Digital data management system based on artificial intelligence

Abstract

The invention relates to a digital data management system based on artificial intelligence, belongs to the field of data management, and aims to solve the problems that an existing digital data management system based on artificial intelligence cannot accurately identify high-energy-consumption links or abnormal energy consumption behaviors, cannot predict future load demands, and cannot accurately identify the high-energy-consumption links or abnormal energy consumption behaviors. The system comprises a data collection and integration module, a data storage and processing module, an analysis and optimization module, an anomaly detection module and a visualization module, and the output end of the data collection and integration module is connected with the input end of the data storage and processing module. By accurately predicting the load demand and combining with the optimization algorithm, efficient scheduling and management of the power system can be realized, so that the operation efficiency of the system can be improved, the energy cost can be reduced and sustainable development can be promoted on the premise of ensuring the safety of energy supply.

Description

Technical Field

[0001] The present invention relates to a digital data management system based on artificial intelligence and belongs to the field of data management. Background Art

[0002] The digital energy data management system is an energy management solution based on information technology, aiming to achieve comprehensive collection, processing, analysis, and visualization of energy data to improve energy management efficiency, optimize energy use, and promote green and low-carbon development. This system can help enterprises, institutions, and government departments better monitor, analyze, and control the production, consumption, and distribution of energy, thereby reducing costs, improving efficiency, reducing waste, and promoting sustainable development.

[0003] However, there are some problems in the existing digital energy data management system during its operation. For example, a digital energy data management system and method based on artificial intelligence with the publication number CN116339204B can, when the actual energy consumption does not exceed the threshold, automatically make more detailed analysis and processing response measures for the actual energy consumption situation. However, it cannot accurately identify high-energy-consuming links or abnormal energy consumption behaviors, and at the same time, it cannot predict future load demands, so it cannot reasonably allocate or dispatch power resources to minimize costs. Therefore, we make improvements and propose a digital data management system based on artificial intelligence. Summary of the Invention

[0004] (1) The technical problem to be solved by the present invention is that the existing digital data management system based on artificial intelligence cannot accurately identify high-energy-consuming links or abnormal energy consumption behaviors, and at the same time, it cannot predict future load demands, so it cannot reasonably allocate or dispatch power resources to minimize costs.

[0005] (2) Technical Solution To achieve the above invention object, the present invention provides a digital data management system based on artificial intelligence, including a data collection and integration module, a data storage and processing module, an analysis and optimization module, an anomaly detection module, and a visualization module. The output end of the data collection and integration module is connected to the input end of the data storage and processing module, the output end of the data storage and processing module is connected to the input end of the analysis and optimization module, the output end of the analysis and optimization module is connected to the input end of the anomaly detection module, and the output end of the anomaly detection module is connected to the input end of the visualization module; The data collection and integration module is used to collect real-time energy data, and at the same time, unify and integrate the data from different energy devices, sensors, and monitoring systems and transmit them to the data storage and processing module; the analysis and optimization module includes an energy consumption analysis module, a load forecasting module, and a load optimization module. The energy consumption analysis module is used to analyze the energy consumption patterns of different time periods, different devices, or regions, calculate and identify high-energy-consuming links or abnormal energy-using behaviors. The load forecasting module is used to calculate and predict future load demands based on historical load data. The load optimization module is used to reasonably allocate or schedule power resources and calculate the minimum cost. The energy consumption analysis module includes: According to the formula: Where: x is the current data point, μ is the mean of the data, and σ is the standard deviation of the data set; If |Z| > z threshold , then the data point x is considered abnormal.

[0006] The load forecasting module includes: According to the formula: Y t = c + ϕ 1 Y t-1 + ϕ 2 Y t-2 +…+ ϕ p Y t-p + θ 1 e t-1 + θ 2 e t-2+ …+ θ q e t-q + e t Where: Y t is the actual value (load data) of the time series; ϕ 1 , ϕ 2 ,…, ϕ p are autoregressive (AR) coefficients; θ 1 , θ 2 ,…, θ q are moving average (MA) coefficients; e t is the white noise error term; c is the constant term.

