A cloud platform-based digital mine energy efficiency management and control method, device and medium

Abstract

The application discloses a kind of digital mine energy efficiency management and control method, equipment and medium based on cloud platform, it is related to energy management technical field, including, perception fusion module, for real-time acquisition and cleaning multi-source heterogeneous operation data, and input mine energy efficiency model, mine energy efficiency model is kept synchronous mapping with physical entity by Kalman filtering algorithm, generates virtual mine real-time state package;Simulation prediction module, for industrial cloud platform, virtual mine real-time state package is used as simulation benchmark, to mine energy efficiency model injects preset equipment fault and load mutation and carries out multidimensional emergency simulation test, generates dynamic emergency plan library;Strategy matching module, for by multidimensional feature comparison algorithm, multi-source heterogeneous operation data is matched with dynamic emergency plan library in real time, when completing matching, output corresponding structured plan.The application generates virtual mine real-time state package by perception fusion module, realizes high-precision data fusion and real-time state mapping.

Description

Technical Field

[0001] This invention relates to the field of energy management technology, and in particular to a digital mine energy efficiency management method, equipment and medium based on a cloud platform. Background Technology

[0002] With the rapid development of the Industrial Internet and cloud computing technologies, cloud-based digital mine energy efficiency management equipment has gradually become a core direction for the intelligent transformation of the mining industry. By integrating IoT devices, big data analytics, and cloud services, real-time monitoring and optimization of mine energy consumption can be achieved. In recent years, related methods have focused on data acquisition, model building, and remote control. For example, sensor networks are used to collect equipment operating parameters, and energy efficiency assessments and scheduling are conducted through cloud platforms. These methods aim to improve the automation and precision of mine energy efficiency management, reduce manual intervention, and lower operating costs.

[0003] Existing methods for energy efficiency management in mines have several shortcomings. Regarding data fusion and model synchronization, most methods rely on historical data or periodic updates, making it difficult to achieve real-time and accurate mapping between the physical entities of the mine and the virtual model. This results in energy efficiency models lagging behind actual operating conditions, failing to reflect dynamic events such as equipment failures or sudden load changes in a timely manner, thus affecting the accuracy and timeliness of emergency response. Furthermore, in terms of emergency simulation and strategy generation, existing methods are mostly based on fixed contingency plan libraries or linear simulation tests, lacking multi-dimensional and adaptive emergency mechanisms. This limits the diversity and flexibility of contingency plans and fails to cover the complex and ever-changing mine environment. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a cloud-based digital mine energy efficiency management method, equipment, and medium to address the problems of model synchronization lag and rigid emergency mechanisms.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a cloud-based digital mine energy efficiency management and control device, comprising: a perception fusion module for real-time acquisition and cleaning of multi-source heterogeneous operating data and inputting it into a mine energy efficiency model, wherein the mine energy efficiency model maintains synchronous mapping with physical entities through a Kalman filter algorithm to generate a virtual mine real-time status package; a simulation prediction module for the industrial cloud platform to use the virtual mine real-time status package as a simulation benchmark, inject preset equipment faults and load mutations into the mine energy efficiency model for multi-dimensional emergency simulation testing, and generate a dynamic emergency plan library; a strategy matching module for real-time matching of multi-source heterogeneous operating data with the dynamic emergency plan library through a multi-dimensional feature comparison algorithm, and outputting the corresponding structured plan after matching is completed; and a collaborative control module for the industrial cloud platform to parse the corresponding structured plan into a distributed control instruction set, and drive a multi-agent coordination mechanism with the distributed control instruction set to generate a collaborative operation strategy; The decision report module is used by the equipment controller to execute collaborative operation strategies and monitor energy efficiency response indicators in real time after the execution of collaborative operation strategies; by comparing energy efficiency response indicators before and after the execution of collaborative operation strategies, a control decision report is generated.

[0007] As a preferred embodiment of the cloud-based digital mine energy efficiency management and control equipment of the present invention, the sensing fusion module is connected to the simulation prediction module via a data bus; the simulation prediction module includes a fault injection unit, a load mutation simulation unit, and an emergency plan generation unit. The fault injection unit is used to inject preset equipment faults into the mine energy efficiency model, simulate abnormal conditions, and generate fault simulation data; the load mutation simulation unit is used to simulate load mutation events, calculate the impact of load mutation events on energy efficiency response indicators, and output energy efficiency response calculation results; the emergency plan generation unit is used to generate a dynamic emergency plan library based on the fault simulation data and energy efficiency response calculation results.

[0008] As a preferred embodiment of the cloud-based digital mine energy efficiency management and control equipment of the present invention, the simulation prediction module and the strategy matching module are connected through a high-speed message queue; the strategy matching module includes a feature extraction unit, an algorithm comparison unit, and a matching output unit; the collaborative control module includes an instruction parsing unit, a multi-agent coordination unit, and a strategy generation unit. The feature extraction unit is used to extract energy efficiency feature parameters from multi-source heterogeneous operation data; the algorithm comparison unit is used to match the energy efficiency feature parameters with the plan features in the dynamic emergency plan library through a multi-dimensional feature comparison algorithm, and calculate the matching degree; the matching output unit is used to output the corresponding structured plan when the matching degree exceeds the preset matching degree threshold. The instruction parsing unit is used to parse the corresponding structured plan into a distributed control instruction set; the multi-agent coordination unit is used to construct a multi-agent architecture according to the standard multi-agent definition principles, drive the coordination and communication between device controllers through the multi-agent architecture, and generate coordination results; the strategy generation unit is used to integrate the distributed control instruction set and coordination results to generate a cooperative operation strategy.

[0009] As a preferred embodiment of the cloud-based digital mine energy efficiency management equipment of the present invention, the specific steps for generating the virtual mine real-time status package are as follows: Real-time acquisition of multi-source heterogeneous operational data and execution of data cleaning operations; construction of a data input layer based on the cleaned multi-source heterogeneous operational data, a feature processing layer based on feature extraction and transformation methods, a core computing layer based on machine learning algorithms, and an output layer based on standard data formats; A mine energy efficiency model is constructed based on the input layer, feature processing layer, core computing layer, and output layer. After data cleaning, the multi-source heterogeneous operating data is input into the mine energy efficiency model. The Kalman filter algorithm is used to keep the mine energy efficiency model and the physical entity synchronously mapped, generating a virtual mine real-time status package.

