A mine intelligent management and control system and method based on artificial intelligence

Abstract

The present invention relates to the field of intelligent mine management, and in particular to an intelligent mine management and control system and method based on artificial intelligence. The specific contents include: collecting and preprocessing the production data and environmental parameter data of the mine to generate a high-dimensional feature representation; constructing a multi-layer dynamic feedback network based on the high-dimensional feature representation; calculating the probability of failure of mining equipment at a future time based on the output of the multi-layer dynamic feedback network, and obtaining a failure prediction result; introducing a multi-objective scheduling optimization model for production scheduling and resource allocation, and generating an optimal scheduling strategy; real-time monitoring of the production data, environmental parameter data and failure probability of mining equipment; feeding back the monitored production data and environmental parameter data of the mining equipment to the preprocessing operation step in step S1, and further optimizing the scheduling strategy. It solves the problem of low accuracy in predicting equipment status and optimizing scheduling, and inability to effectively warn of potential equipment failures.

Description

Technical Field

[0001] The present invention relates to the field of intelligent mine management, and in particular to an intelligent mine management and control system and method based on artificial intelligence. Background Art

[0002] As the scale of mining resources in the world continues to expand, the demand for automated and intelligent mining production is growing. In the modern mining production process, a large number of equipment are involved in the coordinated operation, such as mining equipment, ventilation equipment, power equipment, etc. The operating status of these equipment has a vital impact on the production efficiency and safety of the mine. At the same time, the production environment of mines is often complex and changeable, and is greatly affected by environmental factors such as temperature, humidity, and air quality. Therefore, how to achieve efficient management of mining equipment in such a complex environment has become a core issue that the industry needs to solve urgently.

[0003] The traditional mine management model relies on manual monitoring and operation. Operators need to schedule and manage according to the production data and equipment status on site. This method is not only inefficient, but also has the risk of errors caused by human factors, and it is difficult to adapt to the needs of large-scale mine production. With the development of sensor technology, Internet of Things technology and automation control technology, the dataization and automation of mine production have gradually improved. In modern mine production, it is possible to collect equipment operation status data, production data and environmental parameter data in real time to provide data support for intelligent scheduling and decision-making. However, simple data collection cannot meet actual needs. How to effectively analyze and process the large amount of collected data and provide scientific decision-making support for production scheduling has become the key to the design of intelligent mine management and control systems.

[0004] In summary, the existing mine management and control systems have the following technical problems: the accuracy in predicting equipment status and optimizing scheduling is not high, and it is unable to effectively warn of potential equipment failures, which increases the risk of production interruptions and equipment downtime, and easily leads to unreasonable resource allocation and excessive or underloaded equipment, thereby affecting production efficiency and equipment life. It may even cause safety problems due to failure to avoid equipment failure risks in a timely manner, making it difficult to form an efficient, safe, and intelligent mine management and control system. Summary of the invention

[0005] The present invention provides an intelligent mine management and control system and method based on artificial intelligence to solve the technical problems of low accuracy in predicting equipment status and optimizing scheduling, inability to effectively warn of potential equipment failures, increased risks of production interruptions and equipment downtime, and easy lead to unreasonable resource allocation and excessive or insufficient equipment load, thereby affecting production efficiency and equipment life, and may even cause safety problems due to failure to timely avoid equipment failure risks, making it difficult to form an efficient, safe, and intelligent mine management and control system.

[0006] The present invention provides an artificial intelligence-based mine intelligent management and control system and method, which specifically includes the following technical solutions:

[0007] An artificial intelligence-based mine intelligent management and control method comprises the following steps:

[0008] S1. Collect and preprocess the production data and environmental parameter data of the mine to generate high-dimensional feature representation; based on the high-dimensional feature representation, build a multi-layer dynamic feedback network;

[0009] S2. Based on the output of the multi-layer dynamic feedback network, the failure probability of the mining equipment at a future time is calculated to obtain the failure prediction result; based on the failure prediction result, a multi-objective scheduling optimization model is introduced to perform production scheduling and resource allocation to generate an optimal scheduling strategy; based on the optimal scheduling strategy, the operation tasks of the mining equipment are allocated and the resource usage is adjusted, and the production data, environmental parameter data and failure probability of the mining equipment are monitored in real time; the monitored production data and environmental parameter data of the mining equipment are fed back to the preprocessing operation step in step S1 to further optimize the scheduling strategy.

