A smart power utilization information management system

Abstract

This invention proposes a smart electricity information management system. The system acquires and predicts power supply-related emergencies in real time through a data integration unit, a feature engineering unit preprocesses and extracts power data features, a multi-model fusion unit selects a combination of prediction models for prediction based on demand, an adaptive learning unit builds or optimizes prediction models through deep learning, and a reinforcement learning unit continuously trains the models, thereby achieving accurate prediction of electricity demand and optimization of supply.

Description

Technical Field

[0001] This invention relates to the field of electricity consumption behavior visualization technology, and specifically to a smart electricity consumption information management system. Background Technology

[0002] With social development and technological advancements, the power system plays an increasingly vital role in people's lives. The continuous growth in electricity demand and the volatility of power supply pose numerous challenges to the safe and stable operation of the power system. Traditional power information management systems primarily rely on manual experience for power dispatch and forecasting, which has limitations such as low forecast accuracy and insufficient ability to respond to emergencies. To address these issues, in recent years, researchers have proposed intelligent electricity information management systems based on big data and artificial intelligence.

[0003] While existing smart electricity information management systems have improved the accuracy of electricity forecasting and the ability to respond to emergencies to some extent, they still have the following problems:

[0004] Insufficient data integration capabilities: When acquiring external data, existing systems often can only process data from a single source, making it difficult to achieve real-time integration of multi-source data, resulting in inaccurate prediction results.

[0005] Feature engineering processing methods are limited: Existing systems often use fixed feature extraction methods when processing power data, which are difficult to adapt to the prediction needs of different scenarios.

[0006] Single prediction model: Existing systems typically use a single prediction model for power forecasting, which is difficult to cope with complex power system environments.

[0007] Insufficient model adaptability: When faced with environmental changes, sudden events, or nonlinear effects, the existing system's prediction model has poor adaptability, which can easily lead to increased prediction errors.

[0008] Insufficient model learning and updating capabilities: After acquiring new data samples, the existing system fails to effectively utilize these data to update and optimize the prediction model. Summary of the Invention

[0009] The purpose of this invention is to propose a smart electricity information management system to solve the above-mentioned technical problems.

[0010] To achieve the above objectives, the present invention provides a smart electricity information management system, comprising:

[0011] The data integration unit is used to acquire external data from different sources in real time and to predict power supply-related emergencies based on the external data.

[0012] The feature engineering unit is used to acquire power data, preprocess the power data to obtain a power dataset, and extract features from the power dataset to obtain feature inputs for prediction.

[0013] The feature engineering unit is used to select a matching combination of prediction models from the model database according to the prediction requirements, perform prediction processing on the feature inputs used for prediction, and obtain the prediction results; the combination of prediction models includes at least one prediction model, and the prediction requirements include electricity demand prediction and electricity supply prediction.

[0014] An adaptive learning unit is used to construct a prediction model for the prediction requirements through deep learning or to select a prediction model to be adaptively learned from the model database. After adaptively training the constructed or selected prediction model based on environmental changes, sudden events or nonlinear effects, it is added to the model database.

[0015] The reinforcement learning unit is used to acquire incremental data samples and perform reinforcement learning training on multiple prediction models in the model database based on the incremental data samples.

[0016] Preferably, the data integration unit includes:

[0017] The real-time data acquisition module is used to acquire external data from different sources in real time; the external data from different sources includes meteorological data, market demand data, and socio-economic activity data.

[0018] The time prediction module is used to analyze the external data based on a time-driven prediction method to predict whether a sudden event will occur, and to trigger an early warning mechanism to issue an early warning notification when a sudden event occurs.

[0019] Preferably, the feature engineering unit includes:

[0020] The timeliness adjustment module is used to obtain power data within the most recent preset time period;

[0021] The data augmentation module is used to generate synthetic data based on the power data obtained by the timeliness adjustment module, so as to increase the data samples used for model training; the synthetic data is a data sample similar to the obtained power data generated by the simulation model, or a data sample simulating a sudden event or abnormal situation.

[0022] The preprocessing module is used to preprocess the power data obtained by the timeliness adjustment module to obtain a power dataset by removing missing values, outliers, and duplicate data, and to extract features from the power dataset to obtain feature inputs for prediction; it is also used to extract features from the synthetic data generated by the data augmentation module to obtain feature inputs for training the model.

[0023] Preferably, the multi-model fusion unit includes:

[0024] Select the module and choose the best combination of models based on the historical performance of the models and the current data distribution of the feature inputs;

[0025] The learning method fusion module is used to fuse multiple prediction models of the optimal model combination by weighting or stacking.

[0026] The outlier handling module is used to automatically identify and correct anomalies or deviations in the prediction results of the prediction model.

