A smart grid monitoring and alarming method and system

Abstract

The present application relates to the technical field of smart grid, and particularly relates to a smart grid monitoring and alarming method and system. The present application is not based on traditional image analysis technology, but based on real-time operation data of the power grid, including current, voltage, temperature and humidity data for fault prediction of power grid equipment. The prediction mode combining ANN prediction model and LSTM prediction model can extract static features of collected data, correlate to time series data, retain time dimension features, and improve prediction accuracy. Moreover, based on the predicted data and equipment state, the health degree of the power grid in the future period of time can be evaluated, and alarm information can be sent to relevant personnel, which can give relevant personnel early warning of equipment failure at the next moment and the state of the power grid in the future period of time, achieving the effect of early prediction.

Description

A Smart Grid Monitoring and Alarm Method and System Technical Field

[0001] This invention relates to the field of smart grid technology, and in particular to a smart grid monitoring and alarm method and system. Background Technology

[0002] With technological advancements and the accumulation of big data, AI-based smart grid fault detection and prediction methods have become a key technology for the safe and stable operation of power grids. AI possesses advantages such as real-time performance, high accuracy, big data processing capabilities, and comprehensive analysis and prediction, providing timely, efficient, and reliable fault detection and prediction solutions for smart grids. In today's era of rapid development in digitalization, automation, and internet technologies, the power system is undergoing a profound transformation, with traditional power grids gradually transitioning to smart grids, providing a more efficient, flexible, and reliable power supply. However, this also increases the complexity of the power grid, leading to more faults and unpredictable challenges. Therefore, real-time monitoring and prediction of the power grid are crucial to ensuring its stable and safe operation.

[0003] While existing technologies have made some progress in power grid fault detection and prediction, some significant shortcomings remain. Traditional inspection image analysis techniques typically employ classical image processing algorithms. The handcrafted features extracted by these algorithms are primarily low-level visual features. Compared to features extracted through pure learning methods, these features offer strong interpretability but weak adaptability to data. Furthermore, the power grid's operating environment is complex and variable, frequently encountering ambiguity and uncertainty in fault detection and prediction, and existing technologies have limited capabilities in handling this ambiguity and uncertainty. Summary of the Invention

[0004] To address the problems in existing technologies, this invention provides a smart grid monitoring and alarm method and system, the specific technical solution of which is as follows:

[0005] A smart grid monitoring and alarm method includes the following steps:

[0006] Step S1: Collect real-time operating data of the power grid, including current, voltage, temperature and humidity data;

[0007] Step S2: Process and analyze the collected data to obtain processed current, voltage, temperature and humidity data, as well as the trend data of current, voltage, temperature and humidity changes obtained from the analysis.

[0008] Step S3: Input the collected data into the trained ANN prediction model to obtain the first prediction result; the first prediction result includes the predicted current, voltage, temperature and humidity data for the next moment.

[0009] Step S4: Input the trend data of current, voltage, temperature and humidity into the trained LSTM prediction model to obtain the second prediction result; the second prediction result includes the predicted current, voltage, temperature and humidity data at the next moment, and the trend data of current, voltage, temperature and humidity starting from the next moment.

[0010] Step S5: Combine the first prediction result and the second prediction result to obtain the final current, voltage, temperature and humidity data for the next moment, whether the equipment is faulty at the next moment, and the trend data of current, voltage, temperature and humidity changes starting from the next moment.

[0011] Step S6: When a device failure is predicted in the next moment, an alarm message is sent to the relevant personnel.

[0012] Step S7: Accumulate the cumulative number of equipment failures based on the prediction information, and assess the health of the power grid and predict the cumulative number of failures in the future period based on the trend data of current, voltage, temperature and humidity changes starting from the next moment. When the health is less than the set threshold or the predicted cumulative number of failures reaches the set threshold, send alarm information to relevant personnel.

