Artificial Intelligence-based Distribution Intelligent Management System

Abstract

The present invention discloses a distribution intelligent management system based on artificial intelligence, including: a collection module for collecting the current power consumption data of the power consumption area; a collection switching module connected to the power distribution processing module and the collection module respectively; a power distribution processing module connected to the collection switching module and the power conversion module respectively, controlling the collection switching module to work, so as to send a control signal to the corresponding power conversion module according to the current power consumption data of the input power consumption area; a power conversion module connected to the energy storage module, controlling the energy storage module to perform charge and discharge operations according to the control signal; a preset AI model is stored in the power distribution processing module, the current power consumption data and historical power consumption data of the power consumption area are input into the AI model, and the corresponding future power consumption demand and equipment hidden danger information of the power consumption area are output, and a control signal is generated according to the future power consumption demand, equipment hidden danger information of each power consumption area and the corresponding future weather information of the power consumption area.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply systems, and particularly to a distribution intelligent management system based on artificial intelligence. Background Art

[0002] At present, the distribution system supplies power according to the standard operating load of different power consumption areas to meet the usage needs of end users to a certain extent. However, it cannot adapt to and solve the problems such as insufficient power supply during peak power consumption periods and excessive electric energy during low power consumption periods, which easily lead to the problems of users lacking power for use and wasting electric power resources. Secondly, the distribution system uses multiple separate control lines for multiple power consumption areas to collect power consumption data for subsequent regulation and control processing. However, this method will result in high line costs, a large number of control terminals required for the control board, high overall electrical costs, low regulation and control efficiency, and affect the regulation and control effect. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a distribution intelligent management system based on artificial intelligence, which can improve the intelligent regulation and control efficiency, greatly reduce the waste rate and waste amount of electric energy, relieve the phenomenon of insufficient power supply during peak power consumption periods to the greatest extent, and improve the power consumption demand and satisfaction of users.

[0004] To solve the above technical problem, the present invention provides a distribution intelligent management system based on artificial intelligence, including a collection module, a collection switching module, a power distribution processing module, a power conversion module, and an energy storage module; the collection module is used to collect the current power consumption data of different power consumption areas in different regions; the collection switching module is respectively connected to the power distribution processing module and the collection module, and is used to switch to collect the current power consumption data of multiple different power consumption areas in the same region or the current power consumption data of multiple different power consumption areas in different regions and send it to the power distribution processing module; the power distribution processing module is respectively connected to the collection switching module and the power conversion module, and is used to control the operation of the collection switching module to send a control signal to the corresponding power conversion module according to the current power consumption data of the input power consumption area; the power conversion module is connected to the energy storage module, and is used to control the energy storage module to perform charge and discharge operations according to the control signal; wherein, a preset AI model is stored in the power distribution processing module, the current power consumption data and historical power consumption data of the power consumption area are input into the AI model, and the AI model outputs the corresponding future power consumption demand and equipment hidden danger information of the power consumption area according to the current power consumption data and historical power consumption data, and generates the control signal according to the future power consumption demand, equipment hidden danger information of each power consumption area and the future weather information corresponding to the power consumption area.

[0005] As an improvement to the above solution, the acquisition switching module includes a plurality of acquisition switching modules, and the acquisition switching module is used to switch and acquire the current power consumption data of a plurality of different power consumption areas in the same region or the current power consumption data of a plurality of different power consumption areas in different regions; the acquisition switching module includes a plurality of acquisition switching circuits, and the acquisition switching circuits are arranged in one-to-one correspondence with the acquisition modules.

[0006] As an improvement to the above solution, the acquisition module includes a voltage transformer circuit and a current transformer circuit. The voltage transformer circuit and the current transformer circuit are respectively used to acquire the voltage data and current data of the power consumption area, and the output ends of the voltage transformer circuit and the current transformer circuit are both connected to the acquisition switching circuit.

[0007] As an improvement to the above solution, every two groups of the acquisition switching modules are a first acquisition switching module and a second acquisition switching module with opposite switching structures; any control end of the power distribution processing module is respectively connected to the first acquisition switching module and the second acquisition switching module, and is used to switch and control the operation of the first acquisition switching module or the second acquisition switching module.

[0008] As an improvement to the above solution, the first acquisition switching circuit in the first acquisition switching module includes a first acquisition main switch circuit and a first acquisition sub-switch circuit, and the second acquisition switching circuit in the second acquisition switching module includes a second acquisition main switch circuit and a second acquisition sub-switch circuit; the first acquisition main switch circuit includes a first switching tube and a first resistor. The first end of the first switching tube is connected to the working power supply end, the second end of the first switching tube is connected to the first acquisition sub-switch circuit and grounded through the first resistor, and the control end of the first switching tube is connected to the power distribution processing module; the second acquisition main switch circuit includes a second switching tube and a second resistor. The first end of the second switching tube is connected to the working power supply end through the second resistor and is connected to the second acquisition sub-switch circuit, the second end of the second switching tube is grounded, and the control end of the second switching tube is connected to the power distribution processing module; both the first acquisition sub-switch circuit and the second acquisition sub-switch circuit are respectively connected to the power distribution processing module and the acquisition module. When the first switching tube is turned on, the first acquisition sub-switch circuit conducts the loop between the power distribution processing module and the acquisition module. When the second switching tube is turned on, the second acquisition sub-switch circuit conducts the loop between the power distribution processing module and the acquisition module.

[0009] As an improvement of the above solution, the first acquisition sub-switching circuit includes a third switching transistor, a fourth switching transistor, a third resistor, and a fourth resistor; the control terminals of the third switching transistor and the fourth switching transistor are both connected to the second terminal of the first switching transistor, the first terminal of the third switching transistor is connected to the output terminal of the voltage transformer circuit, the second terminal of the third switching transistor is connected to the power distribution processing module and grounded through the third resistor; the first terminal of the fourth switching transistor is connected to the voltage output terminal of the acquisition module, and the second terminal of the fourth switching transistor is connected to the power distribution processing module through the fourth resistor.

