are problems in how to sense the parameters of the designed mine; how to judge whether the design method proposed by the system is executable; how to make analysis and decisions according to the actual conditions of the working face; and how to optimize the roadway formation design according to the real-time monitoring data during the execution control. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent design method and system for non-pillar self-forming roadway based on machine learning to solve problems such as resource waste and low roadway drivage rate.

[0006] To achieve the above object, the present invention provides an intelligent design method for non-pillar self-forming roadway based on machine learning, including the following steps:

[0007] S1. Program the designs of all 110 mining method projects, perform machine learning on the data in the database on the platform, and establish an intelligent design platform for the 110 mining method;

[0008] S2. By designing the surrounding rock geomechanical parameters, mine information, and stope conditions of the coal mine working face, construct a surrounding rock geomechanical model. At the same time, monitor the deformation and pressure changes of the surrounding rock in the model, and compare the monitoring results with the actual on-site monitoring data;

[0009] S3. According to the model data and the content of machine learning, judge the stability of the surrounding rock and the full filling of the broken swelling in the goaf, and based on the judgment effect, independently complete the decision-making of support parameters and equipment selection;

[0010] S4. Establish a digital model of the working face and a support model for the two roadways, and establish digital twin models for roof cutting, anchor drilling, support, gangue blocking, and roof control in the formed roadway to realize real-time process mapping of the formed roadway, and integrate a digital intervention model considering geological, panel, disaster, and succession guiding factors;

[0011] S5. According to the decision-making content in S3, transmit it to the robot cluster, complete intelligent roof cutting, drill anchoring, gangue blocking, and temporary support at the working face, and transmit the application effect to the digital platform of the working face in real time, evaluate the effect and feedback it to the perception information section for analysis;

[0012] S6. According to the analysis results of the perception information section, conduct a deduction of the production process, combine the production conditions feedback by the equipment, evaluate the implementation effect of the plan and the production process, and conduct autonomous learning and optimization of the model to achieve the optimal decision-making model for intelligent coal mining with the 110 mining method, and optimize and design the safety guarantee for the working face.

[0013] Preferably, S1 includes the following steps:

[0014] S11. Process the data in previous 110 mining method application cases. The data processing includes data cleaning, feature engineering and selection, oversampling, data analysis, and data standardization;

[0015] S12. Use the BP neural network model for training and prediction. During the training process, calculate the error between the predicted value and the true value, transmit the error back layer by layer through the network, adjust the weights and biases according to the error of each layer, and continuously iterate and update the network parameters until the predetermined training goal is reached.

[0016] Preferably, in S11, the Synthetic Minority Over-sampling Technique (SMOTE) is used to optimize the sample distribution characteristics, oversample the original data set, and the data standardization processing method for the oversampled data is as follows:

[0017]

[0018] Where, represents the sample mean, s represents the sample standard deviation, and n represents the sample size.

[0019] Preferably, in S12, the particle swarm optimization algorithm is used to optimize the initial weights and thresholds of the neural network. Given N particles and a Z-dimensional solution space, initially, a population consisting of N particles is randomly initialized. After calculating the individual extreme value and the global extreme value, the velocity and position are updated according to the following formulas:

[0020]

[0021] u t+1 = u t + u t+1 ;

[0022] where ω t , c1 t , c2 t respectively represent the magnitudes of the influence of particle movement on the local optimal solution and the global optimal solution, v t is the velocity vector, u t is the position vector, p j is the individual extreme value vector, and p k is the global extreme value vector.

[0023] Preferably, in S12, the particle swarm optimization algorithm is used to optimize the initial weights and thresholds of the neural network, and the 6-fold cross-validation method is used to train and predict the model.

[0024] Preferably, the model evaluation indices adopt the mean absolute error EMA, the root mean square error ERMS, and the coefficient of determination R 2 to comprehensively evaluate the prediction effect of the model. The calculation methods are as follows:

[0025]

[0026] Preferably, in S2, the three-dimensional modeling software FLAC3D is used to model the designed working face. During the simulation of the excavation process, the surrounding rock deformation and the change of the surrounding rock pressure are monitored, and a visualization chart is generated.

[0027] Preferably, in S5, the robot cluster consists of a 110-method intelligent cutting slot drill rig, a 110-method intelligent constant resistance cable bolt drill rig, a 110-method multi-functional drill rig support, and a 110 cutting top and rib protection support, forming a 110-method intelligent roadway forming four-machine equipment system.

[0028] The coal pillar-free self-forming roadway intelligent system based on machine learning includes a sensing and recognition layer, an analysis and decision-making layer, and an execution and control layer.