[0027] The multi-model fusion unit includes:

[0028] The deep neural network module is used to construct a prediction model for the prediction requirements through deep learning or to select a prediction model to be adaptively learned from the model database.

[0029] The adaptive algorithm module is used to adaptively learn and train the constructed or selected prediction model based on environmental changes, sudden events, or nonlinear effects, and then add it to the model database.

[0030] Preferably, it further includes:

[0031] A mobile terminal is used to log in, log out, and operate the information management system.

[0032] Preferably, the mobile terminal integrates an archived work order data upload module and a query module. The archived work order data upload module is used to store and display the archived work orders, and the query module is used to query the archived work orders.

[0033] Preferably, the query conditions of the query module are: power supply unit, work order number, user number, and user name;

[0034] The query module outputs the following query results: attachment details, work order number, user name, user number, and electricity address. When the user clicks on the attachment details, a pop-up window displays the attachments for the business expansion archiving stage under this work order, which corresponds to the attachment management of the business expansion archiving stage in the marketing system.

[0035] The intelligent electricity information management system of the present invention has the following beneficial effects:

[0036] Improve forecast accuracy: By acquiring multi-source external data in real time through the data integration unit and combining it with the feature engineering unit to preprocess and extract features from the power data, the factors affecting power demand and supply can be captured more comprehensively, thereby significantly improving the accuracy of forecasts.

[0037] Enhancing the ability to respond to emergencies: This invention can predict emergencies related to power supply based on external data, enabling the power system to respond quickly to emergencies and reduce or avoid the resulting losses.

[0038] Enhanced system adaptability: The multi-model fusion unit can select appropriate model combinations according to different forecasting needs, making the system more adaptable to complex and ever-changing power market and environmental conditions.

[0039] Optimize the prediction model: The adaptive learning unit can adaptively learn and train the prediction model according to environmental changes, sudden events or nonlinear effects, ensuring that the model always maintains high prediction performance.

[0040] Continuous improvement of prediction models: The reinforcement learning unit uses incremental data samples to train the prediction model in the model database, enabling the model to continuously learn and adapt to new data features, thereby continuously improving the prediction results. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings required in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a framework diagram of a smart electricity information management system according to an embodiment of the present invention. Detailed Implementation

[0043] The detailed description of the accompanying drawings is intended to illustrate the present embodiments of the invention and is not intended to represent only the forms in which the invention can be implemented. It should be understood that the same or equivalent functions can be accomplished by different embodiments intended to be included within the spirit and scope of the invention.

[0044] See Figure 1 This invention provides a smart electricity information management system, comprising:

[0045] Data integration unit 1 is used to acquire external data from different sources in real time and to predict power supply-related emergencies based on external data.

[0046] Specifically, the main function of Feature Engineering Unit 2 is to process raw power data, transforming it into feature inputs that can be used for predictive analysis through preprocessing and feature extraction steps. The specific steps and functions are as follows:

[0047] Obtain power data:

[0048] Data Sources: The Feature Engineering Unit obtains real-time and historical power data from sources such as power system monitoring equipment, smart meters, and historical databases.

[0049] Data types: These data may include voltage, current, power, frequency, electricity consumption, load rate, etc.

[0050] Preprocessing power data:

[0051] Data cleaning: Remove invalid data, outliers, noise, etc., to ensure data quality.

[0052] Data imputation: Handling missing data by filling in missing values ​​using methods such as interpolation, averaging, and regression.

[0053] Data transformation: Converting data into a format suitable for analysis, such as converting date and time into a uniform format, or encoding categorical data into numerical values.

[0054] Data normalization: Standardizing or normalizing data to bring it to the same order of magnitude, making it easier for models to process.

[0055] Building an electricity dataset:

[0056] Dataset formation: The preprocessed data is organized into a structured dataset, usually in tabular form, with each row representing a sample and each column representing a feature.

[0057] Feature extraction:

[0058] Feature selection: Selecting features from the preprocessed data that have a significant impact on the prediction target.

[0059] Feature construction: Based on business knowledge and data analysis, create new features, such as extracting trend, periodic, and seasonal features from time series data.

[0060] Feature transformation: Applying mathematical transformations (such as logarithmic transformations and square root transformations) to improve the distribution or relationship of features.

[0061] Obtain feature input:

[0062] Final output: The dataset after feature engineering, i.e. the feature input, will be provided as input to the subsequent prediction model for training and prediction.

[0063] For example, feature engineering unit 2 might obtain hourly electricity consumption data from smart meters. During the preprocessing stage, it removes abnormal readings caused by meter malfunctions and fills in data gaps due to communication interruptions. Then, it might extract features such as peak electricity consumption periods during the day and differences in electricity consumption between weekdays and weekends, using these features as input to a predictive model to forecast future electricity demand.

