District guarantee power supply optimization system and method based on multivariate data fusion analysis, terminal equipment and storage medium

Abstract

The invention relates to the technical field of smart power grids, in particular to a district power guarantee and supply optimization system and method based on multivariate data fusion analysis, terminal equipment and a storage medium, and the method comprises the steps: obtaining various types of original data from a data platform, the original data including power load, meteorological data, historical operation data, equipment parameters, user load classification and fault work order data; fusing the multivariate data, extracting key features and carrying out deep analysis; performing prediction and classification based on the input data and the analysis result, and making a power supply optimization decision; the power supply state is monitored in real time, and feedback and adjustment suggestions are provided. According to the invention, intelligent scheduling and accurate optimization of transformer area power supply are realized, and the reliability of transformer area power supply is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grids, and is a system, method, terminal equipment and storage medium for optimizing power supply in a substation area based on multivariate data fusion analysis. Background Art

[0002] With the rapid development of smart grid technology, the power system has higher and higher requirements in terms of safety, stability and efficiency. Especially in the distribution network, power supply guarantee has become a key task. As the basic unit of the distribution network, the substation often faces multiple challenges such as load fluctuations, equipment aging, and bad weather, which leads to reduced power supply reliability. Therefore, how to effectively use historical data, equipment parameters, meteorological data, fault work orders, user load classification and other multi-dimensional data to accurately guide the power supply guarantee of the substation is an important technical issue facing the current power industry.

[0003] In the prior art, the power supply management of the substation mainly relies on the power dispatching system and equipment monitoring system, such as the load monitoring and dispatching management system. The traditional load monitoring system usually performs simple load dispatching optimization based on the collected real-time data such as voltage, current, power, etc.; the dispatching system collects equipment data regularly or in real time, analyzes the load status of each substation, and then provides certain dispatching suggestions. However, these systems are mainly based on historical average loads or static load models, lack dynamic load forecasting capabilities, and do not comprehensively consider the impact of multivariate data on load changes. Moreover, the prior art lacks the ability to comprehensively analyze multivariate data (such as equipment parameters, meteorological data, fault work orders, etc.), and cannot accurately evaluate the probability of occurrence of substation loads and related faults; the use of a single algorithm or fixed model leads to insufficient accuracy in load forecasting; it is difficult to accurately predict the reliability of substation power supply in a complex environment, and the reliability of substation power supply is insufficiently predicted. In addition, existing technologies only use data from a single source for load forecasting and fault analysis, and lack the comprehensive use of multiple data, which leads to inaccurate assessments of substation load and fault probability. Traditional load forecasting methods usually rely on fixed models and fail to adapt to dynamically changing environments and load patterns, resulting in insufficient timeliness of forecast results. Traditional power supply solutions lack flexibility and are difficult to adapt to sudden load changes and fault risks, and can often only respond passively.

[0004] Some existing power supply dispatching and substation equipment monitoring systems can perform preliminary analysis through the collected real-time current, voltage, and load data to help the power dispatching center make dispatching decisions. These systems usually use static analysis methods, lack of in-depth analysis of historical data and comprehensive consideration of external factors such as weather and faults. In addition, the existing system is relatively simple in the classification and capacity management of user loads, and cannot implement refined power supply optimization solutions. For example, in the Chinese invention patent application with publication number CN113496315B, a substation load forecasting method based on big data analysis is proposed, which uses historical load data for short-term load forecasting and provides basic load optimization, but fails to integrate information such as equipment parameters, meteorological data, and user load classification, and the prediction results are still insufficient in accuracy and stability. Summary of the invention

[0005] The present invention provides a substation power supply optimization system, method, terminal equipment and storage medium based on multivariate data fusion analysis, which overcomes the shortcomings of the above-mentioned prior art and can effectively solve the problem that the prior art has the singleness and limitations and cannot make full use of multivariate heterogeneous data to comprehensively optimize the substation power supply.

[0006] One of the technical solutions of the present invention is achieved by the following measures: a substation power supply optimization system based on multivariate data fusion analysis, including a data acquisition module, a multivariate data fusion and analysis module, a prediction model application module and a monitoring and feedback module; The data acquisition module is used to obtain various raw data from the data center, including power load, meteorological data, historical operation data, equipment parameters, user load classification and fault work order data; Among them, the multivariate data fusion and analysis module is used to fuse multivariate data, extract key features and conduct in-depth analysis; Among them, the prediction model application module is used to make predictions and classifications based on the input data and analysis results, and make power supply optimization decisions; Among them, the monitoring and feedback module is used to monitor the power supply status in real time and provide feedback and adjustment suggestions.