[0010] As a preferred embodiment of the cloud-based digital mine energy efficiency management equipment described in this invention, the specific steps for generating the dynamic emergency response plan database are as follows: The virtual mine real-time status package is used as the initial state for multi-dimensional emergency simulation testing; historical data is retrieved from the mine operation history database, and the values ​​of equipment failure and load mutation are defined based on the historical data; By injecting numerical values ​​of equipment failures and load mutations into the mine energy efficiency model, the internal state and parameters of the mine energy efficiency model are adaptively adjusted, thus forming an adjusted mine energy efficiency model. The adjusted mine energy efficiency model is dynamically simulated using numerical simulation methods to generate energy efficiency response indicators and stability data under different preset scenarios. Optimized response strategies are extracted from energy efficiency response indicators and stability data, and these strategies are compiled into structured contingency plans. These structured contingency plans are then integrated into a dynamic emergency response plan library.

[0011] As a preferred embodiment of the cloud-based digital mine energy efficiency management equipment described in this invention, the step of using a multi-dimensional feature comparison algorithm to perform real-time matching of multi-source heterogeneous operating data with a dynamic emergency plan database, and outputting the corresponding structured plan after matching is completed, is detailed below. Real-time feature vectors and plan feature vectors are extracted from multi-source heterogeneous operational data and dynamic emergency plan database, respectively; A multi-dimensional feature comparison algorithm is applied to calculate the matching degree between the real-time feature vector and the pre-plan feature vector. When the matching degree exceeds the preset matching degree threshold, it is considered a successful match, and the structured pre-plan corresponding to the highest matching degree is output. When the matching degree does not exceed the preset matching degree threshold, it is considered a failed match. In this case, the real-time feature vector and the pre-plan feature vector are extracted again, and the matching degree is recalculated until a successful match is achieved.

[0012] As a preferred embodiment of the cloud-based digital mine energy efficiency management equipment described in this invention, the industrial cloud platform parses the corresponding structured plan into a distributed control instruction set, and uses the distributed control instruction set to drive a multi-agent coordination mechanism to generate a collaborative operation strategy. The specific steps are as follows. Control commands and parameters are extracted from the structured plan, and a distributed control command set is generated. The distributed control instruction set is distributed to each device controller through a communication protocol, a multi-agent coordination mechanism is initiated, multiple coordination results are generated, and the multiple coordination results are integrated into a cooperative operation strategy.

[0013] As a preferred embodiment of the cloud-based digital mine energy efficiency management and control equipment of the present invention, the step of generating a management and control decision report by comparing energy efficiency response indicators before and after the execution of the collaborative operation strategy includes the following specific steps. The real-time monitoring energy efficiency response index after the execution of the collaborative operation strategy refers to the energy efficiency response index calculated based on the equipment operation data after the execution of the collaborative operation strategy; comparing the energy efficiency response index before and after the execution of the collaborative operation strategy, the change in energy efficiency is calculated. The effectiveness of collaborative operation strategies is assessed based on changes in energy efficiency, and optimization suggestions are generated based on these results. A management decision report is generated by integrating changes in energy efficiency, the effectiveness of the strategies, and optimization suggestions through a pre-set report template.

[0014] Secondly, the present invention provides a digital mine energy efficiency management method based on a cloud platform, including real-time collection and cleaning of multi-source heterogeneous operating data, inputting it into a mine energy efficiency model, and using a Kalman filter algorithm to maintain synchronous mapping with physical entities to generate a virtual mine real-time status package. The industrial cloud platform uses the real-time status package of the virtual mine as the simulation benchmark, injects preset equipment failures and load mutations into the mine energy efficiency model to conduct multi-dimensional emergency simulation tests, and generates a dynamic emergency plan library. The multi-dimensional feature comparison algorithm is used to match multi-source heterogeneous operational data with a dynamic emergency plan database in real time. Once the matching is completed, the corresponding structured plan is output. The industrial cloud platform parses the corresponding structured plans into a distributed control instruction set, and uses the distributed control instruction set to drive a multi-agent coordination mechanism to generate collaborative operation strategies. The equipment controller executes the collaborative operation strategy and monitors the energy efficiency response indicators after the execution of the collaborative operation strategy in real time; by comparing the energy efficiency response indicators before and after the execution of the collaborative operation strategy, a control decision report is generated.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the cloud-based digital mine energy efficiency management method as described in the second aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By generating a real-time status package of a virtual mine through the perception fusion module, the problem of model synchronization lag is solved, high-precision data fusion and real-time status mapping are achieved, providing an accurate digital twin foundation, and achieving the effects of improving data accuracy, reducing mapping errors and enhancing real-time response; By generating a dynamic emergency plan library through the simulation prediction module, predictive emergency simulation and dynamic plan generation are realized, enhancing risk resistance and adaptability, providing a diversified plan foundation for real-time strategy matching, and achieving the beneficial effects of improving emergency response efficiency, reducing downtime risk, optimizing resource utilization and enhancing plan flexibility. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a cloud-based digital mine energy efficiency management and control device.

[0019] Figure 2 A flowchart for generating a dynamic emergency response plan library.

[0020] Figure 3 A flowchart for generating a collaborative operation strategy.

[0021] Figure 4 A flowchart for generating control decision reports. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a digital mine energy efficiency management method, equipment, and medium based on a cloud platform, including the following steps: The perception fusion module is used to collect and clean multi-source heterogeneous operating data in real time and input it into the mine energy efficiency model. The mine energy efficiency model maintains synchronous mapping with the physical entity through the Kalman filter algorithm to generate a virtual mine real-time status package. Deploy a sensor network to collect multi-source heterogeneous operating data in real time, and perform data cleaning operations on the multi-source heterogeneous operating data; It should be noted that various types of sensor networks are deployed on key energy-consuming components in the mine, such as ventilators, drainage pumps, air compressors, and hoists. These networks include temperature sensors, current sensors, flow sensors, and position sensors, covering data collection points for equipment energy consumption, environmental parameters, and production status. An industrial wireless communication protocol is configured to ensure network connectivity. The sensor network is activated for real-time data acquisition, continuously collecting multi-source heterogeneous operating data, including parameters such as voltage, current, temperature, pressure, and flow rate, and transmitting this data to the data receiving end of the industrial cloud platform via the industrial wireless communication protocol. The industrial cloud platform performs data cleaning operations on the received multi-source heterogeneous operational data. It calculates the mean and standard deviation of each data point in the multi-source heterogeneous operational data, calculates the difference between each data point and the mean, and uses the ratio of the difference to the standard deviation as the Z-score. It checks whether the absolute value of each Z-score is greater than the Z-score threshold. If it is greater than the Z-score threshold, the current data point is marked as an outlier. All marked outliers are deleted from the multi-source heterogeneous operational data, completing the cleaning operation. All data in the multi-source heterogeneous operational data are uniformly converted into JSON format to eliminate heterogeneity and ensure the consistency and quality of the multi-source heterogeneous operational data.