[0010] Preferably, the S1 specifically includes:

[0011] The mining production data includes mining equipment status data and production progress data; the received mining production data and environmental parameter data are preprocessed, including data standardization, denoising and mapping; high-dimensional feature representation is obtained by performing high-dimensional mapping on the standardized and denoised mining production data and environmental parameter data.

[0012] Preferably, the S1 specifically includes:

[0013] In the implementation process of multi-layer dynamic feedback network, each layer of feedback network not only depends on the current input, but also combines the feedback information of the previous layer of feedback network. The specific implementation formula is:

[0014] ;

[0015] in, It is The output of the layer feedback network; For the Adaptive feedback weights for layer feedback networks; It is The output of the layer feedback network; is the element-wise product; is the feedback update function; It is The high-dimensional feature representation of the layer feedback network input; It is the environmental parameter data after standardization and denoising.

[0016] Preferably, the S1 specifically includes:

[0017] An adaptive weight adjustment mechanism is introduced to adjust the adaptive feedback weight. The adjustment formula of the adaptive feedback weight is:

[0018] ;

[0019] in, is the adaptive rate.

[0020] Preferably, the S2 specifically includes:

[0021] The output of the multi-layer dynamic feedback network is used as the input variable of the fault prediction model to calculate the failure probability of the equipment at the future moment; the formula of the fault prediction model is:

[0022] ;

[0023] in, Mining equipment in The probability of failure occurrence at the time, indicating the failure prediction result; is the weight matrix for fault prediction; is the bias term; For the The output of the multi-layer dynamic feedback network at time t.

[0024] Preferably, the S2 specifically includes:

[0025] A multi-objective scheduling optimization model is introduced to optimize the scheduling strategy, and the loss function of the multi-objective scheduling optimization is used as the objective function; the objective function is defined as follows:

[0026] ;

[0027] ;

[0028] in, represents the loss function of multi-objective scheduling optimization; For the Production costs at the moment; For the The health status of mining equipment at all times; For the Energy consumption at each moment; For the Scheduling resources at all times; For the The predicted production progress at the moment; For the Actual production progress at all times; , and is the weight coefficient in scheduling optimization; and are the weight matrix and bias term for production schedule prediction respectively.

[0029] Preferably, the S2 specifically includes:

[0030] Based on the objective function, the current scheduling strategy is updated by gradient descent until the optimal scheduling strategy is obtained.

[0031] An artificial intelligence-based mine intelligent management and control system includes the following parts:

[0032] Data acquisition module, data preprocessing module, multi-layer feedback network module, fault prediction module, scheduling optimization module, execution module;

[0033] The data acquisition module is responsible for collecting the production data and environmental parameter data of the mine. The production data of the mine includes the status data of the mine equipment and the production progress data. The data acquisition module is connected with the data preprocessing module;

[0034] The data preprocessing module preprocesses the collected mine production data and environmental parameter data, including data standardization, denoising and mapping. The data preprocessing module is connected to the multi-layer feedback network module;

[0035] The multi-layer feedback network module builds two or more feedback layers through a multi-layer dynamic feedback network. Each layer conducts comprehensive analysis and modeling of the mining equipment status and environmental parameters by combining the current input data and the feedback information of the previous layer of feedback network. The multi-layer feedback network module is connected with the fault prediction module and the scheduling optimization module.

[0036] The fault prediction module uses the output of the multi-layer dynamic feedback network as the input of the fault prediction model to predict the future probability of faults of mining equipment and obtain the fault prediction results. The fault prediction module is connected to the scheduling optimization module;

[0037] The scheduling optimization module optimizes the production scheduling based on the fault prediction results and the health status of the mining equipment to obtain the optimal scheduling strategy; the scheduling optimization module passes the optimal scheduling strategy to the execution module;

[0038] The execution module executes the optimal scheduling strategy, and in the process of implementing the optimal scheduling strategy, it monitors the operating status and environmental parameters of the mining equipment in real time, and feeds the monitored data back to the data acquisition module.

[0039] The beneficial effects of the technical solution of the present invention are:

[0040] 1. The present invention standardizes and denoises the collected production data and environmental parameter data of the mine, and maps the standardized and denoised data to a high-dimensional feature space through a nonlinear mapping function, thereby removing data noise and unifying the scales of data from different sources. It also provides powerful feature expression capabilities by capturing the complex interactive relationship between mining equipment and the environment. High-dimensional mapping enhances the perception of the mining intelligent management and control system of mining equipment status and environmental changes.