[0013] Preferably, the data processing in step S2 includes cleaning, outlier removal, missing value imputation, and standardization.

[0014] Preferably, the specific steps for training the ANN prediction model in step S3 are as follows:

[0015] Data acquisition and processing: Collect and process historical power grid operation data, and divide the historical operation data into training set, validation set and test set; the historical operation data includes current, voltage, temperature and humidity data and equipment failure status; initialization: establish an ANN prediction model, which includes an input layer, two hidden layers and an output layer, and initialize the weights and biases of each layer of the ANN prediction model with random numbers;

[0016] Forward propagation: Input the training set and validation set into the initialized ANN prediction model and calculate the output of each neuron;

[0017] Loss calculation: Calculate the value of the loss function to evaluate the difference between the model's predictions and the actual values;

[0018] Backpropagation: Calculate the gradient of the loss function with respect to each weight; update the weights and biases using gradient descent;

[0019] Iterative training: Repeat forward propagation, loss calculation and back propagation until the prediction accuracy of the ANN prediction model converges or reaches the preset number of training times, and obtain the trained ANN prediction model.

[0020] Model Evaluation: The trained ANN prediction model is evaluated using a test set. The prediction accuracy and evaluation metrics are calculated. When the prediction accuracy is greater than or equal to the set value and the evaluation metrics meet the set conditions, the trained ANN prediction model is output; otherwise, the model is retrained.

[0021] Preferably, the specific steps for training the LSTM prediction model in step S4 are as follows:

[0022] Data acquisition and processing: Historical power grid operation data is collected and processed for analysis. The historical operation data is divided into training set, validation set, and test set. The historical operation data includes current, voltage, temperature, and humidity data, as well as the status of equipment malfunctions. The analysis yields historical trend data on current, voltage, temperature, and humidity changes.

[0023] Initialization: Establish an LSTM prediction model, which includes an input layer, two LSTM layers, and an output layer. Use random numbers to initialize the weights and biases of each layer of the LSTM prediction model.

[0024] Forward propagation: Input the training set and validation set into the initialized LSTM prediction model and compute the output of each LSTM element;

[0025] Loss calculation: Calculate the value of the loss function to evaluate the difference between the model's predictions and the actual values;

[0026] Backpropagation: Calculate the gradient of the loss function with respect to each weight; update the weights and biases using gradient descent;

[0027] Iterative training: Repeat forward propagation, loss calculation and back propagation until the prediction accuracy of the LSTM prediction model converges or reaches the preset number of training times, and obtain the trained LSTM prediction model.

[0028] Model Evaluation: The trained LSTM prediction model is evaluated using a test set. The prediction accuracy and evaluation metrics are calculated. When the prediction accuracy is greater than or equal to the set value and the evaluation metrics meet the set conditions, the trained LSTM prediction model is output; otherwise, it is retrained.

[0029] Preferably, step S5, which involves concatenating the first prediction result and the second prediction result, specifically includes the following steps:

[0030] Let the prediction result at time t+1 be X(t+1), then:

[0031] X(t+1)=ω1X1(t+1)+ω2X2(t+1);

[0032]

[0033] Where X1(t+1) is the predicted value of current / voltage / temperature / humidity in the first prediction result of the ANN prediction model; ω1 is the weight assigned to X1(t+1); X2(t+1) is the predicted value of current / voltage / temperature / humidity in the second prediction result of the LSTM prediction model; ω2 is the weight assigned to X2(t+1); D1 is the prediction accuracy of the ANN prediction model; and D2 is the prediction accuracy of the LSTM prediction model.

[0034] Preferably, the calculation method for the health of the power grid in step S7 is as follows:

[0035]

[0036] Where J represents the health of the power grid; Z i Let be the predicted state variable for device i, which is a 0-1 variable, set to 0 when the device fails and 1 otherwise; n is the number of devices in the power grid.