[0010] As an improvement of the above solution, the second acquisition sub-switching circuit includes a fifth switching transistor, a sixth switching transistor, a fifth resistor, and a sixth resistor; the control terminals of the fifth switching transistor and the sixth switching transistor are both connected to the first terminal of the second switching transistor, the first terminal of the fifth switching transistor is connected to the output terminal of the voltage transformer circuit, the second terminal of the fifth switching transistor is connected to the power distribution processing module and grounded through the fifth resistor; the first terminal of the sixth switching transistor is connected to the current output terminal of the acquisition module, and the second terminal of the sixth switching transistor is connected to the power distribution processing module through the sixth resistor.

[0011] As an improvement of the above solution, the DC terminal of the power conversion module is connected to the energy storage module, and the AC terminal of the power conversion module is connected to the power grid or an AC load, and is used to convert the alternating current of the power grid into direct current to charge the energy storage module, and convert the direct current of the energy storage module into alternating current to supply power to the power grid or an AC load.

[0012] As an improvement of the above solution, the power conversion module includes a DC circuit breaker, a grid-connected inverter, an AC filtering module, a transformer, an AC contactor, and an AC circuit breaker connected in sequence; the DC circuit breaker is connected to the energy storage module, and the AC circuit breaker is connected to the power grid or an AC load.

[0013] As an improvement of the above solution, the power distribution processing module generates a control signal according to the future power consumption demand, equipment hidden danger information, and future weather information corresponding to each power consumption area, including: obtaining the power consumption weight, equipment weight, and weather weight of the power consumption area according to the area position information of each power consumption area; calculating the power consumption demand factor according to the power consumption weight and the future power consumption demand; calculating the equipment hidden danger factor according to the equipment hidden danger information and the equipment weight; calculating the weather factor according to the future weather information and the weather weight; the future weather information includes at least future wind speed, future temperature, and future humidity, and the weather weight includes wind speed weight, temperature weight, and humidity weight; calculating the safety margin coefficient of the power consumption area according to the power consumption demand factor, equipment hidden danger factor, and weather factor; calculating the control signal of the power consumption area according to the safety margin coefficient and historical power consumption data.

[0014] As an improvement of the above solution, the power distribution processing module calculates a comprehensive factor based on the power consumption demand factor, the equipment hidden danger factor, and the weather factor; and calculates the safety margin coefficient according to the equipment hidden danger factor and the comprehensive factor.

[0015] The expression of the comprehensive factor includes: ;

[0016] The expression of the safety margin coefficient includes: ;

[0017] Where , , are the power consumption demand factor, the equipment hidden danger factor, and the weather factor corresponding to the power consumption area at time ; is the comprehensive factor corresponding to the power consumption area at time ; is the safety margin coefficient corresponding to the power consumption area at time ; is the preset margin weight.

[0018] The power distribution intelligent management system based on artificial intelligence provided by the present invention integrates multiple functional modules (including but not limited to the acquisition module, the acquisition switching module, the power distribution processing module, the power conversion module, and the energy storage module), aiming to achieve efficient power resource allocation, reduce power waste, and at the same time maximize power supply during peak power consumption periods, improving the user's power consumption experience and satisfaction. Next, in combination with the provided technical solutions, the technical content and beneficial effects of the present invention will be described in detail:

[0019] Acquisition module: This module is responsible for collecting real-time power consumption data from different regions and different power consumption areas. These data may include but are not limited to power consumption, power consumption time, power consumption patterns, etc., which are the basis for the system to perform intelligent analysis.

[0020] Acquisition switching module: This module serves as a bridge between data collection and processing, and can switch the data collection of different power consumption areas when needed, whether it is multiple areas in the same region or power consumption areas across regions. This function ensures that the system can flexibly respond to changes in power consumption demands in different regions.

[0021] Power Distribution Processing Module: One of the core modules, with a built-in preset AI model. This module receives real-time power consumption data from the acquisition module and, through the acquisition switching module, can selectively receive historical data. Through in-depth analysis of this data, the AI model not only predicts the future power consumption needs of each power consumption area but also identifies potential equipment hazards. In addition, considering the impact of future weather on power consumption demand, this module also incorporates weather forecast information as part of the basis for generating control signals.

[0022] Power Conversion Module: According to the control signals issued by the power distribution processing module, it performs specific power distribution and conversion tasks. This may involve converting some electrical energy into other forms of energy (such as chemical energy) for subsequent use, or converting the energy in the energy storage module back into electrical energy for user use when necessary.

[0023] Energy Storage Module: Used to store excess electrical energy converted during different periods, so as to release it during peak power consumption periods or when the grid power is insufficient, assisting in balancing the overall energy supply.

[0024] Beneficial Effects:

[0025] 1. Improve the efficiency of intelligent regulation: Through the real-time analysis and prediction of the AI model, the system can more accurately grasp the demand change trends of each power consumption area, realize the dynamic management and optimal allocation of electrical energy, and greatly improve the efficiency and accuracy of regulation.

[0026] 2. Significantly reduce the power waste rate and waste volume: Traditional power distribution methods often have difficulty accurately matching power consumption needs, easily causing power waste. The AI-based intelligent management system can effectively avoid this problem and minimize waste through scientific power allocation.

[0027] 3. Effectively alleviate the phenomenon of insufficient power supply during peak power consumption periods: The combination of the system's prediction ability and energy storage function can accumulate excess electrical energy during low-power consumption periods and release the stored energy during peak demand periods, thus effectively alleviating the situation of insufficient power supply and ensuring that the power consumption needs of users can be met at any time.

[0028] 4. Improve users' power consumption needs and satisfaction: Through the application of the above technical means, it can not only ensure the stable supply of electricity but also further optimize the power consumption cost, reduce unnecessary expenditures, and thus improve users' power consumption experience and satisfaction.

[0029] In summary, the intelligent power distribution management system based on artificial intelligence provided by the present invention realizes the efficient management and utilization of power resources through intelligent data acquisition, analysis, processing, and storage technologies, and is of great significance for promoting energy conservation and emission reduction, optimizing power services, etc.