[0026] It should also be noted that the Z-score threshold is defined based on the theoretical characteristics of the standard normal distribution, and its value ranges from 3.0 to 3.2. If the Z-score threshold is lower than 3.0, more normal data may be mistakenly marked as anomalies, increasing the risk of false positives, leading to excessive data cleaning and information loss. If the Z-score threshold is higher than 3.2, some real outliers may be missed, increasing the risk of false negatives and reducing the reliability of data cleaning.

[0027] A data input layer is constructed based on multi-source heterogeneous operating data, a feature processing layer is constructed based on feature extraction and transformation methods, a core computing layer is constructed based on machine learning algorithms, and an output layer is constructed based on standard data formats. It should be noted that a data interface structure is defined to receive multi-source heterogeneous operating data. The interface supports multiple industrial wireless communication protocols to ensure the input of multi-source heterogeneous operating data. A buffer mechanism is set up to temporarily store the incoming multi-source heterogeneous operating data using a circular queue or memory pool structure to prevent the loss of multi-source heterogeneous operating data and handle traffic peaks. Data verification logic is integrated to check the integrity and basic consistency of multi-source heterogeneous operating data, thus completing the construction of the data input layer. Principal component analysis (PCA) is used for feature identification and extraction to extract energy consumption trend features and load fluctuation index features from multi-source heterogeneous operation data. Based on the Z-score normalization method, the features are standardized to convert the energy consumption trend features and load fluctuation index features into numerical features of a uniform scale to ensure consistency and comparability. The PCA method and the Z-score normalization method are integrated to complete the construction of the feature processing layer. Choose random forest regression or neural network for energy efficiency prediction and calculation; set up the computing structure and parameter storage space to complete the structural configuration of the core computing layer to support algorithm execution; integrate machine learning algorithms into the structure of the core computing layer to complete the construction of the core computing layer; Choosing the JSON standard format defines the output data format for unified data representation; this includes configuring the output interface structure by setting up the data sending mechanism and buffer, enabling the output layer to receive computation results from the core computing layer; and converting the computation results into the standard format by writing conversion code, thus completing the construction of the output layer.

[0028] A mine energy efficiency model is constructed based on the data input layer, feature processing layer, core computing layer, and output layer. It should be noted that the data input layer receives multi-source heterogeneous operating data, and the output is directly used as the input to the feature processing layer; the feature processing layer extracts and transforms features from the input data, and passes the processed feature vector to the core computing layer; the core computing layer uses machine learning algorithms to perform calculations and sends the calculation results to the output layer; the output layer formats the results into a standard data format and outputs them; the input layer, feature processing layer, core computing layer, and output layer are connected in series to construct the mine energy efficiency model. Historical multi-source heterogeneous operation data is obtained by accumulating historical data from sensor networks over a long period of time, and the historical multi-source heterogeneous operation data is divided into training set, validation set and test set. The training set is input into the data input layer for data reception and initial standardization to ensure a consistent data format. The output of the data input layer is passed to the feature processing layer for feature extraction and transformation, generating feature vectors. These feature vectors are then input into the core computation layer, where machine learning algorithms are applied to train the mine energy efficiency model. The internal parameters of the machine learning algorithm are optimized through iterative learning. During training, a validation set is processed through the same data input and feature processing layers and input into the core computation layer for performance verification. Hyperparameters (such as the learning rate) are adjusted based on the verification results to prevent overfitting of the mine energy efficiency model. After training, a test set is processed through the data input and feature processing layers and input into the core computation layer for final evaluation using the trained machine learning algorithm. This generates model predictions, and the coefficient of determination is calculated to confirm the model's accuracy and generalization ability, completing the training of the mine energy efficiency model. The expression for calculating the coefficient of determination is... ; ; ; in, This represents the coefficient of determination, which ranges from 0 to 1. The closer the coefficient of determination is to 1, the better the training effect of the mine energy efficiency model. This represents the sum of squared residuals, reflecting the cumulative magnitude of the prediction error in the mine energy efficiency model. The smaller the value, the better the fit of the mine energy efficiency model. Indicates the first The actual data value of each data point; Indicates the first Model prediction values ​​for each data point; Indicates the index of the data point; The sum of squares represents the total variation of the actual data relative to the mean. This represents the average of the actual data values ​​for all data points. The term "residual" indicates that the current parameter belongs to the residual. "Total" refers to the sum of squares of all data for the current parameter.

[0029] After data cleaning, the multi-source heterogeneous operating data is input into the mine energy efficiency model. The Kalman filter algorithm is used to keep the mine energy efficiency model and the physical entity synchronously mapped, generating a virtual mine real-time status package.

[0030] It should be noted that the cleaned multi-source heterogeneous operational data is input into the data input layer of the mine energy efficiency model through the data interface of the industrial cloud platform. The data input layer performs format unification and buffering processing and outputs a standardized data stream.

[0031] The Kalman filter algorithm is used to dynamically calibrate the current state variables of the mine energy efficiency model with real-time observation data from the physical entity sensor network. The state estimate is calculated through a prediction-update loop: predicting the state of the mine energy efficiency model at the next moment and calculating the prediction uncertainty, expressed as follows: ; ; in, This represents the predicted state vector, which is the estimate of the state vector at the next moment based on the current state vector. The state transition matrix is ​​a standard component of the Kalman filter algorithm. Represents the current state variable; This represents the prediction covariance matrix, i.e., the prediction uncertainty, which reflects the reliability and fluctuation range of the predicted state vector. The covariance matrix representing the current time step is a fundamental component of the Kalman filter algorithm; Represents the matrix transpose symbol; The process noise covariance matrix is ​​a fundamental component of the Kalman filtering algorithm.

[0032] The model predictions are compared with the standardized data stream to generate residuals. The model predictions and observations are then fused using Kalman gain weighting to generate state estimates, which are used to adjust the internal parameters of the mine energy efficiency model to reduce prediction errors. The calibrated mine energy efficiency model state is synchronously mapped with the physical entity. Based on the synchronous mapping results, energy efficiency indicators and equipment operating parameters are extracted, compiled into structured data packets, and used to generate a virtual mine real-time state packet.