[0029] Therefore, the present invention adopts the above-mentioned coal pillar-free self-forming roadway intelligent design method and system, and has the following beneficial effects:

[0030] 1. The method adopted by the present invention can fundamentally realize the rational development and utilization of mine resources, which can not only reduce the number of workers and improve efficiency, but also avoid problems such as resource waste and low roadway drivage rate, and comprehensively form a new intelligent mine construction model with Chinese characteristics.

[0031] 2. The 110 tunneling method safety and high-efficiency roadway forming robot system adopted by the present invention can realize real-time monitoring and early warning of the construction site during the construction process, effectively preventing accidents. In addition, the robot has an autonomous risk avoidance function and can take timely measures when encountering danger to ensure construction safety; the system adopts automation and intelligent technologies to achieve efficient operation of roadway construction. The robot can accurately complete various construction tasks, reduce manual intervention, and greatly improve the construction speed and quality.

[0032] 3. The method adopted by the present invention can not only reduce labor costs, equipment investment and material consumption, and reduce the overall construction cost, but also contribute to promoting the innovation and development of underground engineering technology in China and improving China's competitiveness in the international underground engineering field.

[0033] The technical solution of the present invention will be further described in detail below through the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the 110 tunneling method intelligent design platform of the present invention;

[0035] Figure 2 It is a data processing flow chart of the present invention;

[0036] Figure 3 It is a topological structure diagram of the BP neural network of the present invention;

[0037] Figure 4 It is a flow chart for establishing the BP neural network model of the present invention

[0038] Figure 5 It is a schematic diagram of the cross-validation principle of the present invention

[0039] Figure 6 It is a process diagram of the four-machine linkage roadway forming of the 110 tunneling method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The technical solution of the present invention will be further described below through the drawings and embodiments.

[0041] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those with ordinary skills in the field to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. The terms such as "comprising" or "including" mean that the elements or objects appearing before the term cover the elements or objects listed after the term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to indicate relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0042] Embodiment

[0043] Please refer to Figures 1-6 , the present invention provides an intelligent design method for coal pillar-free self-forming roadway based on machine learning, including the following steps:

[0044] S1. Combining the promotion and application data of the previous 110 mining method, developing an intelligent design platform for the 110 mining method, and constructing a massive data resource governance system for all-round acquisition, full-network aggregation and full-dimensional integration. Program the designs of all projects that have implemented the 110 mining method, perform machine learning on the data in the database in the platform, so as to establish an intelligent design application support service platform for the 110 mining method, and realize unified data, unified services and unified applications for each project.

[0045] For the data resource governance system, it is necessary to migrate the data stored in the traditional database, such as the stope conditions, mine information, surrounding rock geological conditions and the data of the supporting equipment in the 110 mining method, from the previous 110 mining method application cases to the Hbase distributed database using the Sqoop open-source system. As Figure 2 shown, process the data in the previous 110 mining method application cases. The data processing includes data cleaning, feature engineering and selection, oversampling, data analysis and data standardization. Data cleaning mainly uses the Spark distributed computing system to remove missing values and data error values caused by human reasons; feature engineering refers to the data that is highly correlated with the target data, such as mining depth, mining height, coal seam dip angle, working face obliquity length, etc.; oversampling optimizes the sample distribution characteristics and unifies the data range into a certain interval; data standardization processing is to remove the influence of different dimensions of the samples, reduce the data difference, and improve the stability of the data set.

[0046] Use the Synthetic Minority Over-sampling Technique (SMOTE) to optimize the sample distribution characteristics and perform oversampling on the original data set. The method for data standardization processing after oversampling is shown in the following formula:

[0047]

[0048] As shown Figure 3 in the figure, a BP neural network model is used for training and prediction. During the training process, the error between the predicted value and the true value is calculated, and then the error is passed back layer by layer through the network. According to the error of each layer, the weights and biases are adjusted, and the network parameters are continuously iteratively updated until the predetermined training goal is reached.

[0049] As shown Figure 4 in the figure, the main structure of the BP neural network is determined, and the particle swarm optimization algorithm is further used to optimize the initial weights and thresholds of the neural network. Suppose there are N particles and a Z-dimensional solution space. Initially, a population composed of N particles is randomly initialized. After calculating the individual extreme value and the global extreme value, the speed and position of each particle are updated according to the following formula:

[0050]

[0051] u t+1 = u t + u t+1 ;

[0052] In the formula, v t is the velocity vector, u t is the position vector, p j is the individual extreme value vector, and p k is the global extreme value vector.