[0007] The load optimization module includes: According to the formula: Wherein: is the total cost; is the load output of the i-th generator; is the unit cost of the i-th generator.

[0008] The load optimization module further includes: According to the formula: Wherein: is time the load demand of.

[0009] Wherein, the data collection and integration module includes an intelligent metering module, a multi-source data integration module, and a data transmission module. The output end of the intelligent metering module is connected to the input end of the multi-source data integration module, and the output end of the multi-source data integration module is connected to the input end of the data transmission module.

[0010] Wherein, the intelligent metering module collects real-time energy data through intelligent meters. The multi-source data integration module unifies and integrates data from different energy devices, sensors, and monitoring systems to form a complete energy data stream (for example, load data of electricity, wind energy or solar power generation data, equipment operation status, etc.), and transmits the data in real time through Internet of Things technology.

[0011] Wherein, the data storage and processing module includes a distributed storage module, a cloud database, and a data privacy protection module. The distributed storage module can store a large amount of energy data through distributed storage technology and supports fast retrieval and access. The cloud database uses big data aggregation to unify and classify and store the scattered and fragmented data.

[0012] Wherein, the load optimization module includes an engine output limit constraint condition: According to the formula: P imin ≤P i ≤P imax Wherein: P imin and P imax are respectively the minimum and maximum output powers of the i-th generator.

[0013] Among them, the anomaly detection module includes a real-time data monitoring module, a fault diagnosis module, and an alarm module. The output end of the real-time data monitoring module is connected to the input end of the fault diagnosis module, and the output end of the fault diagnosis module is connected to the input end of the alarm module.

[0014] Among them, the real-time data monitoring module monitors the data calculation results of the analysis and optimization module in real time. Combining with the fault diagnosis module, it can automatically analyze the status of energy equipment, identify potential faults or performance degradation in advance, so as to perform maintenance or scheduling in a timely manner. And after detecting an abnormal energy usage trend, it automatically issues an alarm through the alarm module and provides an adjustment plan.

[0015] Among them, the visualization module visualizes complex energy data through charts and dashboards, helps decision-makers understand the energy flow and usage patterns, and generates energy management reports, including energy consumption, cost analysis, efficiency improvement suggestions, and environmental protection compliance content, to assist management decision-making.

[0016] The fault diagnosis module uses a deep learning model to predict the types and times of faults that occur in energy equipment. The deep learning model adopts a long short-term memory network, and its specific implementation steps are as follows: Data collection and annotation: Collect historical operation data covering various energy equipment, including electrical parameters, temperature data, vibration data, and operation time-related information of the equipment. For each set of data, perform annotation to clarify its corresponding fault type and the time point when the fault occurs, and construct a training data set and a test data set, where the proportion of the training data set is not less than 70%, and the proportion of the test data set is not higher than 30%; Data preprocessing: Thoroughly clean the collected data, remove obvious error and missing data points, use mean filling and interpolation methods to supplement a small amount of missing data, normalize all data, map data with different physical dimensions to a specific interval to improve the model training effect. For time series data, resample it at a fixed time interval to ensure the consistency of data in the time dimension, and divide the processed data into sample sequences of a fixed length according to time series characteristics. Each sample sequence corresponds to a device operation status label, and construct a three-dimensional data format for input to the LSTM model.

[0017] The visualization module adopts interactive data visualization technology, including a Web-based visualization library, and its specific implementation method is as follows: Visualization interface design: Build a user-friendly Web interface, adopt a responsive layout design, so that it can adapt to the screen sizes of different devices. The interface mainly includes a data display area, an operation interaction area, and an information prompt area.