[0033] The simulation prediction module is used by the industrial cloud platform to use the real-time status package of the virtual mine as the simulation benchmark, inject preset equipment failures and load changes into the mine energy efficiency model to conduct multi-dimensional emergency simulation tests, and generate a dynamic emergency plan library. The virtual mine real-time status package is used as the initial state for multi-dimensional emergency simulation testing; historical data is retrieved from the mine operation history database, and the values ​​of equipment failure and load mutation are defined based on the historical data; It should be noted that the industrial cloud platform includes an integrated multi-dimensional emergency simulation testing environment. This environment is essentially a virtual space integrating data, algorithms, and business processes, used for simulation, testing, and optimization. It can access the underlying resources of the industrial cloud platform. The real-time status package of the virtual mine is loaded into the multi-dimensional emergency simulation testing environment through the industrial cloud platform's simulation interface. This serves as the initial state for the multi-dimensional emergency simulation test. The system calls the mine's historical operation database and retrieves historical data, including equipment fault records, load fluctuation data, and production logs, through the database query interface, ensuring that historical data covers common risk scenarios. The historical data is then grouped using a K-means clustering algorithm. During the clustering process, the elbow rule is used to determine the optimal number of clusters, and Euclidean distance is used as the similarity metric to ensure that data points are divided into different clusters based on the similarity of fault and load patterns, generating clustering results. The coordinates of the center point and the boundary range of each cluster are extracted from the clustering results. For example, the mean vector of data points within a cluster is calculated as a cluster feature. High-frequency fault features and load mutation patterns are defined based on the cluster features. Equipment faults and load mutations are defined based on the high-frequency fault features and load mutation patterns. For example, "efficiency decrease of 35%" corresponding to the center point of a cluster is defined as a high-frequency fault feature of the fan, and "load fluctuation of 40%-60%" corresponding to the boundary range of another cluster is defined as a load mutation pattern.

[0034] By injecting numerical values ​​of equipment failures and load mutations into the mine energy efficiency model, the internal state and parameters of the mine energy efficiency model are adaptively adjusted, thus forming an adjusted mine energy efficiency model. It should be noted that the numerical values ​​of equipment failures and load mutations are injected into the mine energy efficiency model through the control interface of the industrial cloud platform, and the input parameters or internal variables of the mine energy efficiency model (equipment efficiency coefficient, load level, energy consumption baseline value, environmental parameter threshold, and fault status flag, etc.) are modified. Specifically, the input parameters or internal variables are dynamically adjusted through the API or configuration tools of the industrial cloud platform. For example, for equipment failures, the simulation performance of equipment efficiency coefficient is reduced, or the load level is increased to simulate mutation conditions. Internal variables such as state variables and response coefficients are updated in real time through the Kalman filter algorithm to reduce prediction errors and ensure that the mine energy efficiency model is synchronized with the physical entity. The injected equipment failure and load surge values ​​serve as trigger signals to activate the adaptive adjustment mechanism of the mine energy efficiency model. The Kalman filter algorithm is used to perform dynamic calibration based on real-time observation data (such as actual energy consumption readings or equipment temperature) from the physical entity sensor network. After the dynamic calibration is completed, the internal state and parameters of the mine energy efficiency model are synchronized with the injected scenario to form the adjusted mine energy efficiency model. The adaptive adjustment mechanism refers to the dynamic adjustment of internal state variables (such as equipment efficiency coefficients) and internal parameters (such as energy consumption baseline values) of the mine energy efficiency model through the integrated Kalman filter algorithm, in order to respond to injected equipment failures and load changes, and ensure that the mine energy efficiency model quickly converges to a new synchronization state under abnormal operating conditions, thereby maintaining consistency with the physical entity. It should also be noted that the Kalman filter algorithm is a general real-time data assimilation tool in the scheme. During the generation of the virtual mine's real-time state package, the Kalman filter algorithm is used for basic synchronization between the mine energy efficiency model and the physical entity. Real-time data assimilation ensures that the mine energy efficiency model accurately reflects the actual state under normal operating conditions. After injecting values ​​for equipment failures and load mutations, the Kalman filter algorithm is used for recalibration under abnormal operating conditions. When the injected failure causes a sudden change in the mine energy efficiency model's state, the Kalman filter algorithm re-performs the prediction-update loop based on real-time observation data from the physical entity's sensor network, dynamically adjusting the internal parameters of the mine energy efficiency model so that it quickly converges to a new synchronized state under failure conditions. In the dynamic calibration stage, the Kalman filter algorithm acts as a continuous maintenance mechanism, correcting the deviation between the mine energy efficiency model's predictions and observations in real time, ensuring the reliability of the mine energy efficiency model throughout the entire process. Basic synchronization solves the mapping "from nothing to something," recalibration solves the adaptation to "mutational disturbances," and dynamic calibration solves the maintenance of "continuous operation." These three together form a closed loop.

[0035] The adjusted mine energy efficiency model is dynamically simulated using numerical simulation methods to generate energy efficiency response indicators and stability data under different equipment failures and load mutations. It should be noted that the adjusted mine energy efficiency model is used as the basis for the numerical simulation method. The simulation time step and simulation initial conditions are set. The simulation initial conditions usually include the state variable values ​​of the mine energy efficiency model at the start of the simulation, such as the energy consumption, efficiency status score and temperature of the mining equipment. The simulation initial conditions can be obtained through physical entity sensor network. The existing Runge-Kutta method is used to initiate a dynamic simulation. At each simulation time step, the existing Runge-Kutta method calculates four slopes based on differential equations. These four slopes are then substituted into the weighted average formula to update the state variable values. The dynamic response of the adjusted mine energy efficiency model under numerical scenarios of equipment failure and load mutation is simulated. The expression for the weighted average formula is as follows: ; in, Indicates the first The state variable values ​​at each simulation time step; Indicates the first The state variable values ​​at each simulation time step; This indicates the current simulation time step, which is the starting point of the calculation; This represents the simulation time step, i.e., the time interval between each iteration of the calculation; The starting slope refers to the slope between the current time point and the corresponding state variable value. The intermediate correction slope refers to the slope of the state variable value after adding half the time step to the current time point and correcting it using the starting slope. The optimization slope refers to the slope of the state variable value after adding half the time step to the current time point and correcting it using the intermediate correction slope. The endpoint slope refers to the slope of the state variable value after correction using the optimized slope at the next complete time step; the four slopes are weighted and averaged together to ensure the accuracy and stability of the simulation and reduce prediction errors. During the dynamic simulation, energy efficiency response indicators and stability data are calculated in real time. Energy efficiency response indicators include instantaneous energy consumption, efficiency change rate, and load fluctuation index. Stability data includes oscillation amplitude, recovery time, and steady-state error. The expression for calculating instantaneous energy consumption is: ; in, Indicates time The instantaneous energy consumption value reflects the energy consumption of the equipment at a specific moment; Indicates a specific moment in time; Indicates time The state variable value; The expression for calculating the rate of change in efficiency is: ; in, Indicates time The rate of change of efficiency reflects the rate at which efficiency changes over time. Indicates time Efficiency status score; Indicates time Efficiency status score; Indicates the simulation time step; The expression for calculating the load fluctuation index is as follows: ; in, Indicates time The load fluctuation index; This indicates the number of time points within the current simulation time step; This represents a point-in-time index, with values ​​ranging from 1 to... ; Indicates time The load state variables are the core state variables within the mine energy efficiency model. This represents the average load over the current simulation time step; The term "indicator" is an identifier. The expression for calculating the oscillation amplitude is as follows: ; in, Indicates the amplitude of oscillation; Indicates the start time of the simulation, subscript Indicates "beginning"; Indicates the end time of the simulation, subscript Indicates "finally"; The expression for calculating the recovery time is: ; in, Indicates recovery time, which is the time required for the disturbance to recover to a steady state. The subscript indicates the recovery time. Indicates restoration; Indicates the time when the disturbance occurred, subscript Indicates a disturbance; The subscript represents the steady-state value, that is, the value of the state variable at a stable state. Indicates steady state; This represents the allowable error, defined based on steady-state performance requirements and measurement noise characteristics, and is set to 0.5. If it is greater than 0.5, the judgment standard will be lowered, leading to misjudgment as if it has recovered before it has fully recovered, thus masking residual errors. If it is less than 0.5, it will be oversensitive, leading to prolonged recovery time or incorrect judgment that it cannot be recovered, wasting resources and reducing efficiency. The expression for calculating the steady-state error is as follows: ; in, Indicates steady-state error; The subscript represents the desired steady-state value, i.e., the change in the target state at steady state. Expressing expectations; Optimized response strategies are extracted from energy efficiency response indicators and stability data, and these strategies are compiled into structured contingency plans. These structured contingency plans are then integrated into a dynamic emergency response plan library.