[0041] 2. Through the multi-layer dynamic feedback network, combined with the historical status of mining equipment, the current status of mining equipment and environmental parameters, the status of mining equipment is deeply modeled and dynamically adjusted; the multi-layer progressive mechanism in the multi-layer dynamic feedback network can effectively capture the minor anomalies and failure trends that may occur in the equipment during long-term operation, and automatically correct and optimize according to the operation history and environmental changes of the mining equipment; through feedback updates, the intelligent mine management and control system can accurately predict the future status of the equipment, and provide reliable data support for subsequent scheduling optimization and fault prediction.

[0042] 3. Through the fault prediction model, the output of the multi-layer dynamic feedback network is used to calculate the failure probability of mining equipment; during the operation of mining equipment, the fault prediction results can provide early warning of the potential failure risks of mining equipment, help the mining intelligent management and control system to make fault avoidance decisions early, avoid downtime or safety accidents caused by mining equipment failure, and help improve the overall operation safety of mining equipment and ensure production continuity.

[0043] 4. The present invention proposes a multi-objective scheduling optimization model that comprehensively considers production costs, mining equipment health status, energy consumption, and production progress differences to dynamically optimize production scheduling and resource allocation; the multi-objective scheduling optimization model ensures that the mining intelligent management and control system maximizes the utilization efficiency of equipment while ensuring production safety, reduces mining equipment downtime, and reduces production costs and energy consumption; the gradient descent update mechanism of the scheduling strategy enables the mining intelligent management and control system to continuously iterate and optimize in a complex production environment and adapt to changes in the status of mining equipment and the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a structural diagram of an artificial intelligence-based mine intelligent management and control system according to the present invention;

[0045] Figure 2 This is a flow chart of the artificial intelligence-based mine intelligent management and control method described in the present invention. DETAILED DESCRIPTION

[0046] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0048] The following is a detailed description of a specific scheme of an artificial intelligence-based mine intelligent management and control system and method provided by the present invention in conjunction with the accompanying drawings.

[0049] See attached Figure 1 , which shows a structure diagram of an artificial intelligence-based mine intelligent management and control system provided by an embodiment of the present invention, the system includes the following parts:

[0050] Data acquisition module, data preprocessing module, multi-layer feedback network module, fault prediction module, scheduling optimization module, execution module;

[0051] The data acquisition module is responsible for collecting the mine's production data (mine equipment status data, production progress data) and environmental parameter data. The data acquisition module is connected to the data preprocessing module;

[0052] The data preprocessing module preprocesses the collected mine production data and environmental parameter data, including data standardization, denoising and mapping. The data preprocessing module is connected to the multi-layer feedback network module;

[0053] The multi-layer feedback network module builds multiple feedback layers through a multi-layer dynamic feedback network. Each layer conducts comprehensive analysis and modeling of the mining equipment status and environmental parameters by combining the current input data and the feedback information of the previous layer of feedback network. The multi-layer feedback network module is connected with the fault prediction module and the scheduling optimization module.

[0054] The fault prediction module uses the output of the multi-layer dynamic feedback network as the input of the fault prediction model to predict the future probability of faults of mining equipment and obtain the fault prediction results. The fault prediction module is connected to the scheduling optimization module;

[0055] The scheduling optimization module optimizes production scheduling based on the fault prediction results and the health status of mining equipment to ensure efficient and safe production, including multi-objective scheduling optimization model, dynamic resource allocation, gradient descent update, and obtains the optimal scheduling strategy; the scheduling optimization module passes the optimal scheduling strategy to the execution module;

[0056] The execution module executes the optimal scheduling strategy calculated by the scheduling optimization module. In the process of implementing the optimal scheduling strategy, the execution module monitors the operating status and environmental parameters of the mining equipment in real time and feeds the monitored data back to the data acquisition module.

[0057] See attached Figure 2 , which shows a flow chart of a mine intelligent management and control method based on artificial intelligence provided by an embodiment of the present invention, the method comprises the following steps:

[0058] S1. Collect and preprocess the production data and environmental parameter data of the mine to generate high-dimensional feature representation; based on the high-dimensional feature representation, build a multi-layer dynamic feedback network;

[0059] Through the sensors deployed at the mine production site and the mine equipment monitoring interface, the mine's production data (mine equipment status data, production progress data) and environmental parameter data (temperature, humidity, air quality, etc.) are collected, including the multi-dimensional operation information of the mine equipment and the changes in environmental conditions. The mine equipment includes mining equipment, ventilation equipment, power equipment, etc.