[0030] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present invention. Brief Description of the Drawings

[0031] Figure 1 is a schematic structural diagram of the power distribution intelligent management system of the present invention;

[0032] Figure 2 is a schematic structural diagram of the acquisition module and the acquisition switching module of the present invention;

[0033] Figure 3 is a schematic structural diagram of the acquisition module, the acquisition switching module and the main controller of the present invention;

[0034] Figure 4 is a schematic structural diagram of the power conversion module and the energy storage module of the present invention;

[0035] Figure 5 is a schematic structural diagram of the power conversion module of the present invention.

[0036] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present invention. Detailed Embodiments

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0038] The flowcharts shown in the drawings are only illustrative examples, and do not necessarily include all the contents and operations / steps, nor do they necessarily need to be executed in the described order. For example, some operations / steps can be decomposed, combined or partially merged, so the actual execution order may change according to the actual situation.

[0039] It should be understood that, in order to facilitate the clear description of the technical solutions in the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. Those skilled in the art can understand that the terms "first" and "second" do not limit the quantity and execution order, and the terms "first" and "second" do not necessarily mean different.

[0040] It should be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0041] It should also be understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0042] Currently, the power distribution system supplies power accordingly according to the standard operating load of different power consumption areas to meet the usage needs of end users to a certain extent. However, it cannot adapt to solve the problems such as insufficient power supply during peak power consumption periods and excess electric energy during low power consumption periods, which easily leads to the problems of users lacking power for use and wasting power resources. Secondly, the power distribution system uses multiple separate control lines for multiple power consumption areas to collect power consumption data for subsequent regulation and control processing. However, this method will result in high line costs, a large number of control terminals required for the control board, high overall electrical costs, low regulation and control efficiency, and affect the regulation and control effect.

[0043] To solve the above problems, as Figure 1 shown, the present invention provides a power distribution intelligent management system based on artificial intelligence, including a collection module 1, a collection switching module 2, a power distribution processing module 3, a power conversion module 4 and an energy storage module 5;

[0044] The collection module 1 is used to collect the current power consumption data of different power consumption areas in different regions; wherein, the current power consumption data includes voltage and current data.

[0045] The collection switching module 2 is respectively connected to the power distribution processing module 3 and the collection module 1, and is used to switch and connect multiple collection modules 1 to collect the current power consumption data of multiple different power consumption areas in the same region or the current power consumption data of multiple different power consumption areas in different regions and send it to the power distribution processing module 3 to improve the collection efficiency.

[0046] The power distribution processing module 3 is respectively connected to the collection switching module 2 and the power conversion module 4, and is used to control the operation of the collection switching module 2, receive the current power consumption data of multiple power consumption areas collected, perform data processing according to the current power consumption data of the input power consumption area, and send corresponding control signals to the corresponding power conversion module 4, and can synchronously control the operation of multiple power conversion modules 4 to improve the regulation and control efficiency.

[0047] The power conversion module 4 is connected to the energy storage module 5 and is used to control the charging and discharging operation of the energy storage module 5 according to a control signal. It can store the excess electric energy in the power grid into the energy storage module 5 and can also feedback and transmit the electric energy of the energy storage module 5 to the power grid to achieve the regulation of electric energy resources in the power consumption area and maximize the satisfaction of the actual use needs of the power consumption area. Among them, a corresponding power conversion module 4 and energy storage module 5 are provided for each power consumption area.

[0048] Among them, a preset AI model is stored in the power distribution processing module 3. The current power consumption data and historical power consumption data of the power consumption area are input into the AI model. The AI model outputs the corresponding future power consumption demand and equipment hidden danger information of the power consumption area according to the current power consumption data and historical power consumption data. The power distribution processing module 3 generates a control signal according to the future power consumption demand, equipment hidden danger information of each power consumption area and the corresponding future weather information of the power consumption area.

[0049] Specifically, the acquisition module 1: is responsible for acquiring the current voltage and current data from different power consumption areas in different regions. It can be an intelligent electric meter installed at the user end or real-time data obtained through sensors. The acquisition switching module 2: by connecting with the acquisition module 1 and the power distribution processing module 3, can switch and connect multiple acquisition modules 1, so as to realize the efficient data acquisition of multiple power consumption areas. This can be achieved through a multiplexer or an intelligent switching device. The power distribution processing module 3: as the central processing unit of the system, it receives the power consumption data from the acquisition switching module 2 and conducts data processing. A preset AI model is stored inside this module. This model can predict the future power consumption demand and equipment hidden danger information according to the current power consumption data and historical power consumption data. In addition, the power distribution processing module 3 also generates a control signal according to the future power consumption demand, equipment hidden danger information of each power consumption area and the future weather information to control the operation of the power conversion module 4. The power conversion module 4: is connected to the energy storage module 5 and controls the charging and discharging operation of the energy storage module 5 according to the control signal sent by the power distribution processing module 3. The power conversion module 4 can be a bidirectional inverter or an intelligent charge and discharge controller, which can realize the efficient conversion and management of electric energy. The energy storage module 5: is used to store the excess electric energy in the power grid and feedback and transmit the stored electric energy to the power grid when needed. The energy storage module 5 can be a battery energy storage system, a supercapacitor or a flywheel energy storage system, etc.

[0050] The input data of the AI model are the current power consumption data (voltage, current), historical power consumption data, future weather information, etc. The output data are the future power consumption demand prediction, equipment hidden danger information, etc. The model type can be a time series prediction model (such as LSTM, GRU), a machine learning model (such as random forest, support vector machine) or a deep learning model (such as neural network). A large amount of historical power consumption data and weather data are used for model training to improve the prediction accuracy.