[0018] (III) Beneficial effects The beneficial effects of a digital data management system based on artificial intelligence provided by the present invention are as follows: Through the data collection and integration module, data from different energy devices, sensors, and monitoring systems are uniformly collected and integrated, and then transmitted to the data storage and processing module. Combining with the energy consumption analysis module, load forecasting module, and load optimization module in the analysis and optimization module, calculations for identifying high-energy-consuming links or abnormal energy consumption behaviors, predicting future load demands, and calculating how to reasonably allocate or schedule power resources to minimize costs are respectively carried out. By accurately predicting load demands and combining optimization algorithms, efficient scheduling and management of the power system can be achieved. Furthermore, on the premise of ensuring energy supply security, the operating efficiency of the system can be improved, energy costs can be reduced, sustainable development can be promoted, more efficient resource utilization and operation optimization can be realized, which can not only save energy costs for enterprises or governments, but also support the promotion of green energy and the reduction of carbon emissions, and promote the digitalization and intelligentization process of energy management. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 It is a schematic diagram of the overall system of the present invention; Figure 2 It is a schematic diagram of the data collection and integration module of the present invention; Figure 3 It is a schematic diagram of the data storage and processing module of the present invention; Figure 4 It is a schematic diagram of the analysis and optimization module of the present invention; Figure 5 It is a schematic diagram of the anomaly detection module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The following will further describe in detail the specific embodiments of the present invention in conjunction with the drawings in the specification and the embodiments. The following embodiments are only used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0022] Embodiment 1: As Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 and Figure 5As shown in the figure, this embodiment proposes a digital data management system based on artificial intelligence, including a data collection and integration module, a data storage and processing module, an analysis and optimization module, an anomaly detection module, and a visualization module. The output end of the data collection and integration module is connected to the input end of the data storage and processing module. The output end of the data storage and processing module is connected to the input end of the analysis and optimization module. The output end of the analysis and optimization module is connected to the input end of the anomaly detection module. The output end of the anomaly detection module is connected to the input end of the visualization module; The data collection and integration module is used to collect real-time energy data, and at the same time, unify and integrate the data from different energy devices, sensors, and monitoring systems and transmit them to the data storage and processing module; the analysis and optimization module includes an energy consumption analysis module, a load prediction module, and a load optimization module. The energy consumption analysis module is used to analyze the energy consumption patterns of different time periods, different devices, or regions, calculate and identify high-energy-consuming links or abnormal energy consumption behaviors. The load prediction module is used to calculate and predict future load demands based on historical load data. The load optimization module is used to reasonably allocate or schedule power resources and calculate the minimum cost; The energy consumption analysis module includes: According to the formula: Where: x is the current data point, μ is the mean of the data, and σ is the standard deviation of the data set.

[0023] If |Z| > z threshold , then the data point x is considered abnormal.

[0024] The load prediction module includes: According to the formula: Y t =c + ϕ 1 Y t-1 + ϕ 2 Y t-2 +…+ ϕ p Y t-p + θ 1 e t-1 + θ 2 e t-2+ …+ θ q e t-q + e t Where: Y t is the actual value (load data) of the time series; ϕ 1 , ϕ 2 ,…, ϕ p are autoregressive (AR) coefficients; θ 1 , θ 2 , …, θ q are moving average (MA) coefficients; e t is a white noise error term; c is a constant term.

[0025] The load optimization module includes: According to the formula: Where: is the total cost; is the load output of the i-th generator; is the unit cost of the i-th generator.

[0026] The load optimization module further includes: According to the formula: Where: is time the load demand at.

[0027] The data collection and integration module uniformly collects and integrates the data of different energy devices, sensors, and monitoring systems, and transmits it to the data storage and processing module. Combining with the energy consumption analysis module, load prediction module, and load optimization module in the analysis and optimization module, the identification and calculation of high-energy consumption links or abnormal energy consumption behaviors, the prediction calculation of future load demands, and the calculation of how to reasonably allocate or schedule power resources to minimize costs are respectively carried out. By accurately predicting the load demand and combining with optimization algorithms, the efficient scheduling and management of the power system can be achieved. Furthermore, on the premise of ensuring the security of energy supply, the operating efficiency of the system can be improved, the energy cost can be reduced, and sustainable development can be promoted.