[0037] Preferably, the method further includes step S8, which involves visually displaying on a large screen whether the equipment is faulty at the next moment, the trend data of changes in current, voltage, temperature and humidity starting at the next moment, the health of the power grid over a future period, and the predicted cumulative number of faults.

[0038] A smart grid monitoring and alarm system, applied to the method described, includes:

[0039] The data acquisition module is used to collect real-time operating data of the power grid, including current, voltage, temperature and humidity data;

[0040] The data processing and analysis module is used to process and analyze the collected data to obtain processed current, voltage, temperature and humidity data, as well as the trend data of current, voltage, temperature and humidity changes obtained from the analysis.

[0041] The first prediction module is used to input the collected data into the trained ANN prediction model to obtain the first prediction result; the first prediction result includes the predicted current, voltage, temperature and humidity data for the next moment.

[0042] The second prediction module is used to input the trend data of current, voltage, temperature and humidity into the trained LSTM prediction model to obtain the second prediction result; the second prediction result includes the predicted current, voltage, temperature and humidity data at the next moment, as well as the trend data of current, voltage, temperature and humidity starting from the next moment.

[0043] The result stitching module is used to stitch together the first prediction result and the second prediction result to obtain the final current, voltage, temperature and humidity data for the next moment, whether the equipment is faulty at the next moment, and the trend data of current, voltage, temperature and humidity changes starting from the next moment.

[0044] The predictive alarm module is used to send alarm information to relevant personnel when a device failure is predicted to occur in the next moment.

[0045] The assessment and alarm module is used to accumulate the cumulative number of equipment failures based on the prediction information, and to assess the health of the power grid and predict the cumulative number of failures over a future period based on the changing trends of current, voltage, temperature and humidity starting from the next moment. When the health is less than a set threshold or the predicted cumulative number of failures reaches the set threshold, an alarm message is sent to the relevant personnel.

[0046] A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to execute the smart grid monitoring and alarm method.

[0047] A processor for running a program, wherein the program executes the smart grid monitoring and alarm method during runtime.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] This invention provides a smart grid monitoring and alarm method and system. Instead of employing traditional image analysis techniques, it predicts grid equipment faults based on real-time grid operation data, including current, voltage, temperature, and humidity data. The method combines ANN and LSTM prediction models, extracting static features from the collected data while also correlating it with time-series data, preserving time-dimensional features and improving prediction accuracy. Furthermore, based on the predicted data and equipment status, the system can assess the grid's health over a future period and send alarm messages to relevant personnel, providing early warnings of potential equipment failures and the grid's future condition, achieving the effect of early prediction. Attached Figure Description

[0050] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0051] Figure 1 is a flowchart of the method of the present invention.

[0052] Figure 2 is a system schematic diagram of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0055] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0056] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0057] Example 1:

[0058] As shown in Figure 1, this embodiment provides a smart grid monitoring and alarm method, including the following steps:

[0059] Step S1 involves collecting real-time operational data of the power grid, including current, voltage, temperature, and humidity data. Specifically: Voltage (V): Collects voltage values ​​at each node of the power grid to monitor voltage fluctuations and anomalies; Current (A): Collects current data to monitor current anomalies and load changes; Temperature (°C): Collects the temperature of key equipment (such as transformers and distribution cabinets) to monitor overheating; Humidity (%): Collects ambient humidity data to monitor the impact of humidity on equipment operation. Data acquisition is achieved by installing sensors on the equipment to continuously collect data. Different sensors are required for different data types, including temperature and humidity sensors, current transformers, and voltage transformers. The data acquisition equipment should possess high accuracy, high reliability, and real-time transmission capabilities.

[0060] Step S2 involves processing and analyzing the collected data to obtain processed current, voltage, temperature, and humidity data, as well as the trend data of current, voltage, temperature, and humidity changes obtained from the analysis.

[0061] Data processing includes cleaning, outlier removal, missing value imputation, and standardization.

[0062] The cleaning process includes removing noisy data.