[0051] Suppose there are three power consumption areas A, B, and C, and a collection module 1 is installed in each area. The collection module 1 collects voltage and current data every 15 minutes and transmits it to the power distribution processing module 3 through the collection switching module 2. After receiving the data, the power distribution processing module 3 processes the data of each power consumption area using the internal AI model. For example, an LSTM model is used to predict the power consumption demand within the next 24 hours. Suppose based on the current and historical data, the AI model predicts that there will be a power consumption peak in area A at 8 am tomorrow, while the power consumption demands in areas B and C are relatively low. The power distribution processing module 3 generates a control signal according to the prediction result. The control signal instructs the power conversion module 4 to transfer the excess power from the energy storage modules 5 in areas B and C to area A to ensure that area A has sufficient power supply during the peak power consumption period. At the same time, the power distribution processing module 3 also adjusts the control signal according to the future weather information (such as high temperature or low temperature) to cope with the impact of extreme weather on power consumption demand. After receiving the control signal, the power conversion module 4 controls the energy storage module 5 in area A to discharge and transfer the stored power to the power grid to meet the power consumption demand in area A. At the same time, the power conversion module 4 controls the energy storage modules 5 in areas B and C to charge and store the excess power in the power grid for future use.

[0052] By intelligently predicting the future power consumption demand, the system can store the excess power during the low power consumption period and release it during the peak power consumption period, thus effectively balancing the supply and demand and improving the power supply efficiency. Using the collection switching module 2, the system can efficiently collect the data of multiple power consumption areas, reduce the number of individual control lines, and lower the requirements for the number of control terminals on the control board and the line cost. The introduction of the AI model enables the system to more accurately predict the future power consumption demand and timely adjust the power distribution and storage, improving the regulation effect and response speed. By intelligently managing the storage and distribution of power, the system can minimize the waste of power during the low power consumption period and ensure the stability of the power grid during the peak power consumption period. The AI model can detect the abnormal operation of equipment, discover and report potential equipment hazards in advance, thus reducing the power outages and maintenance costs caused by equipment failures. By combining the future weather information, the system can more flexibly adjust the power distribution strategy to cope with the impact of extreme weather such as high temperature and low temperature on power consumption demand and ensure that the power consumption demands of users are met under different weather conditions.

[0053] In summary, the provided artificial intelligence-based power distribution intelligent management system can effectively solve the problems of insufficient power supply during the peak power consumption period and excess power during the low power consumption period in the traditional power distribution system through intelligent prediction and flexible regulation, improve the power supply efficiency, reduce costs, and enhance the user experience.

[0054] Specifically, for the AI model training process of this application, it is necessary to carefully consider data preparation, model selection, training strategies, evaluation methods, as well as model optimization and deployment. The following is a detailed training process design:

[0055] First, collect historical electricity consumption data such as voltage, current, and power from the historical data of multiple electricity consumption areas. The historical electricity consumption data should include historical electricity consumption data during different time periods (such as peak hours and off-peak hours) and under different weather conditions.

[0056] Historical weather data: Collect historical weather data for each electricity consumption area, including temperature, humidity, wind speed, precipitation, etc.

[0057] Historical equipment status data: Collect historical equipment status data of the equipment from the power grid and electricity-consuming equipment, including equipment operation time, fault records, maintenance records, etc.

[0058] Then, handle missing values and outliers in the historical electricity consumption data, historical weather data, and historical operation status data by filling or deleting the missing values in the data. Interpolation methods or statistical methods based on historical data can be used for filling. Detect and handle outliers. Statistical methods (such as the standard deviation method) or machine learning methods (such as Isolation Forest) can be used for outlier detection. Then, standardize the data to ensure that different features are on the same scale. Z-score standardization or Min-Max standardization can be used.

[0059] Furthermore, manually annotate or automatically calculate the actual electricity consumption demand of each electricity consumption area during different time periods based on the historical electricity consumption data, historical weather data, and historical operation status data. Mark the potential hazard level of each equipment during different time periods according to the equipment's fault records and maintenance records.

[0060] Extract statistical features such as the maximum value, minimum value, average value, and standard deviation of voltage, current, and power from the historical electricity consumption data. Extract time-related features in the historical equipment status data, such as hour, day of the week, holiday, etc. Extract weather features such as temperature, humidity, wind speed, and precipitation from the historical weather data, and perform appropriate data processing (such as converting temperature to relative temperature difference). Extract features such as the operation time, number of faults, and maintenance records of the equipment from the historical equipment status data.

[0061] Use feature selection methods (such as recursive feature elimination, LASSO regression) to select the most relevant features, reduce redundant features, and improve the training efficiency and generalization ability of the model.

[0062] The AI model includes a time series prediction model and a classification model. Time series prediction model: such as LSTM (Long Short-Term Memory Network), GRU (Gated Recurrent Unit), etc., which are used to predict future electricity consumption demands. Classification model: such as random forest, support vector machine, deep neural network, etc., which are used to identify potential hazards of equipment.

[0063] Design a multi-task learning model that can simultaneously predict electricity consumption demands and identify potential hazards of equipment. Such a model can share the underlying feature extraction layer to improve the efficiency and accuracy of the model.

[0064] Training set: used to train the model, accounting for 70% of the total data. Validation set: used to tune hyperparameters and evaluate the performance of the model, accounting for 15% of the total data. Test set: used to finally evaluate the performance of the model, accounting for 15% of the total data.

[0065] Use grid search or random search methods to tune the hyperparameters of the model, such as the hidden layer size, learning rate, batch size, etc. of LSTM. Adopt the cross-validation method to ensure the stability of the tuned hyperparameters on different data sets.

[0066] For electricity consumption demand prediction, use the mean squared error (MSE) or mean absolute error (MAE) as the loss function to minimize the error between the predicted value and the actual value.

[0067] For equipment potential hazard identification, use the cross-entropy loss function to minimize classification errors. Standardize the data and convert it into the format required for model input. Adopt batch gradient descent or stochastic gradient descent methods for training to ensure that the model can effectively learn the features of the data. During the training process, simultaneously optimize the loss functions of the two tasks of electricity consumption demand prediction and equipment potential hazard identification. Set regular checkpoints during the training process to save the intermediate state of the model for subsequent recovery or tuning.