[0036] It should be noted that cluster analysis is used to identify characteristic patterns in energy efficiency response indicators and stability data (such as sudden increases in energy consumption, periodic declines in efficiency, and continuous deviations from the baseline in load), and trend prediction methods (such as time series analysis) are used to determine the direction of state evolution, thereby extracting diverse optimization strategies. For example, a strategy of "reducing equipment operating power" is generated for sudden increases in energy consumption, a strategy of "adjusting equipment maintenance cycle" is generated for periodic declines in efficiency, a strategy of "dynamically adjusting production plan" is generated for load deviations from the baseline, and a strategy of "optimizing control parameters" is generated for increased oscillations. By using contingency plan templates, strategic actions are serialized, specific operational instructions, triggering conditions, execution parameters, and expected goals are defined to form an executable structured contingency plan. These structured contingency plans are then integrated into a dynamic emergency contingency plan library. It should also be noted that the contingency plan template (such as JSON or XML architecture) is created based on the actual needs of mine energy efficiency management. The contingency plan template defines the necessary fields and structure of the structured contingency plan. For example, the operation instruction field is used to store the action type (such as "adjust speed"), the trigger condition field is based on pattern characteristics (such as "efficiency decline trend") rather than a fixed threshold, the execution parameter field contains specific values ​​(such as power adjustment range), and the expected target field describes the expected result. Existing technologies for emergency simulation in mine energy efficiency management typically rely on static testing, such as offline simulation using predefined fault scenarios. This approach has several drawbacks, including a lack of real-time adaptability, inability to dynamically respond to changing operating conditions, and a disconnect between simulation results and reality. This leads to delayed updates to the contingency plan library and low emergency response efficiency. This solution utilizes an industrial cloud platform to use a virtual mine's real-time status package as a simulation benchmark. Pre-defined equipment faults and load surges are injected into the mine's energy efficiency model for multi-dimensional emergency simulation testing. By employing real-time data assimilation and dynamic adjustment mechanisms, it simulates energy efficiency responses under real-world scenarios, thereby generating a dynamic emergency contingency plan library. This enables real-time generation and precise matching of contingency plans, improving the agility and reliability of emergency response, resolving issues of insufficient flexibility and response delays, and ensuring the smooth operation of mine energy efficiency management.

[0037] The strategy matching module is used to match multi-source heterogeneous operational data with a dynamic emergency plan library in real time using a multi-dimensional feature comparison algorithm. Once the matching is complete, the corresponding structured plan is output. Real-time feature vectors and plan feature vectors are extracted from multi-source heterogeneous operational data and dynamic emergency plan database, respectively; It should be noted that, based on standardized data such as real-time collected equipment energy consumption data, environmental parameter data, and production status data, real-time dynamic features are generated through feature extraction methods: Real-time energy consumption trend features are generated by calculating the changing trend through a sliding window analysis of energy consumption data, reflecting short-term energy consumption fluctuations; a dynamic equipment efficiency score is generated by the ratio of output to input energy from temperature, current, flow, and position sensors; a load fluctuation intensity feature is generated by calculating the standard deviation of load data (equipment energy consumption data, environmental parameter data, and production status data) in real time, quantifying the fluctuation intensity; and the real-time energy consumption trend features, dynamic equipment efficiency score, and load fluctuation intensity features are combined into a numerical vector to form a real-time feature vector. One-hot encoding is used to convert each unique action type in the structured emergency plan into a binary vector. For example, if there are three action types in the dynamic emergency plan library, "adjust speed" may be encoded as [1, 0, 0] and "activate backup equipment" may be encoded as [0, 1, 0]. For numerical parameters (such as power adjustment range or delay time), Z-score normalization is used to convert them into numerical features. For example, the power value is mapped from the original range to the interval [0, 1]. The binary vector and the numerical features are concatenated to form the plan feature vector.

[0038] A multi-dimensional feature comparison algorithm is applied to calculate the matching degree between the real-time feature vector and the plan feature vector. When the matching degree exceeds the preset matching degree threshold, it is considered a successful match, and the structured plan corresponding to the highest matching degree is output.