[0063] Outlier removal involves eliminating abnormal values ​​(such as abnormal readings caused by sensor malfunctions), specifically using statistical methods (such as...).

[0064] The 3σ principle or machine learning methods are used to detect and handle outliers.

[0065] Missing values ​​are filled using methods such as interpolation or mean filling.

[0066] Standardization scales data to the range [0,1] or [-1,1] to accelerate model training and improve convergence speed. Specific applications...

[0067] Min-Max normalization or Z-score standardization methods.

[0068] Step S3: Input the collected data into the trained ANN prediction model to obtain the first prediction result; the first prediction result includes the predicted current, voltage, temperature and humidity data for the next moment.

[0069] The specific steps for training an ANN prediction model are as follows:

[0070] (1) Data acquisition and processing: Collect and process historical operating data of the power grid, and divide the historical operating data into training set, verification set and test set; the historical operating data includes current, voltage, temperature and humidity data and whether the equipment is faulty.

[0071] (2) Initialization: Establish an ANN prediction model, which includes an input layer, two hidden layers and an output layer. Use random numbers to initialize the weights and biases of each layer of the ANN prediction model. The hidden layers use the tanh activation function and the output layer uses the linear activation function.

[0072] (3) Forward propagation: Input the training set and validation set into the initialized ANN prediction model and calculate the output of each neuron.

[0073] (4) Loss calculation: Calculate the value of the loss function and evaluate the difference between the model prediction results and the actual values.

[0074] (5) Backpropagation: Calculate the gradient of the loss function with respect to each weight; update the weights and biases using gradient descent.

[0075] (6) Iterative training: Repeat forward propagation, loss calculation and back propagation until the prediction accuracy of the ANN prediction model converges or reaches the preset number of training times, and obtain the trained ANN prediction model.

[0076] (7) Model evaluation: The trained ANN prediction model is evaluated using the test set. The prediction accuracy and evaluation index are calculated. When the prediction accuracy is greater than or equal to the set value and the evaluation index meets the set conditions, the trained ANN prediction model is output. Otherwise, the hyperparameters of the model are adjusted, such as the learning rate, batch size, number of network layers and number of units, and the model is retrained.

[0077] Evaluation metrics include mean squared error (MSE) and mean absolute error (MAE). These metrics assess the model's accuracy and robustness, ensuring its reliability in practical applications.

[0078] Step S4: Input the trend data of current, voltage, temperature and humidity into the trained LSTM prediction model to obtain the second prediction result; the second prediction result includes the predicted current, voltage, temperature and humidity data at the next moment, and the trend data of current, voltage, temperature and humidity starting from the next moment.

[0079] The specific steps for training an LSTM prediction model are as follows:

[0080] (1) Data acquisition and processing: Collect historical operating data of the power grid and process and analyze it. Divide the historical operating data into training set, validation set and test set. The historical operating data includes current, voltage, temperature and humidity data and whether the equipment is faulty. Analyze the historical current, voltage, temperature and humidity change trend data. Specifically, the sliding window method can be used to generate time series data so that the LSTM model can make time series predictions.

[0081] (2) Initialization: Establish an LSTM prediction model, which includes an input layer, two LSTM layers and an output layer. Use random numbers to initialize the weights and biases of each layer of the LSTM prediction model. Use the tanh activation function for the LSTM layers.

[0082] (3) Forward propagation: Input the training set and validation set into the initialized LSTM prediction model and calculate the output of each LSTM element.

[0083] (4) Loss calculation: Calculate the value of the loss function and evaluate the difference between the model's prediction results and the actual values;

[0084] Backpropagation: Calculate the gradient of the loss function with respect to each weight; update the weights and biases using gradient descent;

[0085] (5) Iterative training: Repeat forward propagation, loss calculation and back propagation until the prediction accuracy of the LSTM prediction model converges or reaches the preset number of training times, and obtain the trained LSTM prediction model.