[0068] Use the mean squared error (MSE) to evaluate the error magnitude of the model when predicting electricity consumption demands.

[0069] Use the R² score to evaluate the goodness of fit of the model. The closer R² is to 1, the better the fit of the model.

[0070] Compare the prediction results of the model with the actual electricity consumption data to evaluate the prediction accuracy at different time periods (such as peak hours, off-peak hours). Accuracy evaluates the correct rate of the model when classifying equipment potential hazards. Precision and Recall evaluate the precision and coverage of the model when identifying equipment potential hazards.

[0071] Use existing pre-trained models (such as pre-trained LSTM models) for transfer learning, and quickly improve the prediction ability of the model by fine-tuning on the basis of the pre-trained model. Use multiple models (such as LSTM, GRU, and random forest) for ensemble learning, and improve the stability and prediction accuracy of the model through ensemble methods (such as voting method, stacking method).

[0072] Adopt interpretability methods (such as LIME or SHAP) to explain the prediction results of the model, ensure the transparency of the decision-making process of the model, and facilitate the understanding and trust of grid managers.

[0073] Deploy the trained model to the distribution processing module 3 to ensure that the model can operate efficiently in the actual environment. Use containerization technology (such as Docker) to deploy the model to improve the flexibility and maintainability of the deployment. Monitor the running status and performance of the model in real time, and collect feedback data of the model in actual applications.

[0074] Use anomaly detection methods (such as statistics-based methods or machine learning methods) to detect the prediction errors of the model, and adjust the model parameters or retrain the model in a timely manner. Regularly update the dataset of the model, including new power consumption data, weather data, and equipment status data. Use incremental learning methods (such as online learning) to continuously optimize the model to ensure that the model can adapt to the dynamic changes of the power grid. During the training process, dynamically select the most relevant features, adjust the feature selection strategy according to the performance feedback of the model, and improve the adaptability and accuracy of the model. Use reinforcement learning methods to optimize the decision-making process of the model. By defining a reward function (such as the reward for successfully smoothing the load during peak power consumption periods), let the model continuously learn and optimize during actual operation to improve the control effect. Combine the power demand prediction with the optimization of power distribution, and design a closed-loop control system. The model not only predicts the future power demand, but also optimizes the power distribution and storage strategies according to the prediction results to improve the overall performance of the system.

[0075] Through the above detailed training process design, it can be ensured that the distribution intelligent management system based on artificial intelligence has efficient, accurate, and reliable performance in actual applications, effectively solve the deficiencies of traditional distribution systems, and improve the user experience and the operation efficiency of the power grid.

[0076] In some embodiments, the power distribution processing module 3 generates a control signal based on the future power consumption demand, equipment potential hazard information, and future weather information corresponding to each power consumption area, including: obtaining the power consumption weight, equipment weight, and weather weight of the power consumption area according to the area location information of each power consumption area; calculating the power consumption demand factor according to the power consumption weight and the future power consumption demand; calculating the equipment potential hazard factor according to the equipment potential hazard information and the equipment weight; calculating the weather factor according to the future weather information and the weather weight; the future weather information includes at least future wind speed, future temperature, and future humidity, and the weather weight includes wind speed weight, temperature weight, and humidity weight; calculating the safety margin coefficient of the power consumption area according to the power consumption demand factor, the equipment potential hazard factor, and the weather factor; and calculating the control signal of the power consumption area according to the safety margin coefficient and the historical power consumption data.

[0077] The power consumption location information is the location information of each power consumption area, which may include geographical coordinates, geographical location types (such as city center, suburb, rural area, etc.). The equipment location information is the equipment location information within each power consumption area, which may include the installation location of the equipment, the equipment type (such as large industrial equipment, residential equipment, etc.). The weather location information is the weather station location information of each power consumption area to ensure more accurate weather data acquisition.

[0078] Determine the power consumption weight according to the location information of the power consumption area and the historical power consumption data. For example, the power consumption area in the city center may have a higher power consumption weight. Determine the equipment weight according to the location information of the equipment and the failure frequency. For example, the equipment weight of large industrial equipment may be higher. Determine the weather weight according to the location information of the power consumption area and the accuracy of the weather station. The weather weight can be further divided into wind speed weight, temperature weight, and humidity weight.

[0079] Exemplarily, the power distribution processing module 3 calculates a comprehensive factor according to the power consumption demand factor, the equipment potential hazard factor, and the weather factor; calculates the safety margin coefficient according to the equipment potential hazard factor and the comprehensive factor.

[0080] The expression of the comprehensive factor includes: ;

[0081] The expression of the safety margin coefficient includes: ;

[0082] Wherein, , , are the power consumption demand factor, the equipment potential hazard factor, and the weather factor corresponding to the power consumption area at time ; is the comprehensive factor corresponding to the power consumption area at time ; The electricity consumption area At a time The corresponding safety margin coefficient; Is a preset margin weight.

[0083] Furthermore, the expression of the control signal includes:

[0084] ;

[0085] Is a threshold dynamically adjusted according to historical electricity consumption data and the overall grid load condition. Its expression includes: ;

[0086] It should be noted that Is the electricity consumption area In the past Average comprehensive factor within the time period, Is the standard deviation, Is a preset adjustment coefficient for balancing the strictness and flexibility of the threshold.

[0087] It should be noted that when the electricity consumption obtained based on the collected electricity consumption data is within the first preset threshold range, it indicates that the current electricity consumption area is in the low - valley electricity consumption working state; when the electricity consumption is within the second preset threshold range, it indicates that the current electricity consumption area is in the normal electricity consumption working state; when the electricity consumption is within the third preset threshold range, it indicates that the current electricity consumption area is in the peak - electricity consumption working state. The power distribution processing module 3 issues relevant control instructions to the power conversion module 4 according to the current working state of the electricity consumption area to control the power conversion module 4 to perform corresponding operations.