[0064] Through the processing of Feature Engineering Unit 2, the intelligent electricity information management system can more accurately capture key information in the electricity data, thereby improving the performance of the prediction model and the reliability of the prediction results. Feature Engineering Unit 2 is used to acquire electricity data, preprocess the electricity data to obtain an electricity dataset, and extract features from the data in the electricity dataset to obtain feature inputs for prediction.

[0065] Specifically, the main function of Feature Engineering Unit 2 is to process raw power data, transforming it into feature inputs that can be used for predictive analysis through preprocessing and feature extraction steps. The specific steps and functions are as follows:

[0066] Obtain power data:

[0067] Data Sources: The Feature Engineering Unit obtains real-time and historical power data from sources such as power system monitoring equipment, smart meters, and historical databases.

[0068] Data types: These data may include voltage, current, power, frequency, electricity consumption, load rate, etc.

[0069] Preprocessing power data:

[0070] Data cleaning: Remove invalid data, outliers, noise, etc., to ensure data quality.

[0071] Data imputation: Handling missing data by filling in missing values ​​using methods such as interpolation, averaging, and regression.

[0072] Data transformation: Converting data into a format suitable for analysis, such as converting date and time into a uniform format, or encoding categorical data into numerical values.

[0073] Data normalization: Standardizing or normalizing data to bring it to the same order of magnitude, making it easier for models to process.

[0074] Building an electricity dataset:

[0075] Dataset formation: The preprocessed data is organized into a structured dataset, usually in tabular form, with each row representing a sample and each column representing a feature.

[0076] Feature extraction:

[0077] Feature selection: Selecting features from the preprocessed data that have a significant impact on the prediction target.

[0078] Feature construction: Based on business knowledge and data analysis, create new features, such as extracting trend, periodic, and seasonal features from time series data.

[0079] Feature transformation: Applying mathematical transformations (such as logarithmic transformations and square root transformations) to improve the distribution or relationship of features.

[0080] Obtain feature input:

[0081] Final output: The dataset after feature engineering, i.e. the feature input, will be provided as input to the subsequent prediction model for training and prediction.

[0082] For example, feature engineering unit 2 might obtain hourly electricity consumption data from smart meters. During the preprocessing stage, it removes abnormal readings caused by meter malfunctions and fills in data gaps due to communication interruptions. Then, it might extract features such as peak electricity consumption periods during the day and differences in electricity consumption between weekdays and weekends, using these features as input to a predictive model to forecast future electricity demand.

[0083] Through the processing of feature engineering unit 2, the smart electricity information management system can more accurately capture key information in the electricity data, thereby improving the performance of the prediction model and the reliability of the prediction results.

[0084] The multi-model fusion unit 3 is used to select a matching combination of prediction models from the model database according to the prediction requirements to perform prediction processing on the feature inputs used for prediction and obtain the prediction results; the combination of prediction models includes at least one prediction model, and the prediction requirements include electricity demand prediction and electricity supply prediction.

[0085] Specifically, the main function of the multi-model fusion unit 3 is to select and combine multiple prediction models from the model database according to specific prediction requirements, perform prediction processing on the feature input after feature engineering, and finally output the prediction result. The specific steps and functions are as follows:

[0086] Predictive demand identification:

[0087] Demand types: Forecasted demand may include electricity demand forecasts (such as electricity consumption forecasts for a future period of time) and electricity supply forecasts (such as power generation and power supply reliability forecasts).

[0088] Demand analysis: Analyze the specific objectives of the forecast, such as short-term, medium-term or long-term forecasts, and the time granularity of the forecast (such as hour, day, month).

[0089] Model database:

[0090] Model storage: The model database stores a variety of prediction models, which may include time series models (such as ARIMA, SARIMA), machine learning models (such as random forest, support vector machine), deep learning models (such as neural networks, recurrent neural networks), etc.

[0091] Model characteristics: Each model has its specific advantages and applicable scenarios, and can exhibit different performance under different data characteristics and prediction needs.

[0092] Selecting a combination of prediction models:

[0093] Model matching: Based on the forecasting requirements, the system selects the most suitable forecasting model or combination of models from the model database. For example, for electricity demand forecasting with obvious seasonality, the SARIMA model may be selected.

[0094] Combination strategies: Predictive model combination can be a simple set of models or a more complex ensemble learning method, such as stacking, boosting, or bagging.

[0095] Predictive processing:

[0096] Model training: The selected model is trained using feature inputs, and model parameters are adjusted to optimize performance.

[0097] Model fusion: This involves combining the prediction results of multiple models. Common fusion methods include weighted averaging, voting mechanisms, or more advanced ensemble techniques.

[0098] The prediction results are as follows:

[0099] Output results: The final output of the multi-model fusion unit 3 is the fused prediction result, which is more accurate and reliable than the prediction result of a single model.