[0039] It should be noted that the similarity value between each proposed feature vector and the real-time feature vector is calculated using the cosine similarity formula to obtain the matching degree. The expression is as follows: ; in, Indicates the first The degree of matching between the pre-plan feature vector and the real-time feature vector; The index representing the feature vector of the contingency plan refers to the total number of feature vectors of the contingency plan in the dynamic emergency plan database; This represents the dimension index of the contingency plan feature vector and the real-time feature vector, with a value ranging from 1 to... ; This represents the total number of dimensions of the contingency plan feature vector and the real-time feature vector, that is, the number of elements contained in the contingency plan feature vector and the real-time feature vector respectively; The first feature vector represents the real-time feature vector. The value of each element; Indicates the first In the feature vector of the _th plan The value of each element; Each matching degree is compared with a matching degree threshold. When the matching degree exceeds the matching degree threshold, the corresponding plan feature vector is marked as a successful match. The plan feature vector with the highest matching degree is selected from all the successfully matched plan feature vectors, and the corresponding structured plan is output. Based on performance requirements and industry standards, a matching threshold was defined. Contingency plan matching records were extracted from historical data, and the cosine similarity distribution was calculated. It was found that a matching threshold above 0.8 increases the false negative rate, leading to stricter matching conditions that may ignore some valid contingency plans, resulting in insufficient response or emergency delays. Conversely, a threshold below 0.8 increases the false positive rate, leading to more lenient matching conditions and more successful matching of contingency plans, resulting in decision-making errors or resource waste and impacting response efficiency. Therefore, 0.8 was chosen as the optimal balance point, maximizing matching accuracy while minimizing the risks of false positives and false negatives. It should also be noted that existing technologies typically use methods based on fixed rules or simple queries for data matching, but these methods have some shortcomings, such as the inability to effectively handle the complexity of multi-source heterogeneous operating data, low matching accuracy leading to false alarms or missed alarms, poor real-time performance making it difficult to adapt to dynamic changes, resulting in response delays and insufficient flexibility. This solution uses a multi-dimensional feature comparison algorithm to calculate the matching degree using cosine similarity and combines it with a matching degree threshold for judgment, achieving high-precision matching and real-time response, improving the accuracy and efficiency of emergency decision-making, and solving the problems of insufficient processing capacity, slow response, and high false alarm rate.

[0040] The collaborative control module is used by the industrial cloud platform to parse the corresponding structured plan into a distributed control instruction set, and use the distributed control instruction set to drive the multi-agent coordination mechanism to generate collaborative operation strategies. The industrial cloud platform extracts control instructions and control parameters from structured plans and generates a distributed control instruction set; It should be noted that the industrial cloud platform extracts equipment control instructions (such as equipment start / stop and speed adjustment) and control parameters (such as power settings and time delays) from the structured plan through a data parsing interface; based on a pre-loaded real-time equipment topology distribution configuration file, it decomposes the control instructions into sub-instructions for specific devices, for example, decomposing the global instruction "adjust all fan speeds" into separate instructions for each fan; it assigns a target device identifier (such as device ID) and execution priority to each sub-instruction; and it encapsulates all sub-instructions into a distributed control instruction set in JSON format. It should also be noted that the execution priority is determined based on the equipment's criticality level and the operational urgency level. Equipment criticality is classified into three levels based on its functional importance in mine operations: Level 1 is critical equipment, such as main ventilation fans, drainage pumps, and hoists, because their failure would directly lead to a complete production halt, safety accidents, or significant economic losses; Level 2 is important equipment, such as conveyor belts and air compressors, because their failure would affect local operating efficiency but not cause a complete shutdown; Level 3 is auxiliary equipment, such as lighting systems and monitoring sensors, because their failure would only have a minor impact or be limited to auxiliary functions. Based on the timeliness requirements and potential impact of the operations, the urgency level is classified into three levels: Emergency Operations (Level 1), which refer to operations that need to be performed immediately to avoid risks or prevent accidents, such as fault emergency response, safety alarm handling, and load surge adjustment; Routine Operations (Level 2), which refer to planned or delayed operations, such as regular equipment maintenance, parameter optimization and adjustment, and energy efficiency assessment; and Background Operations (Level 3), which refer to operations that have no direct impact on real-time operations, such as data log updates, historical data backups, and report generation.

[0041] The distributed control instruction set is distributed to each device controller through a communication protocol, and a multi-agent coordination mechanism is initiated to generate multiple coordination results, which are then integrated into a collaborative operation strategy.

[0042] It should be noted that in the industrial cloud platform, a distributed control instruction set is distributed to each device controller via the MQTT communication protocol. The distributed control instruction set includes sub-instructions, target device identifiers, and execution priorities. After receiving the distributed control instruction set, the device controller, acting as an intelligent agent, initiates a multi-agent coordination mechanism. Based on the MQTT communication protocol, it negotiates tasks and resolves conflicts. Each intelligent agent generates coordination results based on its local state and instruction requirements, such as task bidding, parameter adjustment suggestions, or resource allocation schemes. The coordination results are aggregated through a data interface and integrated into a unified collaborative operation strategy. The multi-agent coordination mechanism refers to a distributed collaborative framework defined based on the multi-agent principle in an industrial cloud platform. Multiple agents (such as device controllers) negotiate tasks, resolve conflicts, and allocate resources through communication protocols to achieve common goals. In this solution, the multi-agent coordination mechanism is implemented by a multi-agent coordination unit, driving real-time coordination and communication between device controllers. For example, after receiving a distributed control instruction set, agents generate coordination results through bidding, negotiation, or voting processes, supporting the generation of collaborative operation strategies. The multi-agent coordination mechanism solves the single point of failure and response delay problems of centralized control. It should also be noted that existing technologies typically handle emergency responses through manual intervention or simple automated scripts, but these have shortcomings, such as slow response speed, susceptibility to human error, and lack of multi-device collaboration capabilities, resulting in low emergency response efficiency and poor decision-making consistency. This solution automatically parses structured contingency plans into a distributed control instruction set through an industrial cloud platform and uses a multi-agent coordination mechanism to drive intelligent collaboration between device controllers, thereby generating collaborative operation strategies. This achieves the benefits of improved response speed, enhanced accuracy, and ensured global coordination, solving the problems of insufficient collaboration, poor real-time performance, and high error rates.

[0043] The decision report module is used by the equipment controller to execute collaborative operation strategies and monitor energy efficiency response indicators in real time after the execution of collaborative operation strategies; by comparing energy efficiency response indicators before and after the execution of collaborative operation strategies, a control decision report is generated.

[0044] The system collects equipment operation data after the implementation of the collaborative operation strategy through a sensor network, and calculates the energy efficiency response index in real time after the implementation of the collaborative operation strategy; it compares the energy efficiency response index before and after the implementation of the collaborative operation strategy and calculates the change in energy efficiency. It should be noted that the operating data (including temperature, current and equipment energy consumption) after the equipment executes the collaborative operation strategy is continuously monitored through a sensor network such as temperature sensors and current sensors. Based on the operating data, the instantaneous value of energy consumption, efficiency change rate and load fluctuation index after the collaborative operation strategy is executed are calculated. The energy efficiency response indicators before and after the execution of the collaborative operation strategy are compared. For instantaneous energy consumption, the difference between the instantaneous energy consumption after execution and the instantaneous energy consumption before execution is taken as the energy efficiency change. For efficiency change rate, the ratio of the difference in efficiency change rate before and after execution to the efficiency change rate before execution is calculated and converted into a percentage energy efficiency change. For load fluctuation index, the energy efficiency change of the load fluctuation index is generated by calculating the difference in the load fluctuation index before and after execution.