[0007] The second technical solution of the present invention is achieved by the following measures: a method for optimizing power supply in a substation area based on multivariate data fusion analysis, comprising the following steps: Step S1: Obtain various types of original data from the data center, including power load, meteorological data, historical operation data, equipment parameters, user load classification and fault work order data; Step S2: Fusion of multivariate data, extraction of key features and in-depth analysis; Step S3: Based on the input data and analysis results, prediction and classification are performed to make power supply optimization decisions; Step S4: Monitor the power supply status in real time and provide feedback and adjustment suggestions.

[0008] The following is a further optimization and / or improvement of the second technical solution of the above invention: The above step S1 may specifically include the following steps: Step S1.1, interface call: request data from the data center through the API interface; Step S1.2, data integration: parsing the acquired data in various formats; Step S1.3, data storage: storing the integrated data in a local database.

[0009] The above step S2 may specifically include the following steps: Step S2.1, data cleaning: handle missing values ​​and outliers, and perform cleaning, preprocessing and unified formatting to ensure the consistency of data in time series and structure; Step S2.2, data fusion: fusion of data from different sources to form a comprehensive data set; Step S2.3, feature extraction: extract important features from the data, where important features include equipment operating years, maintenance history, fault work orders, meteorological data, operating data, user load classification, and holiday data; Step S2.4, data analysis: By analyzing the correlation between equipment parameters and meteorological conditions, and the correlation between load types and historical faults, the deep relationship between different data is explored; regression analysis, multivariate regression and interaction effect modeling techniques are used to reveal the complex interactive relationship between multiple factors, providing basic data support for power supply reliability prediction.

[0010] The above step S3 may specifically include the following steps: Step S3.1, data preparation: Based on the pre-processed data output by the multivariate data fusion and association analysis module, further select key features related to the load forecasting and equipment failure warning of the substation area; perform feature interaction, feature combination and add some derived variables according to specific business needs and model requirements to improve the performance of the model; divide the data into training set, validation set and test set; Step S3.2, model construction: based on the specific forecasting requirements and data characteristics, select the appropriate machine learning algorithm; among them, the long short-term memory network can be selected for the load forecast of the substation area, which is suitable for time series data; among them, the random forest is used for equipment failure warning; Step S3.3, model training and validation: input the training set into the model and use the selected algorithm for training; continuously optimize the model parameters so that the model can capture the laws of load fluctuations and equipment failures; adjust the model's hyperparameters through grid search or random search methods; use the validation set to evaluate the model's performance, measure the model's predictive ability, and prevent overfitting; Step S3.4, model optimization: perform feature importance analysis, retain the features that are most helpful for prediction, and reduce model complexity; Step S3.5, prediction and application: predict the future substation load based on the input characteristics of the future time; monitor and predict potential failure risks in real time through equipment status and load data, and issue early warnings; deploy the trained model to the actual system, combine it with real-time data input, and continuously monitor and predict substation load and failure warnings.

[0011] The above step S4 may specifically include the following steps: Step S4.1, real-time monitoring: observe the operation status of the substation in real time through the monitoring system; Step S4.2, feedback mechanism: Establish a feedback mechanism to regularly evaluate the effectiveness of optimization decisions; Step S4.3, Adjustment Suggestions: As data accumulates, further optimization suggestions are made and the model is retrained regularly to maintain its prediction accuracy under changing environments and load patterns.

[0012] The third technical solution of the present invention is achieved through the following measures: a terminal device includes a memory and a processor, the memory stores a program that can be run on the processor, and when the processor executes the program, it implements the above-mentioned substation power supply optimization method based on multivariate data fusion analysis.

[0013] The fourth technical solution of the present invention is achieved through the following measures: a storage medium, the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to realize the above-mentioned substation power supply optimization method based on multivariate data fusion analysis.