[0100] Suppose we need to predict electricity demand for the next week. Multi-model fusion unit 3 selects a prediction model combination from the model database, consisting of an ARIMA model, a random forest model, and a neural network model. Each model processes the feature input separately, and then the predictions from the three models are fused using a weighted average method to obtain the final electricity demand forecast.

[0101] By applying the multi-model fusion unit 3, the smart electricity information management system can fully utilize the advantages of different models to improve the accuracy of predictions and the adaptability of the system, thereby better supporting the operation and management decisions of the power system.

[0102] The adaptive learning unit 4 is used to construct a prediction model for the prediction requirements through deep learning or to select a prediction model to be adaptively learned from the model database. After adaptively learning and training the constructed or selected prediction model based on environmental changes, sudden events or nonlinear effects, it is added to the model database.

[0103] Specifically, the main function of the adaptive learning unit 4 is to build new prediction models or select models from existing model databases and train them adaptively through deep learning. This allows the prediction model to better adapt to environmental changes, sudden events, or the nonlinear characteristics of data, thereby improving the accuracy of predictions and the flexibility of the system. Specific steps and functions are as follows:

[0104] Forecast demand analysis:

[0105] Analyze current forecasting needs to determine whether a new forecasting model needs to be built or an existing model should be selected for adaptive learning.

[0106] Model building or selection:

[0107] Building a new model: If the prediction requirements are special or the existing model cannot meet the requirements, the adaptive learning unit 4 will build a new prediction model.

[0108] Selecting existing models: If suitable models are already available in the model database, Unit 4 will select these models for subsequent adaptive learning.

[0109] Adaptive learning training:

[0110] Data preparation: Collect the latest data, including historical data on environmental changes and emergencies, as well as other variables relevant to the forecast.

[0111] Training process: The selected model is trained using deep learning algorithms. Deep learning algorithms can learn complex patterns and relationships from large amounts of data, enabling the model to adapt to nonlinear effects.

[0112] Adjustment and optimization: Adjust model parameters and optimize model structure based on training results to improve the model's predictive performance under new environmental conditions.

[0113] Model update:

[0114] The adaptively trained model is updated back into the model database to ensure that the models in the database always maintain the latest learning and adaptation status.

[0115] Model database update:

[0116] Add a new model: If a new model is built, add it to the model database.

[0117] Update existing models: If adaptive learning has been performed on an existing model, then update the corresponding model record in the database.

[0118] Suppose the power system has experienced a rare natural disaster, leading to a significant change in power supply patterns. Adaptive learning unit 4 selects relevant predictive models from the model database and trains these models using new data incorporating the disaster's impact. After training, the updated models are better able to predict post-disaster power demand and supply, and are thus added back to the model database for future use.

[0119] Through the adaptive learning unit 4, the smart electricity information management system can continuously improve and optimize the prediction model, ensuring the accuracy of the prediction results and the long-term effectiveness of the system.

[0120] The reinforcement learning unit 5 is used to acquire incremental data samples and perform reinforcement learning training on multiple prediction models in the model database based on the incremental data samples.

[0121] Specifically, the main function of reinforcement learning unit 5 is to train the predictive model in the model database using newly collected data samples (i.e., incremental data samples). This training method can gradually improve the model's predictive ability, enabling it to better adapt to changes and trends in the data. The specific steps and functions are as follows:

[0122] Obtain incremental data samples:

[0123] Incremental data samples refer to data newly collected after the model's last training iteration. This data may contain new market trends, changes in user behavior, technological advancements, and other factors that have a significant impact on the model's predictive performance.

[0124] Data preparation:

[0125] Incremental data samples are preprocessed to ensure data quality and consistency, preparing them for reinforcement learning training.

[0126] Enhance learning and training:

[0127] Model selection: Select a prediction model from the model database that requires reinforcement learning.

[0128] Training process: The selected model is trained using a reinforcement learning algorithm. Reinforcement learning is a method that learns optimal behavioral policies by interacting with the environment, and it can learn even without a complete data distribution.

[0129] Strategy Update: Adjust the model parameters and prediction strategy based on incremental data samples to improve the model's prediction performance.

[0130] Model evaluation:

[0131] During reinforcement learning training, the model's performance is evaluated regularly to ensure that the training direction is correct and the model is improved.

[0132] Model update:

[0133] The model trained through reinforcement learning is updated back into the model database, ensuring that the models in the database always maintain the latest learning status.

[0134] Suppose a new demand pattern emerges in the electricity market, such as increased evening electricity demand due to the proliferation of electric vehicles. Reinforcement learning unit 5 collects incremental data samples during this period and uses this data to train the electricity demand forecasting model in the model database using reinforcement learning. Through this training, the model can learn the new demand pattern and take these changes into account in future forecasts.