[0086] (6) Model evaluation: The trained LSTM prediction model is evaluated using the test set. The prediction accuracy and evaluation index are calculated. When the prediction accuracy is greater than or equal to the set value and the evaluation index meets the set conditions, the trained LSTM prediction model is output. Otherwise, the hyperparameters of the model are adjusted, such as the learning rate, batch size, number of network layers and number of units, and the model is retrained.

[0087] ANNs are used to process static features (such as current voltage, current, temperature, and humidity). ANNs can learn nonlinear relationships and extract high-level feature representations. LSTMs are used to process time-series data (such as historical trends in voltage, current, temperature, and humidity). LSTMs can capture long-term dependencies in time series data and are suitable for predicting future states. The outputs of ANNs and LSTMs can be fused (e.g., through weighted fusion or concatenation) to comprehensively consider both static and time-series features.

[0088] Step S5: Combine the first prediction result and the second prediction result to obtain the final current, voltage, temperature and humidity data for the next moment, whether the equipment is faulty at the next moment, and the trend data of current, voltage, temperature and humidity changes starting from the next moment.

[0089] The specific steps for combining the first and second prediction results are as follows:

[0090] Let the prediction result at time t+1 be X(t+1), then:

[0091] X(t+1)=ω1X1(t+1)+ω2X2(t+1);

[0092]

[0093] Where X1(t+1) represents the predicted values ​​of current / voltage / temperature / humidity in the first prediction result of the ANN prediction model; ω1 represents the weighting of X1(t+1); X2(t+1) represents the predicted values ​​of current / voltage / temperature / humidity in the second prediction result of the LSTM prediction model; ω2 represents the weighting of X2(t+1); D1 represents the prediction accuracy of the ANN prediction model; and D2 represents the prediction accuracy of the LSTM prediction model. In this invention, the higher the prediction accuracy, the larger the proportion of its prediction results.

[0094] Step S6: When a device failure is predicted in the next moment, an alarm message is sent to the relevant personnel.

[0095] Step S7: Accumulate the cumulative number of equipment failures based on the prediction information, and assess the health of the power grid and predict the cumulative number of failures in the future period based on the trend data of current, voltage, temperature and humidity changes starting from the next moment. When the health is less than the set threshold or the predicted cumulative number of failures reaches the set threshold, send alarm information to relevant personnel.

[0096] The health status of the power grid is calculated as follows:

[0097]

[0098] Where J represents the health of the power grid; Z i Let be the predicted state variable for device i, which is a 0-1 variable, set to 0 when the device fails and 1 otherwise; n is the number of devices in the power grid.

[0099] Step S8 involves visually displaying on a large screen whether the equipment will malfunction at the next moment, the changing trends of current, voltage, temperature, and humidity starting from the next moment, the health of the power grid over a future period, and the predicted cumulative number of faults. Specifically, data visualization techniques can be used, such as charts and heat maps.

[0100] This invention specifically provides a graphical interface that displays the power grid's operating status and forecast results. It supports the visualization of real-time and historical data.

[0101] This invention has the following characteristics:

[0102] High-precision prediction: Combining the advantages of ANN and LSTM, it can more accurately predict changes in power grid status.

[0103] Real-time performance: Real-time data acquisition and analysis enable timely detection and handling of anomalies and faults.

[0104] This invention achieves comprehensive monitoring of the power grid's operational status and accurate prediction of future trends by combining real-time data acquisition and predictive models. First, by collecting key parameters such as current, voltage, temperature, and humidity in real time, it ensures a precise understanding of the current state of the power grid. The real-time nature of this data is crucial for rapid response to emergencies in the power grid. Second, the fault identification model trained using this real-time data can promptly identify abnormal states in the power grid, enabling fault prevention and rapid response. Furthermore, the data prediction model, based on historical and real-time data, predicts future power grid parameters and equipment status, allowing maintenance personnel to prepare in advance and take preventative measures, thereby reducing potential faults and losses. This combination of real-time and predictive capabilities not only improves the reliability of the power grid but also provides strong support for intelligent power grid management.