[0088] For example, when the electricity consumption area is in the low - valley electricity consumption working state, the power distribution processing module 3 controls the power conversion module 4 to store the redundant electricity in the power grid of the electricity consumption area into the energy storage module 5 for subsequent use, greatly reducing the waste rate and waste amount of electricity. When the electricity consumption area is in the peak - electricity consumption working state, the power distribution processing module 3 controls the power conversion module 4 to feedback the electricity of the energy storage module 5 back to the power grid to increase the available electric energy during peak - electricity consumption periods, so as to alleviate the power supply shortage phenomenon during peak - electricity consumption periods to the greatest extent and improve the electricity consumption demand and satisfaction of users. By synchronously performing intelligent regulation processing on multiple electricity consumption areas, the intelligent regulation efficiency and intelligent regulation effect can be improved.

[0089] Preferably, the power distribution processing module 3 includes a main controller and a communication module. The main controller is communicatively connected to other devices or apparatuses or modules through a wired communication line module or a wireless communication module to achieve data interaction processing.

[0090] Such as Figures 2-3As shown in the figure, the acquisition module 1 includes a voltage transformer circuit 11 and a current transformer circuit 12. The voltage transformer circuit 11 and the current transformer circuit 12 are respectively used to acquire voltage data and current data of the power consumption area. The output ends of the voltage transformer circuit 11 and the current transformer circuit 12 are both connected to the acquisition switching circuit 22. It should be noted that both the voltage transformer circuit 11 and the current transformer circuit 12 are conventional circuits in the prior art, and their working principles and structures will not be elaborated here one by one.

[0091] The acquisition switching module 2 includes a plurality of acquisition switching modules 21. The acquisition switching module 21 is used to switch and acquire the current power consumption data of multiple different power consumption areas in the same region or the current power consumption data of multiple different power consumption areas in different regions. By switching and controlling the operation of the acquisition switching module 21, data acquisition of more power consumption areas can be realized to improve the intelligent regulation efficiency. Among them, the acquisition switching module 21 includes a plurality of acquisition switching circuits 22 with the same structure. The acquisition switching circuit 22 is arranged in one-to-one correspondence with the acquisition module 1 to obtain the power consumption data of the corresponding power consumption area.

[0092] Every two groups of the acquisition switching modules 21 are the first acquisition switching module 211 and the second acquisition switching module 212 with opposite switching structures to each other; any control end of the main controller U1 is respectively connected to the first acquisition switching module 211 and the second acquisition switching module 212, and is used to switch and control the operation of the first acquisition switching module 211 or the second acquisition switching module 212. Through multiple pairs of the first acquisition switching module 211 and the second acquisition switching module 212, synchronous acquisition of the power consumption data of a large number of power consumption areas can be realized, and the synchronous processing efficiency and the intelligent regulation efficiency can be improved.

[0093] Specifically, the first acquisition switching circuit 221 in the first acquisition switching module 211 includes a first acquisition main switch circuit 2211 and a first acquisition sub-switch circuit 2212, and the second acquisition switching circuit 222 in the second acquisition switching module 212 includes a second acquisition main switch circuit 2221 and a second acquisition sub-switch circuit 2222; the first acquisition main switch circuit 2211 includes a first switching tube Q1 and a first resistor R1. The first end of the first switching tube Q1 is connected to the working power supply terminal VDD, the second end of the first switching tube Q1 is connected to the first acquisition sub-switch circuit 2212 and grounded through the first resistor R1, and the control end of the first switching tube Q1 is connected to the main controller U1; the second acquisition main switch circuit 2221 includes a second switching tube Q7 and a second resistor R7. The first end of the second switching tube Q7 is connected to the working power supply terminal VDD through the second resistor R7 and is connected to the second acquisition sub-switch circuit 2222, the second end of the switching tube is grounded, and the control end of the second switching tube Q7 is connected to the main controller U1; both the first acquisition sub-switch circuit 2212 and the second acquisition sub-switch circuit 2222 are respectively connected to the main controller U1 and the acquisition module 1. When the first switching tube Q1 is turned on, the first acquisition sub-switch circuit 2212 turns on the loop between the main controller U1 and the acquisition module 1. When the second switching tube Q7 is turned on, the second acquisition sub-switch circuit 2222 turns on the loop between the main controller U1 and the acquisition module 1.

[0094] It should be noted that as Figure 3 shown, the main controller includes a first pin group (including the first sub-pin A1 to the nth sub-pin An), a second pin group (including the first auxiliary pin S1 to the nth auxiliary pin Sn), and a third pin group (including the first slave pin C1 to the nth slave pin cn). When the first slave pin C1 of the third pin group of the main controller U1 outputs a high-level signal, the first switching tube Q1 conducts and works, enabling the first acquisition sub-switch circuit 2212 to turn on the loop between the main controller U1 and the acquisition module 1. The first auxiliary pin S1 in the second pin group of the main controller U1 receives the voltage data of the corresponding power consumption area, and the first sub-pin A1 in the first pin group receives the current data of the corresponding power consumption area. At this time, the level of the control end of the second switching tube Q7 is high, and the second switching tube Q7 does not conduct and work. The second acquisition sub-switch circuit 2222 disconnects the loop between the main controller U1 and the acquisition module 1, and the main controller U1 does not obtain the power consumption data of this power consumption area for intelligent regulation.

[0095] Through the first slave pin C1 to the nth slave pin Cn in the third pin group of the main controller U1, the control of n first acquisition switching modules 211 and n second acquisition switching modules 212 can be synchronously switched to perform acquisition work, which can reduce the demand for control terminals, and can further acquire the power consumption data of more power consumption areas to perform intelligent regulation synchronously, improving the efficiency and effect of intelligent regulation, so as to meet the power regulation requirements of multiple power consumption areas. By sharing the first auxiliary pin S1 to the nth auxiliary pin Sn in the second pin group and the first sub-pin A1 to the nth sub-pin An in the first pin group, the power consumption data input through the first acquisition switching module 211 and the second acquisition switching module 212 can be received respectively, improving the pin multiplexing effect, simplifying the line control structure, reducing the line cost, making the overall electrical cost low, and improving the regulation efficiency and regulation effect.