[0060] The received production data and environmental parameter data of the mine are preprocessed, including data standardization, denoising and mapping. The main purpose of data preprocessing is to eliminate data noise, unify the scale of data from different sources, and convert the standardized and denoised production data and environmental parameter data into a high-dimensional space for effective feature extraction and analysis in subsequent steps. Assume that the mine equipment status data after standardization and denoising is , the production data is , the environmental parameter data is , the standardized and denoised mining production data and environmental parameter data are mapped to high dimensions through nonlinear mapping functions to form a more expressive feature space. The specific mapping formula is:

[0061]

[0062] in, Represents high-dimensional feature representation; is the weight matrix, is the bias term, Represents the outer product operation of the mining equipment status and production data, is a non-linear activation function.

[0063] Multiple feedback layers are constructed through multi-layer dynamic feedback networks. Each layer of feedback network not only depends on the current input, but also combines the feedback information of the previous layer of feedback network, so as to dynamically adjust the state of the feedback network to adapt to the complex mining environment and changes in the state of mining equipment. In the layer feedback network, the input data is the high-dimensional feature representation of the output of the previous layer feedback network and the input of the current layer feedback network. The feedback update formula is:

[0064]

[0065] in, It is The output of the layer feedback network; It is The high-dimensional feature representation of the layer feedback network input; For the Adaptive feedback weights for layer feedback networks; It is The output of the layer feedback network; is the element-wise product; is the feedback update function, the formula is:

[0066]

[0067] in, is the feedback weight matrix; and is the adjustment factor of historical feedback and environmental parameter data, which controls the weight of historical feedback and environmental parameter data in the current feedback update. The historical feedback is the output of the previous layer of feedback network; It is the bias term of the feedback network and is used to adjust the benchmark of the feedback output. Dynamically adjust the proportion of the mining equipment status in the prediction to ensure that each layer of the feedback network can automatically make corrections based on the historical status of the mining equipment and environmental changes, dynamically optimize the status prediction and analysis of the mining equipment, and enable it to cope with complex environmental conditions and changeable mining equipment operating conditions.

[0068] In order to further enhance the dynamic adaptability of the mine intelligent management and control system, an adaptive weight adjustment mechanism is introduced. The adaptive feedback weight of each layer of the feedback network is not a fixed value, but is automatically adjusted according to the current mine equipment status, feedback error and environmental parameters of the mine intelligent management and control system. The adjustment formula of the adaptive feedback weight is:

[0069]

[0070] in, is the adaptive rate, which is used to control the update speed of the adaptive feedback weight; It represents the inner product of the output of the previous feedback network and the current environmental parameter data, reflecting the mutual influence between the current environment and the state of the mining equipment; Represents feedback error. A nonlinear adjustment mechanism is introduced through an exponential function, so that the adaptive feedback weight can be automatically adjusted according to the current state of the mine intelligent management and control system and the feedback error. The current state of the mine intelligent management and control system is based on the comprehensive information integration of the mine equipment status data, production data, and environmental parameter data of the mine intelligent management and control system.

[0071] The multi-layer dynamic feedback network conducts in-depth analysis and comprehensive modeling of the status and environmental parameters of mining equipment. The production process, operation of mining equipment and environmental conditions of mines are often extremely complex and dynamic. The feedback network can combine the historical status of mining equipment, the current status of mining equipment and environmental factors at each time step through a progressive feedback mechanism to accurately predict the status of the mining intelligent management and control system; the final output of the multi-layer dynamic feedback network is It represents the comprehensive result of the interaction between the current mining equipment status and the environment after multiple layers of feedback and adjustment.

[0072] For example, during the long-term operation of mining equipment, the wear and aging of mining equipment will gradually become apparent, and environmental changes, such as increased temperature and humidity, will also have varying degrees of impact on the operation of mining equipment. The multi-layer dynamic feedback network can accurately identify potential changes in the status of mining equipment through repeated updates, thereby providing a dynamic basis for subsequent optimized scheduling and fault prediction.

[0073] S2. Based on the output of the multi-layer dynamic feedback network, the failure probability of the mining equipment at a future time is calculated to obtain the failure prediction result; based on the failure prediction result, a multi-objective scheduling optimization model is introduced to perform production scheduling and resource allocation to generate an optimal scheduling strategy; based on the optimal scheduling strategy, the operation tasks of the mining equipment are allocated and the resource usage is adjusted, and the production data, environmental parameter data and failure probability of the mining equipment are monitored in real time; the monitored production data and environmental parameter data of the mining equipment are fed back to the preprocessing operation step in step S1 to further optimize the scheduling strategy.