[0053] In the initial stage of particle movement, ω t and c2 t are set to larger values to ensure particle diversity and fully search the entire space. In the later stage of iteration, the particles should converge to the global optimal solution. To ensure the convergence effect of the particles, at this time, ω t and c1 t should be smaller while c2 t is larger. The update methods of ωt and c1 t and c2 t are shown in the following formula:

[0054] ω t = [(ω0 - ω1)cos(πt / T) + (ω0 + ω1)] / 2;

[0055]

[0056] To further expand the search space, improve particle diversity, and avoid falling into local optimal solutions, after each update, the particles are initialized with a certain probability β t . The direction of particle mutation is determined by a random number a, and the particle mutation method and the calculation method of β t are shown in the following formula:

[0057]

[0058] βt = [(β1 + β) - (β1 - β0)cos(πt / T)] / 2;

[0059] Among them, to further improve the data utilization efficiency and the generalization performance of the model, a 6-fold cross-validation method is used to train and predict the model, as Figure 5 shown. First, the target data set is divided into 6 non-overlapping subsets. Each time, 5 subsets are taken in sequence as the training set, and the remaining 1 subset is used as the validation set. Training and prediction are carried out 6 times in sequence to obtain 6 prediction models. The average prediction effect of the 6 prediction models is selected to evaluate the model and its parameters. The model evaluation indicators use the mean absolute error (EMA), root mean square error (ERMS), and coefficient of determination (R2) to comprehensively evaluate the prediction effect of the model. The calculation methods are as follows:

[0060]

[0061] Step 2: Enter the surrounding rock geomechanical parameters, mine information, and stope conditions of the designed working face coal mine in the perception and recognition layer. Based on the selected and matched 110 working method working face design according to the previous design usage, further increase the design model attributes. Conduct efficient digital modeling for the machine learning of the digital design of the model, and integrate the three-dimensional design platform, process simulation platform, cost management platform, etc. Use the FLAC3D three-dimensional modeling software to model the designed working face. During the simulation of the excavation process, monitor the surrounding rock deformation and the change of surrounding pressure. Through the asynchronous data interaction method and the FLAC3D three-dimensional modeling means, the digital model is called from the database or server and then undergoes three-dimensional visualization conversion to form a visualization model. Communicate, adjust, and modify with experts to form a design plan or solution, which will be saved in the Hbase sub-database through data interaction.

[0062] Step 3: The system makes a judgment on the surrounding rock stability and the full filling of the broken expansion in the rock goaf according to the design plan generated in Step 2 and the content of machine learning. Then, according to the judgment effect, it autonomously completes the decision-making of support parameters, equipment selection, plan design, design optimization, etc.

[0063] Step 4: Collect important parameters of various equipment in the fully mechanized mining face, such as coal mining technology, geological structure, coal seam roof and floor, hydrology, and other factors affecting coal mining, etc. Unify and integrate all the data, and apply digital twin technology to construct digital twin models for roof cutting, drilling and anchoring, support, gangue blocking, and roof control in the drivage roadway, as well as support models for the two roadways, to realize real-time process mapping of the drivage roadway, and a digital intervention model integrating guiding factors such as geology, panel, disasters, and continuation. To achieve intelligent linkage control of systems such as face mining, transportation, support, liquid supply, video, and personnel positioning, and realize intelligent collaborative control of fully mechanized mining equipment at the face gateway monitoring center.

[0064] Step 5: According to the decision-making content in Step 3, transmit it to the robot cluster, and complete the four-machine linkage drivage roadway process shown in Figure 6 the figure, intelligent drilling and anchoring, roof cutting, gangue blocking, and temporary support, and transmit the working effect to the face digital platform in real time, and evaluate the effect and feedback it to the perception information section for analysis.

[0065] Step 6: According to the analysis results of the perception information section, conduct deduction of the production process, combine the production conditions feedback by the equipment, evaluate the implementation effect of the plan and the production process, and conduct autonomous learning and optimization of the model. Through the application of AI large model technology, conduct in-depth learning and manual correction, accurately drive the mining, effectively implement the drivage roadway process, optimize the production process in real time, achieve the optimal decision-making model of the 110 mining method intelligent design system, guide the actual production, and ensure production safety.