[0096] Preferably, the specific number of the acquisition switching circuits 22 in the acquisition switching module 21 can be set and adjusted according to the actual situation, such as according to the number of power consumption areas to be acquired and processed.

[0097] Preferably, the first switching transistor Q1 is preferably an NPN triode, the second switching transistor Q7 is preferably a PNP triode, and the main controller U1 is preferably a PLC controller, but not limited thereto.

[0098] Among them, the first acquisition sub-switching circuit 2212 includes a third switching transistor Q2, a fourth switching transistor Q3, a third resistor R2, and a fourth resistor R3; the control terminals of the third switching transistor Q2 and the fourth switching transistor Q3 are both connected to the second end of the first switching transistor Q1, the first end of the third switching transistor Q2 is connected to the output end of the voltage transformer circuit 11, the second end of the third switching transistor Q2 is connected to the main controller U1 and grounded through the third resistor R2; the first end of the fourth switching transistor Q3 is connected to the voltage output end of the acquisition module 1, and the second end of the fourth switching transistor Q3 is connected to the main controller U1 through the fourth resistor R3.

[0099] It should be noted that when the first switching transistor Q1 is turned on, the electrical levels at the control terminals of the third switching transistor Q2 and the fourth switching transistor Q3 are both in the high level state, and the third switching transistor Q2 and the fourth switching transistor Q3 are turned on to work, so that the pin S1 of the main controller U1 is connected to the voltage output end of the voltage transformer in the acquisition module 1, and the first sub-pin A1 is connected to the current output end of the current transformer in the acquisition module 1, so that the main controller U1 can obtain the power consumption data of this power consumption area and calculate the relevant power consumption for intelligent regulation processing.

[0100] Preferably, both the third switching transistor Q2 and the fourth switching transistor Q3 are preferably N-channel junction field effect transistors, but not limited thereto.

[0101] The second acquisition sub-switching circuit 2222 includes a fifth switching transistor Q8, a sixth switching transistor Q9, a fifth resistor R8, and a sixth resistor R9; the control terminals of the fifth switching transistor Q8 and the sixth switching transistor Q9 are both connected to the first end of the second switching transistor Q7, the first end of the fifth switching transistor Q8 is connected to the output end of the voltage transformer circuit 11, the second end of the fifth switching transistor Q8 is connected to the main controller U1 and grounded through the fifth resistor R8; the first end of the sixth switching transistor Q9 is connected to the current output end of the acquisition module 1, and the second end of the sixth switching transistor Q9 is connected to the main controller U1 through the sixth resistor R9.

[0102] It should be noted that when the second switching transistor Q7 is turned on, the electrical levels at the control terminals of the fifth switching transistor Q8 and the sixth switching transistor Q9 are both in the high-level state, and the fifth switching transistor Q8 and the sixth switching transistor Q9 are turned on to work, so that the pin S1 of the main controller U1 is connected to the voltage output end of the voltage transformer circuit 11 in the acquisition module 1, and the first sub-pin A1 is connected to the current output end of the current transformer circuit 12 in the acquisition module 1, so that the main controller U1 can obtain the power consumption data of the power consumption area and calculate the relevant power consumption for intelligent regulation processing.

[0103] Preferably, both the fifth switching transistor Q8 and the sixth switching transistor Q9 are preferably P-channel junction field effect transistors, but not limited thereto.

[0104] As Figure 4 shown, the DC end of the power conversion module 4 is connected to the energy storage module 5, the AC end of the power conversion module 4 is connected to the power grid, and the power grid is also connected to the power consumption area. The power conversion module 4 is used to convert the alternating current of the power grid into direct current to charge the energy storage module 5, and convert the direct current of the energy storage module 5 into alternating current to supply power to the power grid. Among them, the power conversion module 4 is preferably a PCS energy storage converter, but not limited thereto. The AC end of the power conversion module 4 can also be directly connected to an AC load for users to use to meet the power consumption requirements.

[0105] As Figure 5 shown, the power conversion module 4 includes a DC circuit breaker 41, a grid-connected inverter 42, an AC filtering module 43, a transformer 44, an AC contactor 45, and an AC circuit breaker 46 connected in sequence; the DC circuit breaker 41 is connected to the energy storage module 5, and the AC circuit breaker 46 is connected to the power grid. Among them, the AC filtering module 43 includes inductance, capacitance filtering, etc.

[0106] The AC-DC conversion operation can be achieved by controlling the on / off of the DC circuit breaker 41 or the AC circuit breaker 46. When the power consumption in the power consumption area is at a low level, when the DC circuit breaker 41 is controlled to conduct, the redundant electric energy in the power grid can be stored in the energy storage module 5 for subsequent use, greatly reducing the waste rate and waste amount of electric energy. When the power consumption in the power consumption area is at a peak level, the AC circuit breaker 46 can be controlled to work, and the redundant electric energy in the energy storage module 5 can be fed back to the power grid for use in the power consumption area, increasing the available electric energy during peak power consumption periods, thus alleviating the power supply shortage during peak power consumption periods to the greatest extent and improving the power consumption demand and satisfaction of users.

[0107] In summary, the present invention can switch to collect the current power consumption data of multiple different power consumption areas in the same region or the current power consumption data of multiple different power consumption areas in different regions to improve the regulation efficiency; according to the power consumption data of multiple power consumption areas, it can simultaneously and intelligently regulate the power output and storage of different power consumption areas, improve the intelligent regulation efficiency, greatly reduce the waste rate and waste amount of electric energy, alleviate the power supply shortage during peak power consumption periods to the greatest extent, and improve the power consumption demand and satisfaction of users.

[0108] Secondly, the line control structure can be simplified, the line cost can be reduced, the demand for the control end can be reduced by optimizing and reusing it, the overall electrical cost is low, and the regulation efficiency and regulation effect are improved.