[0105] This invention combines ANN and LSTM models to learn complex patterns and correlations from large amounts of historical data, which are difficult to capture using traditional methods. Through continuous optimization of the training set and adjustment of model parameters, the model's accuracy can be continuously improved. Furthermore, real-time alarm mechanisms and the recording of accumulated fault counts provide immediate feedback for fault diagnosis, enabling maintenance personnel to respond quickly and shortening troubleshooting time. This automated and intelligent fault diagnosis process not only improves efficiency but also reduces the possibility of human error, ensuring the stable operation of the power grid.

[0106] This invention predicts potential faults within a future timeframe, allowing maintenance teams to proactively allocate technical personnel and resources for more efficient maintenance. For example, equipment predicted to be high-risk can be prioritized for inspection and maintenance, thus preventing or reducing the occurrence of faults. Furthermore, by comparing the cumulative number of faults with the actual number of faults, the accuracy of the fault identification model can be evaluated, allowing for model adjustments and optimization to improve predictive accuracy. This data-driven decision support system enables more scientific and rational resource allocation and more precise and effective maintenance strategies, thereby reducing maintenance costs and improving the operational efficiency of the power grid.

[0107] This invention enhances the adaptability and resilience of the power grid by real-time monitoring and prediction of its status. Faced with complex and ever-changing power grid operating environments, it can quickly adapt to various emergencies, such as weather changes and load fluctuations, and adjust power grid operating strategies in a timely manner to ensure stable power supply. The introduction of predictive models makes power grid management more proactive, rather than reactive to fault responses. By predicting future power grid conditions, measures can be taken in advance to avoid or mitigate the impact of faults, improving the power grid's resilience to various risks. Furthermore, the real-time alarms provided by this invention offer data support for power grid health management, helping to promptly identify and resolve potential problems, thereby extending equipment lifespan, reducing the occurrence of large-scale failures, and enhancing the overall resilience of the power grid.

[0108] Example 2:

[0109] As shown in Figure 2, based on the same inventive concept as in Embodiment 1, this embodiment provides a smart grid monitoring and alarm system applied to the method described, including:

[0110] The data acquisition module is used to collect real-time operating data of the power grid, including current, voltage, temperature and humidity data;

[0111] The data processing and analysis module is used to process and analyze the collected data to obtain processed current, voltage, temperature and humidity data, as well as the trend data of current, voltage, temperature and humidity changes obtained from the analysis.

[0112] The first prediction module is used to input the collected data into the trained ANN prediction model to obtain the first prediction result; the first prediction result includes the predicted current, voltage, temperature and humidity data for the next moment.

[0113] The second prediction module is used to input the trend data of current, voltage, temperature and humidity into the trained LSTM prediction model to obtain the second prediction result; the second prediction result includes the predicted current, voltage, temperature and humidity data at the next moment, as well as the trend data of current, voltage, temperature and humidity starting from the next moment.

[0114] The result stitching module is used to stitch together the first prediction result and the second prediction result to obtain the final current, voltage, temperature and humidity data for the next moment, whether the equipment is faulty at the next moment, and the trend data of current, voltage, temperature and humidity changes starting from the next moment.

[0115] The predictive alarm module is used to send alarm information to relevant personnel when a device failure is predicted to occur in the next moment.

[0116] The assessment and alarm module is used to accumulate the cumulative number of equipment failures based on the prediction information, and to assess the health of the power grid and predict the cumulative number of failures over a future period based on the changing trends of current, voltage, temperature and humidity starting from the next moment. When the health is less than a set threshold or the predicted cumulative number of failures reaches the set threshold, an alarm message is sent to the relevant personnel.