[0066] Therefore, the present invention adopts the above-mentioned coal pillar-free self-forming roadway intelligent design method and system based on machine learning, which can fundamentally realize the rational development and utilization of mine resources, can not only reduce the number of personnel and improve efficiency, but also avoid problems such as resource waste and low roadway drivage rate, and comprehensively form a new intelligent mine construction model with Chinese characteristics. The system adopts automation and intelligent technologies to achieve the efficient operation of roadway construction. The robot can accurately complete various construction tasks, reduce manual intervention, and greatly improve the construction speed and quality. The present invention can not only reduce labor costs, equipment investment, and material consumption, reduce the overall construction cost, but also contribute to promoting the innovation and development of underground engineering technology in China and improving China's competitiveness in the international underground engineering field.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent design method for coal pillarless self-forming roadway based on machine learning, characterized in that The following steps are involved: S1. Program the design of all projects that have implemented the 110 construction method, conduct machine learning on the data in the database on the platform, and establish an intelligent design platform for the 110 construction method; S2. By designing the surrounding rock geomechanical parameters, mine information, and stope conditions of the working face coal mine, a surrounding rock geomechanical model is constructed. At the same time, the deformation and pressure changes of the surrounding rock in the model are monitored, and the monitoring results are compared with the actual on-site monitoring data; S3. Based on the model data and machine learning content, make judgments on the stability of the surrounding rock and the sufficient filling of the rock goaf, and independently complete the support parameters and equipment selection decisions based on the judgment results; S4. Establish a digital model of the working face and a two-lane support model, as well as a digital twin model of the top cutting, anchor drilling, support, rock blocking, and top control of the completed lane, to achieve real-time process mapping of the completed lane, and integrate a digital intervention model that integrates geological, panel, disaster, and succession guidance factors; S5: According to the decision content of S3, it is transmitted to the robot cluster to complete intelligent top cutting, drilling and anchoring, rock blocking and temporary support at the working face, and the application effect is transmitted to the working face digital platform in real time, and its effect is evaluated and fed back to the perception information section for analysis; S6. Based on the analysis results of the perception information section, the production process is deduced, and combined with the production conditions fed back by the equipment, the planning execution effect and production process are evaluated and the model is optimized through autonomous learning, so as to achieve the optimal decision-making model for intelligent coal mining of the 110 working method, optimize the working face and design safety assurance; S1 includes the following steps: S11. Process the data from the previous 110 method application cases. Data processing includes data cleaning, feature engineering and selection, oversampling, data analysis and data standardization; S12, using the BP neural network model for training and prediction, in the training process, by calculating the error between the predicted value and the true value, the error is passed back through the network layer by layer, the weight and bias are adjusted according to the error of each layer, and the network parameters are continuously updated iteratively until the predetermined training target is achieved; In S11, synthetic minority oversampling technology SMOTE is used to optimize the sample distribution characteristics, oversample the original data set, and the method for standardizing the oversampled data is as follows: ； where, represents the sample mean, s represents the sample standard deviation, and n represents the sample size.

2. The intelligent design method for coal pillar-free self-forming roadway based on machine learning according to claim 1, wherein: In S12, the particle swarm algorithm is used to optimize the initial weights and thresholds of the neural network. There are N particles and a Z-dimensional solution space. Initially, a population of N particles is randomly initialized. After calculating the individual extreme value and the global extreme value, the speed and position of the particles are updated according to the following formula: ； ； Among them, and respectively represent the magnitudes of the influence of particle motion on the local optimal solution and the global optimal solution, is the velocity vector, is the position vector, is the individual extreme value vector, is the global extreme value vector.

3. The intelligent design method for coal pillarless self-forming roadway based on machine learning according to claim 2, characterized in that: In S12, the particle swarm algorithm is used to optimize the initial weights and thresholds of the neural network, and the 6-fold cross-validation method is used to train and predict the model.

4. The intelligent design method for non-pillar self-forming roadway based on machine learning according to claim 3, characterized in that: The model evaluation metrics adopt the mean absolute error EMA, the root mean square error ERMS, and the coefficient of determination to comprehensively evaluate the prediction effect of the model. The calculation methods are as follows: ； ； 。 5. The intelligent design method for coal pillar-free self-forming roadway based on machine learning according to claim 4, characterized in that: In S2, the three-dimensional modeling software FLAC3D is used to model the designed working face, simulate the excavation process, monitor the surrounding rock deformation and confining pressure changes, and generate visual charts.

6. The intelligent design method for coal pillarless self-forming roadway based on machine learning according to claim 5, characterized in that: The robot cluster in S5 is a 110 method intelligent lane formation four-machine equipment system consisting of a set of core equipment including the 110 method intelligent slotting drill rig, the 110 method intelligent constant resistance anchor cable drill rig, the 110 method multifunctional drilling rig support and the 110 top cutting and side guard support.

7. A system for applying the intelligent design method for non-pillar self-forming roadway based on machine learning according to any one of claims 1-6, characterized in that: It includes a sensing and recognition layer, an analysis and decision-making layer, and an execution and control layer.

Images