[0135] By applying reinforcement learning unit 5, the smart electricity information management system can not only utilize historical data but also learn from the latest data, thereby continuously improving the accuracy and adaptability of its predictive models. This is crucial for the reliable operation and effective management of the power system.

[0136] Furthermore, the data integration unit includes:

[0137] The real-time data acquisition module is used to acquire external data from different sources in real time; the external data from different sources includes meteorological data, market demand data, and socio-economic activity data.

[0138] The time prediction module is used to analyze the external data based on a time-driven prediction method to predict whether a sudden event will occur, and to trigger an early warning mechanism to issue an early warning notification when a sudden event occurs.

[0139] Specifically, the data integration unit is a crucial component of the intelligent electricity information management system. It is responsible for collecting and processing various external data, enabling the system to conduct in-depth analysis and predict potential emergencies. The following is a detailed explanation of the real-time data acquisition module and the time prediction module within the data integration unit:

[0140] Real-time data acquisition module:

[0141] Function: The main task of the real-time data acquisition module is to collect external data from multiple sources in real time. This data is crucial for predicting power supply conditions and potential risks.

[0142] Data source:

[0143] Meteorological data, including temperature, humidity, wind speed, rainfall, and weather conditions, can affect electricity demand and supply. For example, extreme weather can damage power facilities.

[0144] Market demand data: This includes changes in electricity market demand, such as electricity consumption, peak load, and demand data for different user types (e.g., residential, commercial, industrial).

[0145] Socioeconomic activity data: This includes information such as population growth, economic development indicators, holiday schedules, and major events, which may influence electricity demand patterns.

[0146] Function: By acquiring this data in real time, the system can promptly understand external factors that may affect the power system, providing a basis for subsequent analysis and prediction.

[0147] Time prediction module:

[0148] Function: The time prediction module uses a time-driven prediction method to analyze external data in real time to predict whether a sudden event will occur, and activates an early warning mechanism when a sudden event is detected.

[0149] Prediction methods:

[0150] This module may use techniques such as time series analysis, trend analysis, and periodic analysis to predict the changing trends of data.

[0151] By setting thresholds and rules, the module can identify abnormal patterns in the data that may foreshadow impending emergencies.

[0152] Early warning mechanism:

[0153] When the predictive model analysis results indicate that a sudden event (such as a surge in electricity demand or supply disruption) may occur, the module will trigger an early warning mechanism.

[0154] Warning notifications may include sending alerts to power system operators, displaying warning messages on the system interface, and notifying relevant personnel via SMS or email.

[0155] Function: The time prediction module is used to identify potential power supply risks in advance so that power companies can take preventative measures to reduce the impact of emergencies on the power system.

[0156] In summary, the data integration unit, through the collaborative work of the real-time data acquisition module and the time prediction module, provides the smart electricity information management system with real-time monitoring and prediction capabilities of the external environment, thereby improving the reliability and emergency response capabilities of the power system.

[0157] Furthermore, the feature engineering unit includes:

[0158] The timeliness adjustment module is used to obtain power data within the most recent preset time period;

[0159] The data augmentation module is used to generate synthetic data based on the power data obtained by the timeliness adjustment module, so as to increase the data samples used for model training; the synthetic data is a data sample similar to the obtained power data generated by the simulation model, or a data sample simulating a sudden event or abnormal situation.

[0160] The preprocessing module is used to preprocess the power data obtained by the timeliness adjustment module to obtain a power dataset by removing missing values, outliers, and duplicate data, and to extract features from the power dataset to obtain feature inputs for prediction; it is also used to extract features from the synthetic data generated by the data augmentation module to obtain feature inputs for training the model.

[0161] Specifically, the feature engineering unit is a key component of the smart electricity information management system. It is responsible for processing and transforming raw electricity data to provide high-quality input features for predictive models. The following is a detailed explanation of each module within the feature engineering unit:

[0162] Timeliness adjustment module:

[0163] Function: The purpose of the timeliness adjustment module is to obtain the most recent preset time period of power data from the data source. This data is usually real-time or near real-time, and for predictive models, it contains the latest information, which is crucial for capturing the current state and recent trends of the power system.

[0164] Function: By acquiring data from the most recent time period, the module ensures the relevance and timeliness of the data, which is crucial for predicting upcoming events or changes in state.

[0165] Data Augmentation Module:

[0166] Function: The data augmentation module generates synthetic data based on the power data acquired by the timeliness adjustment module. This synthetic data is used to increase the amount of data samples for model training and improve the model's generalization ability.

[0167] Synthetic data generation:

[0168] Simulation Model: Using simulation models to generate data samples that are similar to the actual power data obtained, these samples can simulate the behavior of the power system under normal operating conditions.

[0169] Simulate sudden events or abnormal situations: Generate data samples simulating sudden events (such as power facility failures or extreme weather events) or abnormal situations (such as sudden fluctuations in power demand) to train the model's predictive ability under these conditions.