[0014] The present invention integrates historical load data, equipment parameters, meteorological data, fault work orders and user load classification data through multivariate data fusion analysis technology to form an accurate power supply plan for a specific substation. The present invention provides a more accurate load forecasting and fault warning mechanism by integrating multiple types of data sources; improves the accuracy and real-time performance of the prediction by introducing an improved machine learning algorithm and multi-factor association analysis; and provides a more accurate power supply plan through multivariate data fusion. The present invention comprehensively considers multiple factors such as equipment parameters, meteorological conditions, fault work orders and user load classification through multi-source data fusion analysis technology; can reveal the complex relationship between various factors, thereby improving the accuracy of load forecasting and fault warning. The present invention adopts an improved machine learning algorithm, combines multiple machine learning algorithms such as random forest and LSTM, and utilizes the advantages of each algorithm to improve the overall prediction accuracy. The present invention can dynamically learn the load mode and continuously optimize the prediction model according to the new data; can timely reflect the load changes and fault risks, and improve the real-time performance of the prediction. By continuously accessing real-time data, the dynamic adjustment mechanism of the present invention can actively generate optimized power supply decisions based on real-time load forecasts and equipment status. This active management not only improves the reliability of power supply, but also reduces the occurrence of power outages. The present invention realizes intelligent scheduling and precise optimization of power supply in the substation area through multivariate data fusion analysis and intelligent decision-making scheme, and effectively improves the reliability of power supply in the substation area. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the overall structure of the substation power supply optimization system based on multivariate data fusion analysis according to an embodiment of the present invention.

[0016] Figure 2 The figure is a schematic diagram of the workflow of the prediction model application module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The present invention is not limited by the following embodiments, and specific implementation methods can be determined based on the technical solution of the present invention and actual conditions.

[0018] The present invention will be further described below in conjunction with embodiments: Embodiment 1: As attached Figure 1 As shown, the substation power supply optimization system based on multivariate data fusion analysis includes a data acquisition module, a multivariate data fusion and analysis module, a prediction model application module and a monitoring and feedback module; The data acquisition module is used to obtain various raw data from the data center, including power load, meteorological data, historical operation data, equipment parameters, user load classification and fault work order data; the implementation steps are as follows: a. Interface call: request data from the data center through the API interface; b. Data integration: Use the pandas library to parse the acquired data in various formats (such as csv, json); c. Data storage: Store the integrated data in the local database.

[0019] Among them, the multivariate data fusion and analysis module is used to fuse multivariate data, extract key features and conduct in-depth analysis; the implementation steps are: a. Data cleaning: handle missing values ​​and outliers, and perform cleaning, preprocessing, and unified formatting to ensure the consistency of data in terms of time series and structure; b. Data fusion: Fusion of data from different sources to form a comprehensive data set; c. Feature extraction: Extract important features from the data, including equipment operating years, maintenance history, fault work orders, meteorological data, operating data, user load classification, and holiday data; d. Data analysis: By analyzing the correlation between equipment parameters and meteorological conditions, and the correlation between load types and historical faults, the deep relationship between different data can be explored; technical means such as regression analysis, multivariate regression and interaction effect modeling are used to reveal the complex interactive relationship between multiple factors, thereby providing basic data support for power supply reliability prediction.

[0020] Among them, the prediction model application module is used to make predictions and classifications based on the input data and analysis results to improve the accuracy of load prediction and the real-time nature of fault warning, and to make power supply optimization decisions; the implementation steps are: a. Data preparation Feature selection: Based on the pre-processed data output by the multivariate data fusion and correlation analysis module, key features related to the load forecast and equipment failure warning of the substation are further selected. The key features include equipment status, meteorological data, holiday data, and historical load data. Feature engineering: Based on specific business needs and model requirements, perform feature interactions, feature combinations, and add some derived variables to improve model performance; for example, weather data can be converted into temperature change rates, and historical load data can be used to extract peak load time periods.

[0021] Data segmentation: Divide the data into training, validation, and test sets to ensure that the model can make accurate predictions on unseen data.

[0022] b. Model construction Algorithm selection: Select the appropriate machine learning algorithm based on specific forecasting requirements and data characteristics. For substation load forecasting, you can choose the long short-term memory network (LSTM), which is suitable for time series data. For equipment fault warning, use random forest, which performs well for high-dimensional data.

[0023] Model architecture design: For deep learning models such as LSTM, design a suitable network architecture, where the network architecture includes the number of layers, number of neurons, and activation function.