[0074] Mining equipment may produce abnormalities during long-term operation. If not handled in time, it may gradually develop into a serious fault. In fault prediction, the output of the multi-layer dynamic feedback network is used as the core input variable of the fault prediction model to calculate the probability of failure of mining equipment in the future. The fault prediction model formula is:

[0075]

[0076] in, Indicates that mining equipment is The probability of failure at the time, i.e., the failure prediction result; is the weight matrix for fault prediction; is the bias term; For the The output of the multi-layer dynamic feedback network at the moment. Through exponential function mapping, the output of the multi-layer dynamic feedback network is converted into a probability value, thereby achieving accurate prediction of mining equipment failures.

[0077] The result of fault prediction is an estimate of the probability that mining equipment may fail at some point in the future, with the aim of providing early warning and risk avoidance mechanisms for production scheduling optimization. Specifically, the result of fault prediction will directly affect the choice of scheduling strategy, because the scheduling process not only needs to consider the current status of mining equipment, but also needs to adjust the production plan in advance according to the predicted potential risks, so as to avoid possible losses caused by mining equipment downtime or failure.

[0078] If the probability of failure If the pre-set threshold is exceeded, it indicates that the mining equipment is about to fail. In this case, the scheduling optimization module will activate the fault avoidance mechanism, readjust the production tasks and resource allocation, give priority to the mining equipment with lower failure risk, or arrange maintenance time to reduce the possibility of mining equipment failure. Therefore, the probability of failure directly affects the generation of scheduling strategies, ensuring that the intelligent mine management and control system can avoid risks in production in real time.

[0079] In order to realize automated production scheduling and resource allocation, a multi-objective scheduling optimization model is introduced, which comprehensively considers production costs, mining equipment health status, energy consumption and production progress differences, and optimizes scheduling strategies to ensure efficient operation of mining production. The objective function is defined as follows:

[0080]

[0081]

[0082] in, represents the loss function of multi-objective scheduling optimization; For the Production costs at the moment; For the The health status of mining equipment at all times, and the assessment of the current and future comprehensive operating conditions of mining equipment, not only depends on the current status of mining equipment, but also is affected by the historical status of mining equipment and environmental factors; For the Energy consumption at each moment; For the Scheduling resources at all times; For the The predicted production progress at the moment; For the Actual production progress at all times; , and is the weight coefficient in scheduling optimization; and They are the weight matrix and bias term of production schedule prediction respectively. Through multi-objective trade-off, production scheduling and resource utilization are optimized under the premise of ensuring production safety and the health of mining equipment.

[0083] Based on the above objective function, the scheduling optimization process will adjust the current scheduling strategy Perform gradient descent updates until the optimal scheduling strategy is obtained. The formula is:

[0084]

[0085] in, is the current scheduling policy; is the updated scheduling strategy; represents the learning rate; is the total gradient of the scheduling strategy, and the specific calculation formula is:

[0086]

[0087] It is the time range for scheduling optimization. At each gradient update, resource allocation is dynamically adjusted according to the health status and failure risk of mining equipment. For example, the higher the failure risk of mining equipment, the greater the probability of failure, and the more tasks should be allocated to the current mining equipment or mining equipment maintenance should be arranged in advance.

[0088] The optimal scheduling strategy derived from the multi-objective scheduling optimization model will be handed over to the execution module for actual operation. According to the optimal scheduling strategy, the operation tasks of the mining equipment will be allocated and the resource usage will be adjusted, and the operation status and environmental parameters of the mining equipment will be tracked in real time. At the same time, the mining intelligent management and control system will continuously monitor the failure probability of the mining equipment and the actual production progress, and feed back the monitored operation status data, environmental parameter data and production progress data of the mining equipment to the pre-processing operation step in step S1, and further adjust the health status estimation of the mining equipment and optimize the scheduling strategy.

[0089] In summary, an artificial intelligence-based intelligent mine management and control system and method have been completed.