[0117] Example 3:

[0118] Based on the same inventive concept as Embodiment 1, this embodiment provides a computer-readable storage medium, which includes a stored program, wherein the program controls the device where the computer-readable storage medium is located to execute the smart grid monitoring and alarm method when it is running.

[0119] Example 4:

[0120] Based on the same inventive concept as Embodiment 1, this embodiment provides a processor for running a program, wherein the program executes the smart grid monitoring and alarm method during runtime.

[0121] Those skilled in the art will recognize that the modules of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the invention.

[0122] In the embodiments provided by this invention, it should be understood that the division of modules is only a logical functional division. In actual implementation, there may be other division methods, such as multiple modules can be combined into one module, one module can be split into multiple modules, or some features can be ignored.

[0123] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0124] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A smart grid monitoring and alarm method, characterized in that, Includes the following steps: Step S1: Collect real-time power grid operation data, including current, voltage, temperature, and humidity data; Step S2: Process and analyze the collected data to obtain processed current, voltage, temperature, and humidity data, as well as the analyzed trend data of current, voltage, temperature, and humidity; Step S3: Input the collected data into a trained ANN prediction model to obtain a first prediction result; the first prediction result includes the predicted current, voltage, temperature, and humidity data for the next moment; Step S4: Input the trend data of current, voltage, temperature, and humidity into a trained LSTM prediction model to obtain a second prediction result; the second prediction result includes the predicted current, voltage, temperature, and humidity data for the next moment, as well as the trend data of current, voltage, temperature, and humidity starting from the next moment. Step S5: The first and second prediction results are concatenated to obtain the final current, voltage, temperature, and humidity data for the next time step, whether the equipment is faulty at the next time step, and the trend data of current, voltage, temperature, and humidity changes starting from the next time step. Concatenating the first and second prediction results specifically includes the following steps: Let the prediction result at time t+1 be... ,but: ； ;in, These are the predicted values ​​of current / voltage / temperature / humidity in the first prediction result of the ANN prediction model; for The allocation weights; The predicted values ​​of current / voltage / temperature / humidity in the second prediction result of the LSTM prediction model; for The allocation weights; The prediction accuracy of the ANN prediction model; To improve the prediction accuracy of the LSTM prediction model; Step S6, when predicting equipment failure at the next moment, send alarm information to relevant personnel; Step S7, accumulate the cumulative number of equipment failures based on the prediction information, and evaluate the health of the power grid and the predicted cumulative number of failures in the future period based on the changing trend data of current, voltage, temperature and humidity starting from the next moment. When the health is less than the set threshold or the predicted cumulative number of failures reaches the set threshold, send alarm information to relevant personnel.

2. The smart grid monitoring and alarm method according to claim 1, characterized in that, The data processing in step S2 includes cleaning, outlier removal, missing value imputation, and standardization.

3. The smart grid monitoring and alarm method according to claim 1, characterized in that, The specific steps for training the ANN prediction model in step S3 are as follows: Data acquisition and processing: Collect and process historical operating data of the power grid, and divide the historical operating data into training set, validation set and test set; the historical operating data includes current, voltage, temperature and humidity data and the status of equipment failure; Initialization: Establish an ANN prediction model, which includes an input layer, two hidden layers, and an output layer. Initialize the weights and biases of each layer of the ANN prediction model using random numbers. Forward propagation: Input the training set and validation set into the initialized ANN prediction model and calculate the output of each neuron. Loss calculation: Calculate the value of the loss function and evaluate the difference between the model's prediction results and the true values. Backpropagation: Calculate the gradient of the loss function with respect to each weight; update the weights and biases using gradient descent; Iterative training: Repeat forward propagation, loss calculation, and backpropagation until the prediction accuracy of the ANN prediction model converges or the preset number of training iterations is reached, resulting in a trained ANN prediction model; Model evaluation: Evaluate the trained ANN prediction model using a test set, calculate the prediction accuracy and evaluation metrics. When the prediction accuracy is greater than or equal to a set value and the evaluation metrics meet the set conditions, output the trained ANN prediction model; otherwise, retrain.