[0170] Function: Data augmentation allows models to be trained in a wider range of contexts, thereby improving their robustness and accuracy in real-world applications.

[0171] Preprocessing module:

[0172] Function: The preprocessing module is responsible for cleaning and transforming the raw power data obtained by the timeliness adjustment module, as well as extracting features, in order to generate feature inputs suitable for model training and prediction.

[0173] Data cleaning:

[0174] Missing value handling: Fill or remove missing values ​​from the dataset.

[0175] Outlier handling: Identify and handle outliers in the dataset, which may be caused by measurement errors or system malfunctions.

[0176] Duplicate data processing: Remove duplicate records from the dataset.

[0177] Feature extraction:

[0178] Feature extraction is performed on the cleaned power dataset, such as extracting statistical features (mean, variance, peak value, etc.), frequency features, or other advanced features of the time series.

[0179] The same feature extraction process is performed on the synthetic data generated by the data augmentation module to ensure the consistency of the training data.

[0180] Function: The preprocessing module ensures that the data input into the model is clean, consistent, and predictive, which is crucial for training an efficient and reliable prediction model.

[0181] Through the collaborative work of these three modules, the feature engineering unit can provide high-quality feature inputs for the prediction model of the smart electricity information management system, thereby improving the prediction performance and practicality of the entire system.

[0182] Furthermore, the multi-model fusion unit includes:

[0183] Select the module and choose the best combination of models based on the historical performance of the models and the current data distribution of the feature inputs;

[0184] The learning method fusion module is used to fuse multiple prediction models of the optimal model combination by weighting or stacking.

[0185] The outlier handling module is used to automatically identify and correct anomalies or deviations in the prediction results of the prediction model.

[0186] Specifically, the multi-model fusion unit is an advanced processing module in the smart electricity information management system. It improves the accuracy and reliability of predictions by combining the advantages of multiple prediction models. The following is a detailed explanation of each module within the multi-model fusion unit:

[0187] Select module:

[0188] Function: The selection module's responsibility is to choose the optimal model combination based on the model's historical performance and the current feature input data distribution. This selection process is based on the model's past performance and the characteristics of the current data to ensure that the selected model is best suited to the current prediction needs.

[0189] Selection criteria:

[0190] Historical performance of the model: Consider the model's performance on historical data, including indicators such as prediction accuracy, stability, and computational efficiency.

[0191] Current feature input data distribution: Analyze the characteristics of the current data, such as distribution pattern and feature range, to determine which models may be more suitable for processing this data.

[0192] Function: The selection module ensures that the system uses a validated combination of models that is suitable for the characteristics of the current data, thereby improving the accuracy of predictions.

[0193] Learning Methods Integration Module:

[0194] Function: The learning method fusion module is responsible for fusing the selected best model combination. The purpose of fusion is to combine the advantages of different models to generate a more accurate and robust prediction result.

[0195] Fusion method:

[0196] Weighting: Based on the historical performance of each model or some evaluation criterion, different weights are assigned to the prediction results of each model, and then these results are weighted and averaged.

[0197] Stacking: A hierarchical model ensemble technique that first uses multiple different models to make predictions, and then uses another model (usually a simple linear model) to combine these predictions.

[0198] Function: The learning method fusion module improves the stability and accuracy of prediction results through effective model fusion strategies.

[0199] Outlier handling module:

[0200] Function: The outlier handling module automatically identifies and corrects anomalies or biases in the prediction model results. These outliers may be caused by inaccurate model predictions, data quality issues, or rare events that actually occur.

[0201] Solution:

[0202] Identification: Using statistical methods or machine learning techniques to detect outliers in the prediction results.

[0203] Correction: Correcting identified outliers may be done by replacing, adjusting weights, or rerunning the prediction model.

[0204] Function: The outlier handling module ensures the reliability and usability of the prediction results and reduces the impact of erroneous predictions on power system operation decisions.

[0205] In summary, the multi-model fusion unit improves the overall predictive performance of the smart electricity information management system by selecting the optimal model combination, fusing the prediction results of different models, and handling outliers in the prediction, thus providing strong support for the stable operation and optimized scheduling of the power system.

[0206] Furthermore, the multi-model fusion unit includes:

[0207] The deep neural network module is used to construct a prediction model for the prediction requirements through deep learning or to select a prediction model to be adaptively learned from the model database.

[0208] The adaptive algorithm module is used to adaptively learn and train the constructed or selected prediction model based on environmental changes, sudden events, or nonlinear effects, and then add it to the model database.