[0028] Example 2: The following further introduces the solution in Example 1 in combination with specific working methods, as detailed in the following description: Such as Figure 1 and Figure 2As shown, as a preferred embodiment, on the basis of the above method, further, the data collection and integration module includes an intelligent metering module, a multi-source data integration module, and a data transmission module. The output end of the intelligent metering module is connected to the input end of the multi-source data integration module, and the output end of the multi-source data integration module is connected to the input end of the data transmission module. The intelligent metering module collects real-time energy data through intelligent meters (such as intelligent electricity meters, intelligent gas meters, intelligent water meters). The multi-source data integration module unifies and integrates data from different energy devices, sensors, and monitoring systems to form a complete energy data stream, such as power load data, wind or solar power generation data, equipment operating status, etc., and transmits the data in real time through Internet of Things technology. The data collection and integration module integrates data from different sources. Since different data sources may adopt different formats and structures, the integration module can perform appropriate data conversion and standardization to ensure that the data can be uniformly summarized into an available structure for subsequent analysis, processing, and use.

[0029] As Figure 1 and Figure 3 As shown, as a preferred embodiment, on the basis of the above method, further, the data storage and processing module includes a distributed storage module, a cloud database, and a data privacy protection module; the distributed storage module can store a large amount of energy data through distributed storage technology and supports fast retrieval and access, such as HDFS (Hadoop Distributed File System). In this way, the data is split and stored on multiple machines, increasing the availability and fault tolerance of the data; the cloud database uses big data aggregation to unify and classify and store the scattered and fragmented data, such as Amazon S3, Google Cloud Storage, etc. The data is stored on remote servers and supports high availability and elastic expansion. The data privacy protection module can ensure that the data is protected during storage and transmission to prevent unauthorized access.

[0030] As Figure 1 and Figure 4 As shown, as a preferred embodiment, on the basis of the above method, further, the load optimization module includes an engine output limit constraint condition: According to the formula: P imin ≤P i ≤P imax Where: P imin and P imax are the minimum and maximum output powers of the i-th generator, respectively.

[0031] As Figure 1 andFigure 5 As shown, as a preferred implementation, on the basis of the above method, further, the anomaly detection module includes a real-time data monitoring module, a fault diagnosis module, and an alarm module. The output end of the real-time data monitoring module is connected to the input end of the fault diagnosis module, and the output end of the fault diagnosis module is connected to the input end of the alarm module. The real-time data monitoring module monitors and analyzes the data calculation results of the analysis and optimization module in real time. Combining with the fault diagnosis module, it can automatically analyze the status of energy equipment, identify potential faults or performance degradation in advance, so as to perform maintenance or scheduling in a timely manner. And after detecting an abnormal energy usage trend, it automatically issues an alarm through the alarm module and provides an adjustment plan. Moreover, key information such as energy consumption, equipment operation status, and energy cost can be displayed in real time through the dashboard.

[0032] As Figure 1 shown, as a preferred implementation, on the basis of the above method, further, the visualization module visualizes complex energy data through charts and dashboards, helps decision-makers understand the energy flow and usage patterns, and generates energy management reports, including energy consumption, cost analysis, efficiency improvement suggestions, and environmental compliance content, to assist management decision-making. For example, it converts complex data and information into easy-to-understand visual content in the form of graphs, charts, maps, dashboards, etc. For example, line charts are used to display trend changes, and heat maps are used to show density distributions, etc. This can not only help users grasp key information more quickly, but also reduce the cognitive burden in data processing, and can help reveal potential patterns, trends, or relationships in the data.