[0090] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0092] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A mine intelligent management and control method based on artificial intelligence, characterized in that: The following steps are involved: S1. Collect the production data and environmental parameter data of the mine, where the mine production data includes the mine equipment status data and production progress data; perform preprocessing operations on the received mine production data and environmental parameter data, including data standardization, denoising and mapping; generate high-dimensional feature representation by performing high-dimensional mapping on the mine production data and environmental parameter data after standardization and denoising; the specific mapping formula is: ； in, Represents high-dimensional feature representation; is the weight matrix; is the bias term; Represents the outer product operation of mining equipment status and production data; is a nonlinear activation function; It is the mining equipment status data after standardization and denoising; To produce data; is the environmental parameter data; Based on high-dimensional feature representation, a multi-layer dynamic feedback network is constructed. In the implementation process of the multi-layer dynamic feedback network, each layer of the feedback network not only depends on the current input, but also combines the feedback information of the previous layer of the feedback network. The specific implementation formula is: ； in, It is The output of the layer feedback network; For the Adaptive feedback weights for layer feedback networks; It is The output of the layer feedback network; is the element-wise product; is the feedback update function; It is The high-dimensional feature representation of the layer feedback network input; is the environmental parameter data after standardization and denoising; S2. Based on the output of the multi-layer dynamic feedback network, the output of the multi-layer dynamic feedback network is used as the input variable of the fault prediction model to calculate the failure probability of the mining equipment at the future moment; the formula of the fault prediction model is: ； in, Mining equipment in The probability of failure occurrence at the time, indicating the failure prediction result; is the weight matrix for fault prediction; is the bias term; For the The output of the multi-layer dynamic feedback network at each moment; Based on the fault prediction results, a multi-objective scheduling optimization model is introduced for production scheduling and resource allocation, and the loss function of the multi-objective scheduling optimization is used as the objective function; the objective function is defined as follows: ； ； in, represents the loss function of multi-objective scheduling optimization; For the Production costs at the moment; For the The health status of mining equipment at all times; For the Energy consumption at each moment; For the Scheduling resources at all times; For the The predicted production progress at the moment; For the Actual production progress at all times; , and is the weight coefficient in scheduling optimization; and They are the weight matrix and bias term of production schedule prediction respectively; is the time range for scheduling optimization; Based on the objective function, the current scheduling strategy is updated by gradient descent until the optimal scheduling strategy is obtained. The formula is: ； in, is the current scheduling policy; is the updated scheduling strategy; represents the learning rate; is the total gradient of the scheduling strategy, and the specific calculation formula is: ； is the time range for scheduling optimization; Based on the optimal scheduling strategy, the operation tasks of the mining equipment are allocated and the resource usage is adjusted, and the production data, environmental parameter data and failure probability of the mining equipment are monitored in real time; the monitored production data and environmental parameter data of the mining equipment are fed back to the preprocessing operation step in step S1 to further optimize the scheduling strategy.

2. The method for intelligent mine management and control based on artificial intelligence according to claim 1, characterized in that: The S1 specifically includes: An adaptive weight adjustment mechanism is introduced to adjust the adaptive feedback weight. The adjustment formula of the adaptive feedback weight is: ； in, is the adaptive rate.

3. An artificial intelligence-based intelligent mine management and control system, applied to the artificial intelligence-based intelligent mine management and control method as claimed in claim 1, characterized in that: Includes the following sections: Data acquisition module, data preprocessing module, multi-layer feedback network module, fault prediction module, scheduling optimization module, execution module; The data acquisition module is responsible for collecting the production data and environmental parameter data of the mine. The production data of the mine includes the status data of the mine equipment and the production progress data. The data acquisition module is connected with the data preprocessing module; The data preprocessing module preprocesses the collected mine production data and environmental parameter data, including data standardization, denoising and mapping. The data preprocessing module is connected to the multi-layer feedback network module; The multi-layer feedback network module builds two or more feedback layers through a multi-layer dynamic feedback network. Each layer conducts comprehensive analysis and modeling of the mining equipment status and environmental parameters by combining the current input data and the feedback information of the previous layer of feedback network. The multi-layer feedback network module is connected with the fault prediction module and the scheduling optimization module. The fault prediction module uses the output of the multi-layer dynamic feedback network as the input of the fault prediction model to predict the future probability of faults of mining equipment and obtain the fault prediction results. The fault prediction module is connected to the scheduling optimization module; The scheduling optimization module optimizes the production scheduling based on the fault prediction results and the health status of the mining equipment to obtain the optimal scheduling strategy; the scheduling optimization module passes the optimal scheduling strategy to the execution module; The execution module executes the optimal scheduling strategy, and in the process of implementing the optimal scheduling strategy, it monitors the operating status and environmental parameters of the mining equipment in real time, and feeds the monitored data back to the data acquisition module.

Images