[0209] Specifically, the deep neural network module and the adaptive algorithm module in the multi-model fusion unit are two key components that work together to improve the predictive capabilities of the smart electricity information management system. The following is a detailed explanation of these two modules:

[0210] Deep Neural Network Module:

[0211] Function: The Deep Neural Network module utilizes deep learning techniques to build new predictive models or select existing predictive models from a model database for further training and learning. Deep Neural Networks (DNNs) are complex neural network structures that can learn deep features of data through multi-layer processing.

[0212] Building Predictive Models: This module can design and train one or more deep neural network models that can learn complex patterns and relationships from large amounts of electricity data for use in electricity demand forecasting, electricity supply forecasting, and more.

[0213] Model selection: The module can also select pre-trained deep neural network models from the model database. These models may already be suitable for specific prediction tasks, but need to be adaptively learned based on new data or conditions.

[0214] Adaptive algorithm module:

[0215] Function: The adaptive algorithm module is responsible for adaptively training the prediction model built or selected by the deep neural network module. This training is to enable the model to adapt to environmental changes, sudden events, or the nonlinear effects of data.

[0216] Adaptive learning training:

[0217] Environmental changes: When the external environment (such as weather, economic conditions, etc.) changes, the model needs to adjust its parameters to maintain the accuracy of the prediction.

[0218] Unexpected events: When unexpected events occur (such as natural disasters, power grid failures, etc.), the model needs to adapt quickly to these abnormal situations in order to provide accurate predictions.

[0219] Nonlinear effects: Power systems often have nonlinear characteristics, and adaptive learning can help models better capture these complex nonlinear relationships.

[0220] Update the model database: Models trained through adaptive learning will be added back to the model database so that they can be quickly called and used in future prediction tasks.

[0221] Through the collaboration of these two modules, the multi-model fusion unit ensures that the predictive models in the intelligent electricity information management system are always in optimal condition, effectively handling various complex situations and improving prediction accuracy and system robustness. This is of great significance for the stable operation of the power system, energy management, and emergency response.

[0222] Furthermore, it also includes:

[0223] A mobile terminal is used to log in, log out, and operate the information management system.

[0224] The mobile terminal integrates an archived work order data upload module and a query module. The archived work order data upload module is used to store and display the work orders that have been archived by the business, and the query module is used to query the archived work orders.

[0225] Specifically, in the smart electricity information management system, the addition of mobile terminals provides greater flexibility and convenience for system operation. The following is a detailed explanation of the mobile terminals and their integrated archived work order data upload and query modules:

[0226] Mobile terminal:

[0227] Function: Mobile terminals refer to portable devices such as smartphones and tablets, which can remotely connect to the smart electricity information management system. Users can log in to the system through mobile terminals to perform operations such as viewing data, executing forecasting tasks, and managing work orders, and can also log out of the system after completing their operations.

[0228] Login / Logout: Users log in to the system by entering their username and password or other authentication methods (such as QR code scanning, fingerprint recognition, etc.). Logout is the process of securely exiting the system after the user has completed their actions, in order to protect the security of their account information.

[0229] Archived work order data upload module:

[0230] Function: The archived work order data upload module is a feature on the mobile terminal that allows users to upload archived work order data from business expansion (i.e., power business expansion) to the information management system. These work orders may include records of power facility installation, maintenance, inspection, etc.

[0231] Storage and Display: Uploaded work order information is stored in the system and can be displayed to authorized users. This helps maintain the integrity and traceability of work order records, facilitating future auditing and retrieval.

[0232] Query module:

[0233] Function: The query module allows users to search and query archived work orders on mobile devices. Users can find specific work order records by entering keywords, work order numbers, user names, and other information.

[0234] Search Results: Search results typically display detailed information about the work order, including attachment details, work order number, user name, user ID, and electricity address. Users can click on attachment details to view attachments in the work order's archiving process, such as photos and documents.

[0235] By integrating these modules onto mobile terminals, the smart electricity information management system provides field staff with convenient data entry and query tools, improving work efficiency and data real-time performance. This is of great significance for power companies in areas such as customer service, on-site management, and decision support.

[0236] Furthermore, the query conditions of the query module are: power supply unit, work order number, user number, and user name;

[0237] The query module outputs the following query results: attachment details, work order number, user name, user number, and electricity address. When the user clicks on the attachment details, a pop-up window displays the attachments for the business expansion archiving stage under this work order, which corresponds to the attachment management of the business expansion archiving stage in the marketing system.

[0238] Specifically, in the intelligent electricity information management system, the query module is a tool used to retrieve and display work order information. The following is a detailed explanation of the query conditions and output results of the query module:

[0239] Search criteria:

[0240] Power supply unit: One of the query conditions, allowing users to filter work order records based on the power supply unit (such as a substation or power supply area).

[0241] Work order number: Another search criterion, users can search for specific work orders by entering a specific work order number.

[0242] User ID: Users can use their user ID as a search criterion to find all work orders related to a specific user.