4. The smart grid monitoring and alarm method according to claim 1, characterized in that, The specific steps for training the LSTM prediction model in step S4 are as follows: Data acquisition and processing: Collect historical operating data of the power grid and process and analyze it. Divide the historical operating data into training set, validation set and test set; the historical operating data includes current, voltage, temperature and humidity data and the status of equipment failure. Analysis yielded historical trend data on current, voltage, temperature, and humidity. Initialization: Establish an LSTM prediction model, which includes an input layer, two LSTM layers, and an output layer. Initialize the weights and biases of each layer of the LSTM prediction model using random numbers. Forward propagation: Input the training set and validation set into the initialized LSTM prediction model and calculate the output of each LSTM element. Loss calculation: Calculate the value of the loss function and evaluate the difference between the model's prediction results and the true values. Backpropagation: Calculate the gradient of the loss function with respect to each weight; update the weights and biases using gradient descent; Iterative training: Repeat forward propagation, loss calculation, and backpropagation until the prediction accuracy of the LSTM prediction model converges or the preset number of training iterations are reached, resulting in a trained LSTM prediction model; Model evaluation: Evaluate the trained LSTM prediction model using a test set, calculate the prediction accuracy and evaluation metrics. When the prediction accuracy is greater than or equal to a set value and the evaluation metrics meet the set conditions, output the trained LSTM prediction model; otherwise, retrain.

5. The smart grid monitoring and alarm method according to claim 1, characterized in that, The calculation method for the power grid health in step S7 is as follows: ;in, For the health of the power grid; Let be the predicted state variable for device i, which is a 0-1 variable, set to 0 when the device fails and 1 otherwise; n is the number of devices in the power grid.

6. The smart grid monitoring and alarm method according to claim 1, characterized in that, It also includes step S8, which visualizes the data on whether the equipment will fail at the next moment, the trend of changes in current, voltage, temperature and humidity starting at the next moment, the health of the power grid and the predicted number of cumulative failures over a period of time.

7. A smart grid monitoring and alarm system, characterized in that, The method applied to any one of claims 1 to 6 comprises: a data acquisition module for acquiring real-time operating data of the power grid, including current, voltage, temperature, and humidity data; a data processing and analysis module for processing and analyzing the acquired data to obtain processed current, voltage, temperature, and humidity data, as well as analyzed trend data of current, voltage, temperature, and humidity; a first prediction module for inputting the acquired data into a trained ANN prediction model to obtain a first prediction result; the first prediction result includes predicting the current, voltage, temperature, and humidity data for the next time step; and a second prediction module for inputting the trend data of current, voltage, temperature, and humidity into a trained LSTM prediction model to obtain a second prediction result; the second prediction result includes predicting the current, voltage, temperature, and humidity data for the next time step. The system includes: a humidity data module, and trend data of current, voltage, temperature, and humidity changes starting from the next moment; a result stitching module, which stitches the first and second prediction results to obtain the final current, voltage, temperature, and humidity data for the next moment, whether the equipment is faulty at the next moment, and trend data of current, voltage, temperature, and humidity changes starting from the next moment; a prediction alarm module, which sends alarm information to relevant personnel when equipment failure is predicted at the next moment; and an evaluation alarm module, which accumulates the number of equipment failures based on the prediction information and evaluates the health of the power grid and the number of predicted cumulative failures over a future period based on the trend data of current, voltage, temperature, and humidity changes starting from the next moment. When the health is less than a set threshold or the number of predicted cumulative failures reaches a set threshold, an alarm information is sent to relevant personnel.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the smart grid monitoring and alarm method according to any one of claims 1 to 6.

9. A processor, characterized in that, The processor is used to run a program, wherein the program executes the smart grid monitoring and alarm method according to any one of claims 1 to 6 when it runs.

Images