[0024] c. Model training and validation Model training: The training set is input into the model and trained using the selected algorithm; by continuously optimizing the model parameters, the model can capture the patterns of load fluctuations and equipment failures.

[0025] Hyperparameter tuning: Adjust the model’s hyperparameters through grid search or random search methods, where the hyperparameters include learning rate, regularization coefficient, and depth of decision tree to obtain the best model performance.

[0026] Model validation: Use the validation set to evaluate the performance of the model, measure the predictive ability of the model through indicators such as mean square error (MSE), accuracy, and AUC, and prevent overfitting.

[0027] d. Model optimization Feature importance analysis: For tree models such as random forests, feature importance analysis is performed to retain the features that are most helpful for prediction and reduce model complexity.

[0028] Model integration: If the prediction effect of a single model is not ideal, you can improve the prediction accuracy by integrating multiple models (such as random forest + LSTM). Integration methods include model weighted averaging, voting method, etc.

[0029] e. Prediction and application Load forecasting: Based on input features of future time (such as weather forecasts, equipment status, etc.), the model can predict future substation loads and help power companies plan power supply plans in advance.

[0030] Fault warning: Through equipment status and load data, the model can monitor and predict potential fault risks in real time, issue warnings, and help operators take preventive measures.

[0031] Model deployment: Deploy the trained model to the actual system, combine it with real-time data input, and continuously monitor and predict substation load and fault warnings.

[0032] The monitoring and feedback module is used to monitor the power supply status in real time and provide feedback and adjustment suggestions. The implementation steps are as follows: a. Real-time monitoring: observe the operation status of the substation in real time through the monitoring system; b. Feedback mechanism: Establish a feedback mechanism to regularly evaluate the effectiveness of optimization decisions; c. Adjustment suggestions: As data accumulates, further optimization suggestions are made and the model is retrained regularly to maintain its prediction accuracy under changing environments and load patterns.

[0033] In this embodiment, the substation is a relatively independent power supply area divided by the power company in the distribution network, which usually includes a distribution transformer and the users within its power supply range, and is an important management and operation unit in the power system. In this embodiment, user load classification is to classify the types of electricity consumption according to the nature of the user's electricity consumption, such as: residential electricity consumption, commercial electricity consumption, agricultural irrigation electricity consumption, etc. In this embodiment, the long short-term memory network (LSTM) is a special recursive neural network that can effectively process and predict time series data, and is particularly suitable for capturing long-term dependencies.

[0034] The substation power supply optimization system based on multivariate data fusion analysis uses multi-source data fusion analysis technology, and utilizes historical data, equipment parameters, meteorological data, and user load classification data to build a comprehensive database; implements correlation analysis between data to identify potential load patterns and abnormal conditions; and uses data preprocessing techniques, such as data cleaning, normalization, and feature selection, to improve data quality and analysis results.

[0035] This substation power supply optimization system based on multivariate data fusion analysis is based on the load forecasting and fault warning mechanism of machine learning algorithms. It uses advanced machine learning algorithms such as random forests and long short-term memory networks (LSTM) to improve the accuracy of load forecasting; builds an equipment fault warning model, combines real-time monitoring data, and promptly identifies potential equipment failures to reduce the risk of power outages; and develops an adaptive learning mechanism. As new data is added, the model can be continuously updated and optimized to improve prediction accuracy.

[0036] The substation power supply optimization system based on multivariate data fusion analysis dynamically adjusts the power supply plan, combines different load types (such as peak, valley and high-frequency loads) and equipment status (such as operating time, maintenance status, etc.) to generate personalized power supply strategies; adjusts the power supply plan through real-time meteorological data (such as temperature, humidity, wind speed, etc.) to cope with the impact of different environmental conditions on the load; formulates a comprehensive decision support system, combines historical data analysis and machine learning model output, and provides a scientific basis for power supply decisions.