[0243] Username: Users can also query work orders by entering their username, which is useful when a user knows the name but is unsure of the work order number.

[0244] These query criteria can be used individually or in combination to accurately locate the work order information that the user needs.

[0245] Query results:

[0246] Attachment Details: One of the search results allows users to click and view attachments related to the work order, such as documents, images, and tables.

[0247] Work Order Number: The unique number of the work order displayed in the query results makes it easy for users to confirm the work order information found.

[0248] Username: Displays the user name corresponding to the work order, helping users identify the owner of the work order.

[0249] User ID: The query results include a user ID, which provides another form of identification for the user.

[0250] Electricity address: Displays the user's electricity address, which is very important for on-site staff to perform on-site operations.

[0251] Attachment details:

[0252] When a user clicks "Attachment Details" in the search results, the system will pop up a window displaying all attachments for this work order's archiving process. These attachments may include engineering drawings, acceptance reports, maintenance records, etc., and are important documentary evidence during the work order execution process.

[0253] The attachment management in the business expansion archiving stage of the marketing system corresponds to this: This means that the query module is tightly integrated with the business expansion archiving stage of the marketing system, and users can directly access and manage the attachments of work orders in the information management system without having to switch to other systems.

[0254] In this way, the query module not only provides powerful search functionality, but also ensures the integrity and accessibility of work order information, thereby improving the work efficiency and customer service level of power companies.

[0255] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A smart electricity information management system, characterized in that, include: The data integration unit is used to acquire external data from different sources in real time and to predict power supply-related emergencies based on the external data. The feature engineering unit is used to acquire power data, preprocess the power data to obtain a power dataset, and extract features from the power dataset to obtain feature inputs for prediction. The multi-model fusion unit is used to select a matching combination of prediction models from the model database according to the prediction requirements, perform prediction processing on the feature inputs used for prediction, and obtain the prediction results; the combination of prediction models includes at least one prediction model, and the prediction requirements include electricity demand prediction and electricity supply prediction. An adaptive learning unit is used to construct a prediction model for the prediction requirements through deep learning or to select a prediction model to be adaptively learned from the model database. After adaptively training the constructed or selected prediction model based on environmental changes, sudden events or nonlinear effects, it is added to the model database. The reinforcement learning unit is used to acquire incremental data samples and perform reinforcement learning training on multiple prediction models in the model database based on the incremental data samples. The multi-model fusion unit includes: Select the module and choose the best combination of models based on the historical performance of the models and the current data distribution of the feature inputs; The learning method fusion module is used to fuse multiple prediction models of the optimal model combination by weighting or stacking. The outlier handling module is used to automatically identify and correct anomalies or biases in the prediction results of the prediction model. The multi-model fusion unit includes: The deep neural network module is used to construct a prediction model for the prediction requirements through deep learning or to select a prediction model to be adaptively learned from the model database. The adaptive algorithm module is used to adaptively learn and train the constructed or selected prediction model based on environmental changes, sudden events, or nonlinear effects, and then add it to the model database.

2. The system according to claim 1, characterized in that, The data integration unit includes: The real-time data acquisition module is used to acquire external data from different sources in real time; the external data from different sources includes meteorological data, market demand data, and socio-economic activity data. The time prediction module is used to analyze the external data based on a time-driven prediction method to predict whether a sudden event will occur, and to trigger an early warning mechanism to issue an early warning notification when a sudden event occurs.

3. The system according to claim 2, characterized in that, The feature engineering unit includes: The timeliness adjustment module is used to obtain power data within the most recent preset time period; The data augmentation module is used to generate synthetic data based on the power data obtained by the timeliness adjustment module, so as to increase the data samples used for model training; the synthetic data is a data sample similar to the obtained power data generated by the simulation model, or a data sample simulating a sudden event or abnormal situation. The preprocessing module is used to preprocess the power data obtained by the timeliness adjustment module to obtain a power dataset by removing missing values, outliers, and duplicate data, and to extract features from the power dataset to obtain feature inputs for prediction; it is also used to extract features from the synthetic data generated by the data augmentation module to obtain feature inputs for training the model.

4. The system according to claim 3, characterized in that, Also includes: A mobile terminal is used to log in, log out, and operate the information management system.

5. The system according to claim 4, characterized in that, The mobile terminal integrates an archived work order data upload module and a query module. The archived work order data upload module is used to store and display the work orders that have been archived by the business, and the query module is used to query the archived work orders.

6. The system according to claim 5, characterized in that, The query criteria for the query module are: power supply unit, work order number, user number, and user name; The query module outputs the following query results: attachment details, work order number, user name, user number, and electricity address. When the user clicks on the attachment details, a pop-up window displays the attachments for the business expansion archiving stage under this work order, which corresponds to the attachment management of the business expansion archiving stage in the marketing system.

Images