[0045] The effectiveness of the collaborative operation strategy is assessed based on the change in energy efficiency, and specific optimization suggestions are generated based on the effectiveness. It should be noted that when using trend analysis methods (such as linear regression to calculate the slope) to analyze changes in energy efficiency, preset boundary values ​​are used to classify the effect levels: when the slope is greater than or equal to the improvement boundary, it is assessed as an improvement level, indicating that the coordinated operation strategy effectively promotes energy efficiency improvement; when the slope is less than or equal to the deterioration boundary, it is assessed as a deterioration level, indicating energy efficiency degradation; when the slope is between the deterioration boundary and the improvement boundary, it is assessed as a stable level, indicating that there is no significant change in effect. Collect historical operating data of mining equipment, including time series data such as instantaneous energy consumption, efficiency change rate, and load fluctuation index; apply linear regression analysis to calculate the slope value for each analysis window (e.g., on a weekly basis) to form a slope distribution; determine the significance level through statistical hypothesis testing; analyze whether the slope distribution differs significantly from the null hypothesis that the slope is zero; select the slope value corresponding to 0.05 as the improvement boundary; if the improvement boundary is lower than 0.05, it will increase the risk of false alarms, misjudging noise as a trend, leading to decision-making errors; while if it is higher than 0.05, it will increase the risk of false negatives, ignoring small but real improvements and reducing response sensitivity. Based on historical data analysis and statistical principles, a deterioration boundary is defined. Records of equipment performance deterioration events and corresponding time series data of energy efficiency indicators are selected from historical data. At the same time, time series data of energy efficiency indicators under normal operating conditions are selected for comparison. Statistical distribution analysis (such as fitting the probability distribution of the slope of energy efficiency changes during deterioration and normal periods) is applied to determine that the deterioration boundary value is -0.05. If the deterioration boundary is greater than -0.05, slight deterioration may be ignored, increasing the risk. If it is less than -0.05, it will be oversensitive, misjudging normal fluctuations as deterioration, leading to unnecessary adjustments. For the "Improvement Level" result, it indicates that the strategy is effective, and the generated optimization suggestion is "maintain the current strategy and extend it to similar equipment"; for the "Stable Level" result, it indicates that the effect has not changed significantly, and the generated optimization suggestion is "continue to monitor and collect more data"; for the "Deterioration Level" result, it indicates that energy efficiency has deteriorated, and the generated optimization suggestion is "adjust equipment parameters or increase maintenance frequency".

[0046] By integrating energy efficiency changes, implementation results, and optimization suggestions through preset report templates, a management and control decision report is generated.

[0047] It should be noted that a preset report template (a standardized structure defined according to industry standards and actual business needs) is used to fill in the corresponding fields with energy efficiency changes, implementation results, and optimization suggestions. For example, energy efficiency changes are filled in a data table, implementation results are filled in an evaluation summary, and optimization suggestions are filled in an action guide. The current timestamp and equipment identifier are added to the filled report template to ensure that the report template contains metadata for traceability and identification. The report template is then converted into a structured document using a document generation tool to generate a management and control decision report.

[0048] This embodiment also provides a cloud-based digital mine energy efficiency management method, including: Real-time acquisition and cleaning of multi-source heterogeneous operation data, which is then input into the mine energy efficiency model. The mine energy efficiency model maintains synchronous mapping with the physical entity through the Kalman filter algorithm, generating a virtual mine real-time status package. The industrial cloud platform uses the real-time status package of the virtual mine as the simulation benchmark, injects preset equipment failures and load mutations into the mine energy efficiency model to conduct multi-dimensional emergency simulation tests, and generates a dynamic emergency plan library. The multi-dimensional feature comparison algorithm is used to match multi-source heterogeneous operational data with a dynamic emergency plan database in real time. Once the matching is completed, the corresponding structured plan is output. The industrial cloud platform parses the corresponding structured plans into a distributed control instruction set, and uses the distributed control instruction set to drive a multi-agent coordination mechanism to generate collaborative operation strategies. The equipment controller executes the collaborative operation strategy and monitors the energy efficiency response indicators after the execution of the collaborative operation strategy in real time; by comparing the energy efficiency response indicators before and after the execution of the collaborative operation strategy, a control decision report is generated.

[0049] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the cloud-based digital mine energy efficiency management method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0050] In summary, this invention addresses the problem of model synchronization lag by generating a virtual mine real-time status package through a perception fusion module, achieving high-precision data fusion and real-time status mapping, providing an accurate digital twin foundation, and improving data accuracy, reducing mapping errors, and enhancing real-time response. Furthermore, the invention generates a dynamic emergency plan library through a simulation prediction module, enabling predictive emergency simulation and dynamic plan generation, enhancing risk resistance and adaptability, providing a diversified plan foundation for real-time strategy matching, and achieving the beneficial effects of improving emergency response efficiency, reducing downtime risks, optimizing resource utilization, and enhancing plan flexibility.

[0051] 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A cloud platform-based digital mine energy efficiency management and control device, characterized in that: include, The perception fusion module is used to collect and clean multi-source heterogeneous operating data in real time and input it into the mine energy efficiency model. The mine energy efficiency model maintains synchronous mapping with the physical entity through the Kalman filter algorithm to generate a virtual mine real-time status package. The simulation prediction module is used by the industrial cloud platform to use the real-time status package of the virtual mine as a simulation benchmark. It injects preset equipment failures and load surges into the mine energy efficiency model to conduct multi-dimensional emergency simulation tests, generating a dynamic emergency plan library. The specific steps are as follows: The virtual mine real-time status package is used as the initial state for multi-dimensional emergency simulation testing; historical data is retrieved from the mine operation history database, and the values ​​of equipment failure and load mutation are defined based on the historical data; By injecting numerical values ​​of equipment failures and load mutations into the mine energy efficiency model, the internal state and parameters of the mine energy efficiency model are adaptively adjusted, thus forming an adjusted mine energy efficiency model. The adjusted mine energy efficiency model is dynamically simulated using numerical simulation methods to generate energy efficiency response indicators and stability data under different preset scenarios. Extract optimized response strategies from energy efficiency response indicators and stability data, compile these strategies into structured contingency plans, and integrate these structured contingency plans into a dynamic emergency response plan library. The strategy matching module is used to perform real-time matching of multi-source heterogeneous operational data with a dynamic emergency response plan database using a multi-dimensional feature comparison algorithm. After the matching is completed, the corresponding structured plan is output. The specific steps are as follows. Real-time feature vectors and plan feature vectors are extracted from multi-source heterogeneous operational data and dynamic emergency plan database, respectively; Real-time energy consumption trend characteristics are obtained by analyzing the sliding window change trend of equipment energy consumption data. The dynamic score of equipment efficiency is obtained based on the ratio of equipment output energy to equipment input energy. The load fluctuation intensity characteristics are obtained by calculating the standard deviation of real-time load data. The real-time energy consumption trend characteristics, equipment efficiency dynamic score and load fluctuation intensity characteristics are combined to form a real-time feature vector. The action types in the structured plan are encoded one-hot to form binary vectors. The power adjustment range and delay time in the structured plan are normalized by Z-score to form numerical features. The binary vectors and numerical features are then concatenated to form the plan feature vector. A multi-dimensional feature comparison algorithm is applied to calculate the matching degree between the real-time feature vector and the pre-plan feature vector. When the matching degree exceeds the preset matching degree threshold, it is considered a successful match, and the structured pre-plan corresponding to the highest matching degree is output. If the matching degree does not exceed the preset matching degree threshold, it is considered a matching failure. Then, the real-time feature vector and the pre-planned feature vector are extracted again, and the matching degree is recalculated until the matching is successful. The collaborative control module is used by the industrial cloud platform to parse the corresponding structured plan into a distributed control instruction set, and use the distributed control instruction set to drive the multi-agent coordination mechanism to generate collaborative operation strategies. The decision report module is used by the equipment controller to execute collaborative operation strategies and monitor energy efficiency response indicators in real time after the execution of collaborative operation strategies; by comparing energy efficiency response indicators before and after the execution of collaborative operation strategies, a control decision report is generated.