[0033] The fault diagnosis module uses a deep learning model to predict the possible fault types and times of energy equipment. The deep learning model uses a long short-term memory network (LSTM), and its specific implementation steps are as follows: Data collection and annotation: Collect historical operation data of various energy equipment (such as generators, transformers, motors) under different working conditions (normal operation, different degrees of fault states), including electrical parameters of the equipment (such as voltage, current, power factor), temperature data, vibration data, and operation time-related information. For each set of data, perform annotation to clarify its corresponding fault type (such as short circuit fault, overheat fault, mechanical wear fault) and the time point of fault occurrence, and construct a training data set and a test data set, where the proportion of the training data set is not less than 70%, and the proportion of the test data set is not higher than 30%; Data preprocessing: Thoroughly clean the collected data, remove obvious errors and missing data points, use mean filling and interpolation methods to supplement a small amount of missing data, normalize all data, and map data with different physical dimensions to a specific interval (such as [-1, 1] or [0, 1]) to improve the model training effect. For time series data, resample it at a fixed time interval (such as 5 minutes or 10 minutes) to ensure data consistency in the time dimension. Divide the processed data into sample sequences of a fixed length (such as 100 time steps) according to time series characteristics, and each sample sequence corresponds to a device operating status label (normal or specific fault type), constructing a three-dimensional data format (number of samples, time steps, feature dimensions) suitable for input to the LSTM model.

[0034] Model construction: Build an LSTM network structure, including an input layer, one or more LSTM hidden layers, and an output layer. The number of nodes in the input layer is determined according to the feature dimension of the data. The number of layers of the LSTM hidden layer is set between 2 and 4, and the number of nodes in each layer is adjusted according to the complexity of the dataset and the device type (such as 128 - 512 nodes), and tanh or sigmoid is used as the activation function. The number of nodes in the output layer is the same as the number of predicted fault types, and a softmax classifier is used to output the predicted probability distribution of each fault type. During the model construction process, regularization techniques (such as L1 or L2 regularization) can be introduced to prevent overfitting, and the regularization coefficient ranges from 0.001 to 0.1.

[0035] Model training: Input the sample sequences of the preprocessed training dataset into the LSTM model, and use the backpropagation algorithm to calculate the loss function to update the model parameters. The cross-entropy loss function is selected as the loss function, and its formula is , where is the number of samples, is the true label (using one-hot encoding to represent the fault type), is the probability distribution predicted by the model. During the training process, an adaptive learning rate adjustment strategy is adopted. The initial learning rate is set between 0.001 and 0.1, and the learning rate is dynamically adjusted according to the change of loss during the training process. When the loss does not decrease for several consecutive epochs (such as 5 - 10), the learning rate is appropriately reduced. At the same time, use the Adam optimizer or RMSProp optimizer to optimize the algorithm to accelerate the model convergence. The hyperparameters of the optimizer (such as , ) are set according to empirical values, and the training process continues until the accuracy of the model on the validation set reaches a preset threshold (such as above 90%) or the number of training rounds reaches a predetermined upper limit (such as 100 - 500 rounds).

[0036] Model Evaluation and Optimization: Use the test dataset to evaluate the performance of the trained model. Adopt multiple metrics such as accuracy, recall, and F1-score to comprehensively measure the prediction effect of the model. Optimize and adjust the model according to the evaluation results, such as increasing the amount of training data, adjusting the network structure (adding or reducing hidden layers, adjusting the number of nodes), and optimizing hyperparameters (learning rate, regularization coefficient). Then train and evaluate again until the model performance reaches a satisfactory level.

[0037] Fault Prediction: During actual operation, convert the real-time monitored energy device data into a sample sequence of fixed length according to the same data preprocessing method, and input it into the trained and evaluated and optimized LSTM model. The model outputs the probability distribution of the predicted fault types. Judge whether a fault is likely to occur and the fault type according to the preset probability threshold (such as between 0.7 - 0.9, determined according to the severity of different fault types and the tolerance of false alarms). If the predicted probability exceeds the threshold, issue a fault warning and display the fault prediction results and related data through the visualization module to provide a decision-making basis for maintenance personnel. The visualization module adopts interactive data visualization technology, such as web-based visualization libraries (such as D3.js or Echarts). The specific implementation method is as follows: Visualization Interface Design: Build a user-friendly web interface with a responsive layout design so that it can adapt to the screen sizes of different devices (such as computers, tablets, mobile phones, etc.). The interface mainly includes a data display area, an operation interaction area, and an information prompt area. The data display area is used to present visualization charts of various energy data (such as line charts, bar charts, pie charts, maps, etc.). The operation interaction area provides interactive function buttons such as mouse click, drag, zoom, filter, query, etc. The information prompt area is used to display real-time user operation feedback information, data details, as well as analysis results and suggestions.