[0037] The power supply optimization system for the substation based on multivariate data fusion analysis integrates historical load data, equipment parameters, meteorological data, fault work orders and user load classification data through multivariate data fusion analysis technology to form an accurate power supply plan for a specific substation. The power supply optimization system for the substation based on multivariate data fusion analysis provides a more accurate load forecasting and fault warning mechanism by integrating multiple types of data sources; improves the accuracy and real-time performance of the forecast by introducing improved machine learning algorithms and multi-factor correlation analysis; and provides a more accurate power supply plan through multivariate data fusion (equipment parameters, meteorology, load type, etc.). The power supply optimization system for the substation based on multivariate data fusion analysis comprehensively considers multiple factors such as equipment parameters, meteorological conditions, fault work orders and user load classification through multivariate data fusion analysis technology; it can reveal the complex relationship between various factors, thereby improving the accuracy of load forecasting and fault warning. The power supply optimization system for the substation based on multivariate data fusion analysis adopts an improved machine learning algorithm, combined with multiple machine learning algorithms such as random forest and LSTM, and takes advantage of the advantages of each algorithm to improve the overall prediction accuracy. The power supply optimization system for the substation based on multivariate data fusion analysis can dynamically learn the load pattern and continuously optimize the prediction model according to the new data; it can timely reflect the load changes and fault risks, and improve the real-time nature of the prediction. Through continuous access to real-time data (such as equipment status and meteorological data), the dynamic adjustment mechanism of the power supply optimization system for the substation based on multivariate data fusion analysis can actively generate optimized power supply decisions based on real-time load forecasts and equipment status. This active management not only improves the reliability of power supply, but also reduces the occurrence of power outages. Through multivariate data fusion analysis and intelligent decision-making solutions, intelligent scheduling and precise optimization of substation power supply are realized, effectively improving the reliability of power supply in the substation.

[0038] Embodiment 2: As attached Figure 1 and attached Figure 2 As shown, this embodiment provides a method for optimizing power supply in a substation area based on multivariate data fusion analysis, comprising the following steps: Step S1: Obtain various types of original data from the data center, the original data including power load, meteorological data, historical operation data, equipment parameters, user load classification and fault work order data; in this embodiment, step S1 specifically includes the following steps: Step S1.1, interface call: request data from the data center through the API interface; Step S1.2, data integration: use the pandas library to parse the obtained data in various formats (such as csv, json); Step S1.3, data storage: storing the integrated data in a local database.

[0039] Step S2: Fusing the multivariate data, extracting key features and conducting in-depth analysis; in this embodiment, step S2 specifically includes the following steps: Step S2.1, data cleaning: handle missing values ​​and outliers, and perform cleaning, preprocessing and unified formatting to ensure the consistency of data in time series and structure; Step S2.2, data fusion: fusion of data from different sources to form a comprehensive data set; Step S2.3, feature extraction: extract important features from the data, where important features include equipment operating years, maintenance history, fault work orders, meteorological data, operating data, user load classification, and holiday data; Step S2.4, data analysis: By analyzing the correlation between equipment parameters and meteorological conditions, and the correlation between load types and historical faults, the deep relationship between different data is explored; technical means such as regression analysis, multivariate regression and interaction effect modeling are used to reveal the complex interactive relationship between multiple factors, thereby providing basic data support for power supply reliability prediction.

[0040] Step S3: Based on the input data and analysis results, prediction and classification are performed to improve the accuracy of load prediction and the real-time nature of fault warning, and to make power supply optimization decisions; in this embodiment, step S3 specifically includes the following steps: Step S3.1, data preparation: Based on the pre-processed data output by the multivariate data fusion and association analysis module, further select key features related to the load forecasting and equipment failure warning of the substation, where the key features include equipment status, meteorological data, holiday data, and historical load data; according to specific business needs and model requirements, perform feature interaction, feature combination, and add some derived variables to improve the performance of the model; for example, weather data can be converted into temperature change rate, and historical load data can be used to extract peak load time periods; divide the data into training set, validation set, and test set to ensure that the model can make accurate predictions on unseen data; Step S3.2, model construction: Based on the specific forecasting requirements and data characteristics, select the appropriate machine learning algorithm; among them, the long short-term memory network (LSTM) can be selected for the load forecast of the substation area, which is suitable for time series data; among them, the random forest is used for equipment fault warning, which performs well for high-dimensional data; for deep learning models such as LSTM, design a suitable network architecture, wherein the network architecture includes the number of layers, the number of neurons, and the activation function; Step S3.3, model training and validation: input the training set into the model and use the selected algorithm for training; continuously optimize the model parameters so that the model can capture the laws of load fluctuations and equipment failures; adjust the model's hyperparameters through grid search or random search methods, where the hyperparameters include learning rate, regularization coefficient, and decision tree depth to obtain the best model performance; use the validation set to evaluate the model's performance, measure the model's predictive ability through indicators such as mean square error (MSE), accuracy, and AUC, and prevent overfitting; Step S3.4, model optimization: For tree models such as random forest, feature importance analysis is performed to retain the features that are most helpful for prediction and reduce model complexity; if the prediction effect of a single model is not ideal, the prediction accuracy can be improved by integrating multiple models (such as random forest + LSTM); in this embodiment, the integration method includes model weighted averaging, voting method, etc.; Step S3.5, prediction and application: Predict future substation loads based on input features of future time (such as weather forecasts, equipment status, etc.) to help power companies plan power supply plans in advance; monitor and predict potential failure risks in real time through equipment status and load data, issue warnings, and help operators take preventive measures; deploy the trained model to the actual system, combine it with real-time data input, and continuously monitor and predict substation loads and failure warnings.