2. The cloud-based digital mine energy efficiency management equipment as described in claim 1, characterized in that: The perception fusion module is connected to the simulation prediction module via a data bus; the simulation prediction module includes a fault injection unit, a load mutation simulation unit, and an emergency plan generation unit. The fault injection unit is used to inject preset equipment faults into the mine energy efficiency model, simulate abnormal conditions, and generate fault simulation data; the load mutation simulation unit is used to simulate load mutation events, calculate the impact of load mutation events on energy efficiency response indicators, and output energy efficiency response calculation results; the emergency plan generation unit is used to generate a dynamic emergency plan library based on the fault simulation data and energy efficiency response calculation results.

3. The cloud-based digital mine energy efficiency management equipment as described in claim 2, characterized in that: The simulation prediction module and the policy matching module are connected via a high-speed message queue; the policy matching module includes a feature extraction unit, an algorithm comparison unit, and a matching output unit; the cooperative control module includes an instruction parsing unit, a multi-agent coordination unit, and a policy generation unit. The feature extraction unit is used to extract energy efficiency feature parameters from multi-source heterogeneous operation data; the algorithm comparison unit is used to match the energy efficiency feature parameters with the plan features in the dynamic emergency plan library through a multi-dimensional feature comparison algorithm, and calculate the matching degree; the matching output unit is used to output the corresponding structured plan when the matching degree exceeds the preset matching degree threshold. The instruction parsing unit is used to parse the corresponding structured plan into a distributed control instruction set; the multi-agent coordination unit is used to construct a multi-agent architecture according to the standard multi-agent definition principles, drive the coordination and communication between device controllers through the multi-agent architecture, and generate coordination results; the strategy generation unit is used to integrate the distributed control instruction set and coordination results to generate a cooperative operation strategy.

4. The cloud-based digital mine energy efficiency management equipment as described in claim 3, characterized in that: The specific steps for generating the real-time status packet of the virtual mine are as follows: Real-time acquisition of multi-source heterogeneous operational data and execution of data cleaning operations; construction of a data input layer based on the cleaned multi-source heterogeneous operational data, a feature processing layer based on feature extraction and transformation methods, a core computing layer based on machine learning algorithms, and an output layer based on standard data formats; A mine energy efficiency model is constructed based on the input layer, feature processing layer, core computing layer, and output layer. After data cleaning, the multi-source heterogeneous operating data is input into the mine energy efficiency model. The Kalman filter algorithm is used to keep the mine energy efficiency model and the physical entity synchronously mapped, generating a virtual mine real-time status package.

5. The cloud-based digital mine energy efficiency management equipment as described in claim 1, characterized in that: The industrial cloud platform parses the corresponding structured plan into a distributed control instruction set, which drives a multi-agent coordination mechanism to generate a collaborative operation strategy. The specific steps are as follows. Control commands and parameters are extracted from the structured plan, and a distributed control command set is generated. The distributed control instruction set is distributed to each device controller through a communication protocol, a multi-agent coordination mechanism is initiated, multiple coordination results are generated, and the multiple coordination results are integrated into a cooperative operation strategy.

6. The cloud-based digital mine energy efficiency management equipment as described in claim 5, characterized in that: The process of generating a control decision report by comparing energy efficiency response indicators before and after the implementation of the collaborative operation strategy involves the following steps: The real-time monitoring energy efficiency response index after the execution of the collaborative operation strategy refers to the energy efficiency response index calculated based on the equipment operation data after the execution of the collaborative operation strategy; comparing the energy efficiency response index before and after the execution of the collaborative operation strategy, the change in energy efficiency is calculated. The effectiveness of collaborative operation strategies is assessed based on changes in energy efficiency, and optimization suggestions are generated based on these results. A management decision report is generated by integrating changes in energy efficiency, the effectiveness of the strategies, and optimization suggestions through a pre-set report template.

7. A cloud-based digital mine energy efficiency management method, based on the cloud-based digital mine energy efficiency management equipment according to any one of claims 1 to 6, characterized in that: include, Real-time acquisition and cleaning of multi-source heterogeneous operation data, which is then input into the mine energy efficiency model. The mine energy efficiency model maintains synchronous mapping with the physical entity through the Kalman filter algorithm, generating a virtual mine real-time status package. The industrial cloud platform uses the real-time status package of the virtual mine as the simulation benchmark, injects preset equipment failures and load mutations into the mine energy efficiency model to conduct multi-dimensional emergency simulation tests, and generates a dynamic emergency plan library. The multi-dimensional feature comparison algorithm is used to match multi-source heterogeneous operational data with a dynamic emergency plan database in real time. Once the matching is completed, the corresponding structured plan is output. The industrial cloud platform parses the corresponding structured plans into a distributed control instruction set, and uses the distributed control instruction set to drive a multi-agent coordination mechanism to generate collaborative operation strategies. The equipment controller executes the collaborative operation strategy and monitors the energy efficiency response indicators after the execution of the collaborative operation strategy in real time; by comparing the energy efficiency response indicators before and after the execution of the collaborative operation strategy, a control decision report is generated.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the cloud-based digital mine energy efficiency management method according to any one of claims 7.

Images