[0038] Implementation of Data Interaction Function: The data interaction function is implemented by using the JavaScript programming language in combination with the selected visualization library (such as D3.js or Echarts). When the user clicks the mouse in the operation interaction area, for example, clicks on a data point in the chart, the system captures this operation through the event listening mechanism, triggers a data query request, obtains detailed data from the backend server according to the time point or data range corresponding to the click position, and displays the detailed information of this data point (such as specific value, data source, relevant device information, etc.) in the information prompt area. When the user performs a drag operation, such as dragging to select a specific time period on the time series chart, the system updates the chart in the data display area in real time to show the detailed energy data change trend during this time period, and at the same time displays the key statistical information (such as average energy consumption, maximum load, etc.) during this time period in the information prompt area. The zoom operation shows data views with different granularities by changing the time scale or data range of the chart, facilitating users to observe the macroscopic trends and microscopic details of the data. The filtering and query functions allow users to filter data according to specific conditions (such as device type, region, time range, etc.). The system queries the data in the backend database according to the conditions input by the user and updates the data display area to show the data visualization results that meet the conditions.

[0039] Real-time Data Update and Dynamic Display: By establishing a real-time communication mechanism such as Websocket or long polling to maintain a connection with the backend server, the latest energy data is obtained in real time. When new data arrives, the system automatically updates the chart in the data display area to dynamically display the real-time changes of the energy data, such as the update of the real-time energy consumption curve and the dynamic changes of the device operation status. At the same time, according to the data changes and preset rules, real-time data analysis results and suggestions are provided in the information prompt area. For example, when the energy consumption suddenly increases, possible abnormal situations and troubleshooting directions are prompted.

[0040] Visualization Customization and Personalization: Provide visualization customization functions. Users can choose different chart types, color themes, data display dimensions, etc. according to their own needs. The system dynamically generates corresponding visualization configuration files in the backend according to the user's customization selection and passes them to the front-end visualization library for rendering and display to achieve a personalized data visualization experience.

[0041] The above embodiments are only used to illustrate the present invention, rather than to limit the present invention. Although the present invention has been described in detail with reference to the embodiments, those of ordinary skill in the art should understand that various combinations, modifications or equivalent replacements of the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention and should all be covered within the scope of the claims of the present invention.

Claims

1. A digital data management system based on artificial intelligence, comprising a data collection and integration module, a data storage and processing module, an analysis and optimization module, an anomaly detection module and a visualization module, characterized in that: The output end of the data collection integration module is connected to the input end of the data storage processing module, the output end of the data storage processing module is connected to the input end of the analysis and optimization module, the output end of the analysis and optimization module is connected to the input end of the anomaly detection module, and the output end of the anomaly detection module is connected to the input end of the visualization module; The data collection and integration module is used to collect real-time energy data, and at the same time, integrate the data from different energy devices, sensors and monitoring systems and transmit them to the data storage and processing module; the analysis and optimization module includes an energy consumption analysis module, a load forecasting module and a load optimization module. The energy consumption analysis module is used to analyze the energy consumption patterns of different time periods, different devices or regions, calculate and identify high energy consumption links or abnormal energy consumption behaviors, the load forecasting module is used to calculate and forecast future load demand based on historical load data, and the load optimization module is used to reasonably allocate or dispatch power resources and calculate the minimum cost; The energy consumption analysis module includes: According to the formula: in: x is the current data point, μ is the mean of the data, and σ is the standard deviation of the data set. If |Z|>z threshold , then the data point x is considered to be abnormal. The load forecasting module comprises: According to the formula: Y t =c+ϕ1Y t-1 +ϕ2Y t-2 +…+ϕ p Y t-p +θ1e t-1 +θ2e t-2+ …+θ q e t-q +e t in: Y t is the actual value of the time series (load data); ϕ1, ϕ2, …, ϕ p is the autoregression (AR) coefficient; θ1,θ2,…,θ q is the moving average (MA) coefficient; e t is the white noise error term; c is a constant term. The load optimization module includes: According to the formula: in: is the total cost; is the load output of the i-th generator; is the unit cost of the i-th generator. The load optimization module also includes: According to the formula: in: For time load demand.