[0109] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions recorded in the foregoing embodiments or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. An intelligent power distribution management system based on artificial intelligence, characterized in that: It includes a collection module, a collection switching module, a power distribution processing module, a power conversion module and an energy storage module; A collection module is installed in each different power consumption area; the collection module is used to collect current power consumption data of different power consumption areas in different regions; The collection switching module is connected to the power distribution processing module and the collection module respectively, and is used to switch and collect the current power consumption data of multiple different power consumption areas in the same area or the current power consumption data of multiple different power consumption areas in different areas and send them to the power distribution processing module; The power distribution processing module is connected to the acquisition switching module and the power conversion module respectively, and is used to control the acquisition switching module to work, so as to send a control signal to the corresponding power conversion module according to the current power consumption data of the input power consumption area; The electric energy conversion module is connected to the energy storage module and is used to control the energy storage module to perform charging and discharging operations according to a control signal; Wherein, a preset AI model is stored in the power distribution processing module, and the current power consumption data and historical power consumption data of the power consumption area are input into the AI ​​model. The AI ​​model outputs the future power demand and equipment hidden danger information corresponding to the power consumption area according to the current power consumption data and the historical power consumption data, and generates the control signal according to the future power demand, equipment hidden danger information and future weather information corresponding to each of the power consumption areas, including: obtaining the power consumption weight, equipment weight and weather weight of the power consumption area according to the regional location information of each power consumption area; calculating the power demand factor according to the power weight and future power demand; calculating the equipment hidden danger factor according to the equipment hidden danger information and the equipment weight; calculating the weather factor according to the future weather information and the weather weight; the future weather information includes at least the future wind speed, future temperature and future humidity, and the weather weight includes the wind speed weight, the temperature weight and the humidity weight; calculating the safety margin coefficient of the power consumption area according to the power demand factor, the equipment hidden danger factor and the weather factor; calculating the control signal of the power consumption area according to the safety margin coefficient and the historical power consumption data; Among them, the acquisition module includes a voltage transformer circuit and a current transformer circuit, the voltage transformer circuit and the current transformer circuit are respectively used to collect voltage data and current data of the power consumption area, and the output ends of the voltage transformer circuit and the current transformer circuit are connected to the acquisition switching module; the acquisition switching module includes multiple acquisition switching modules, and the acquisition switching module includes multiple acquisition switching circuits, and the acquisition switching circuits are arranged one by one with the acquisition modules; every two groups of acquisition switching modules are a first acquisition switching module and a second acquisition switching module with opposite switching structures; any control end of the power distribution processing module is respectively connected to the first acquisition switching module and the second acquisition switching module, which is used to switch and control the operation of the first acquisition switching module or the second acquisition switching module, and can also synchronously switch and control multiple first acquisition switching modules and multiple second acquisition switching modules to perform acquisition work.

2. The system according to claim 1, characterized in that The first acquisition switching circuit in the first acquisition switching module includes a first acquisition main switch circuit and a first acquisition auxiliary switch circuit, and the second acquisition switching circuit in the second acquisition switching module includes a second acquisition main switch circuit and a second acquisition auxiliary switch circuit; The first acquisition main switch circuit includes a first switch tube and a first resistor, the first end of the first switch tube is connected to the working power supply end, the second end of the first switch tube is connected to the first acquisition auxiliary switch circuit and grounded through the first resistor, and the control end of the first switch tube is connected to the power distribution processing module; The second acquisition main switch circuit includes a second switch tube and a second resistor, a first end of the second switch tube is connected to the working power supply end through the second resistor and is connected to the second acquisition auxiliary switch circuit, a second end of the second switch tube is grounded, and a control end of the second switch tube is connected to the power distribution processing module; The first acquisition sub-switch circuit and the second acquisition sub-switch circuit are respectively connected to the power distribution processing module and the acquisition module. When the first switch tube is turned on, the first acquisition sub-switch circuit turns on the loop between the power distribution processing module and the acquisition module. When the second switch tube is turned on, the second acquisition sub-switch circuit turns on the loop between the power distribution processing module and the acquisition module.

3. The system according to claim 2, characterized in that The first acquisition auxiliary switch circuit includes a third switch tube, a fourth switch tube, a third resistor and a fourth resistor; The control ends of the third switch tube and the fourth switch tube are both connected to the second end of the first switch tube, the first end of the third switch tube is connected to the output end of the voltage transformer circuit, and the second end of the third switch tube is connected to the power distribution processing module and grounded through a third resistor; The first end of the fourth switch tube is connected to the voltage output end of the acquisition module, and the second end of the fourth switch tube is connected to the power distribution processing module through a fourth resistor.

4. The system according to claim 2, characterized in that The second acquisition auxiliary switch circuit includes a fifth switch tube, a sixth switch tube, a fifth resistor and a sixth resistor; The control ends of the fifth switch tube and the sixth switch tube are both connected to the first end of the second switch tube, the first end of the fifth switch tube is connected to the output end of the voltage transformer circuit, and the second end of the fifth switch tube is connected to the power distribution processing module and grounded through a fifth resistor; The first end of the sixth switch tube is connected to the current output end of the acquisition module, and the second end of the sixth switch tube is connected to the power distribution processing module through a sixth resistor.

5. The system according to claim 1, characterized in that The DC end of the electric energy conversion module is connected to the energy storage module, and the AC end of the electric energy conversion module is connected to the power grid or the AC load, and is used to convert the AC power of the power grid into DC power to charge the energy storage module, and convert the DC power of the energy storage module into AC power to power the power grid or the AC load.

6. The system according to claim 1, characterized in that The power distribution processing module calculates a comprehensive factor based on power demand factors, equipment hidden danger factors and weather factors; Calculate the safety margin coefficient according to the equipment hidden danger factor and the comprehensive factor; The expression of the comprehensive factor includes: ; The expression of the safety margin factor includes: ; in, , , It is the electricity consumption area In time Corresponding electricity demand factors, equipment hidden danger factors and weather factors; It is the electricity consumption area In time The corresponding comprehensive factor; It is the electricity consumption area In time The corresponding safety margin factor; is the preset margin weight.

Images