2. According to claim 1, a digital data management system based on artificial intelligence is characterized in that: The data collection integration module includes an intelligent metering module, a multi-source data integration module and a data transmission module. The output end of the intelligent metering module is connected to the input end of the multi-source data integration module, and the output end of the multi-source data integration module is connected to the input end of the data transmission module.

3. The digital data management system based on artificial intelligence according to claim 2 is characterized in that: The smart metering module collects real-time energy data through smart meters, and the multi-source data integration module integrates data from different energy devices, sensors and monitoring systems to form a complete energy data flow, and transmits the data in real time through the Internet of Things technology.

4. The digital data management system based on artificial intelligence according to claim 3 is characterized in that: The data storage and processing module includes a distributed storage module, a cloud database and a data privacy protection module. The distributed storage module can store massive amounts of energy data through distributed storage technology and support fast retrieval and access. The cloud database uses big data aggregation to uniformly aggregate and collect scattered and fragmented data and store them in a classified manner.

5. The digital data management system based on artificial intelligence according to claim 4 is characterized in that: The load optimization module includes engine output limitation constraints: According to the formula: P imin ≤P i ≤P imax in: P imin and P imax are the minimum and maximum output powers of the ith generator, respectively.

6. The digital data management system based on artificial intelligence according to claim 5 is characterized in that: The abnormality detection module includes a real-time data monitoring module, a fault diagnosis module and an alarm module. The output end of the real-time data monitoring module is connected to the input end of the fault diagnosis module, and the output end of the fault diagnosis module is connected to the input end of the alarm module.

7. The digital data management system based on artificial intelligence according to claim 6 is characterized in that: The real-time data monitoring module monitors the data calculation results of the analysis and optimization module in real time, and combined with the fault diagnosis module, it can automatically analyze the status of energy equipment, identify potential faults or performance degradation in advance, so as to perform maintenance or scheduling in time, and after discovering abnormal energy usage trends, automatically issue an alarm through the alarm module and provide adjustment plans.

8. The digital data management system based on artificial intelligence according to claim 1 is characterized in that: The visualization module visualizes complex energy data through charts and dashboards to help decision makers understand energy flows and usage patterns, and generates energy management reports, including energy consumption, cost analysis, efficiency improvement suggestions, and environmental compliance content, to assist management in decision-making.

9. The digital data management system based on artificial intelligence according to claim 1, characterized in that: The fault diagnosis module uses a deep learning model to predict the type and time of faults occurring in energy equipment. The deep learning model uses a long short-term memory network. The specific implementation steps are as follows: Data collection and labeling: Collect historical operating data covering various energy devices, including electrical parameters, temperature data, vibration data, and operating time related information of the equipment. Label each set of data to identify the corresponding fault type and the time when the fault occurred, and construct training data sets and test data sets, of which the training data set accounts for no less than 70% and the test data set accounts for no more than 30%; Data preprocessing: The collected data is cleaned comprehensively to remove data points with obvious errors and missing data points. A small amount of missing data is supplemented by mean filling and interpolation. All data are normalized and data of different physical dimensions are mapped to specific intervals to improve the model training effect. For time series data, resample at fixed time intervals to ensure the consistency of data in the time dimension. The processed data is divided into sample sequences of fixed length according to the time series characteristics. Each sample sequence corresponds to a device operation status label, which is constructed into a three-dimensional data format for LSTM model input.

10. The digital data management system based on artificial intelligence according to claim 1, characterized in that: The visualization module adopts interactive data visualization technology, including a Web-based visualization library, and its specific implementation is as follows: Visual interface design: Build a user-friendly web interface and adopt a responsive layout design so that it can adapt to the screen sizes of different devices. The interface mainly includes a data display area, an operation interaction area, and an information prompt area.

Images