[0041] Step S4: Real-time monitoring of power supply status, providing feedback and adjustment suggestions. In this embodiment, step S4 specifically includes the following steps: Step S4.1, real-time monitoring: observe the operation status of the substation in real time through the monitoring system; Step S4.2, feedback mechanism: Establish a feedback mechanism to regularly evaluate the effectiveness of optimization decisions; Step S4.3, Adjustment Suggestions: As data accumulates, further optimization suggestions are made and the model is retrained regularly to maintain its prediction accuracy under changing environments and load patterns.

[0042] The power supply optimization method for the substation based on multivariate data fusion analysis integrates historical load data, equipment parameters, meteorological data, fault work orders and user load classification data through multivariate data fusion analysis technology to form an accurate power supply plan for a specific substation. The power supply optimization method for the substation based on multivariate data fusion analysis provides a more accurate load forecasting and fault warning mechanism by integrating multiple types of data sources; improves the accuracy and real-time performance of the prediction by introducing improved machine learning algorithms and multi-factor correlation analysis; and provides a more accurate power supply plan through multivariate data fusion (equipment parameters, meteorology, load type, etc.). The power supply optimization method for the substation based on multivariate data fusion analysis comprehensively considers multiple factors such as equipment parameters, meteorological conditions, fault work orders and user load classification through multivariate data fusion analysis technology; it can reveal the complex relationship between various factors, thereby improving the accuracy of load forecasting and fault warning. The present invention adopts an improved machine learning algorithm, combines multiple machine learning algorithms such as random forest and LSTM, and utilizes the advantages of each algorithm to improve the overall prediction accuracy. The power supply optimization method for the substation based on multivariate data fusion analysis can dynamically learn the load mode and continuously optimize the prediction model according to the new data; it can timely reflect the load changes and failure risks, and improve the real-time nature of the prediction. By continuously accessing real-time data (such as equipment status and meteorological data), the dynamic adjustment mechanism of the power supply optimization method for the substation based on multivariate data fusion analysis can actively generate optimized power supply decisions according to real-time load prediction and equipment status. This active management not only improves the reliability of power supply, but also reduces the occurrence of power outages. Through multivariate data fusion analysis and intelligent decision-making solutions, intelligent scheduling and precise optimization of substation power supply are realized, effectively improving the reliability of power supply in the substation.

[0043] Embodiment 3: This embodiment provides a terminal device, which includes a memory, a processor, a communication interface and a communication bus. The memory stores a program that can be run on the processor. When the processor executes the program, the substation power supply optimization method based on multivariate data fusion analysis in the above embodiment is implemented.

[0044] The processor may be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, field programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0045] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs, non-transient computer executable programs and units, such as the corresponding program units in the above method embodiments of the present invention. The processor executes various functional applications and data processing of the processor by running the non-transient software programs, instructions and modules stored in the memory, thereby realizing the substation power supply optimization method based on multivariate data fusion analysis in the above embodiment.

[0046] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required by at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. The memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. One or more programs are stored in the memory, and when executed by the processor, the power supply optimization method for the substation based on multivariate data fusion analysis in the above embodiment is executed.

[0047] Embodiment 4: This embodiment provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement a substation power supply optimization method based on multivariate data fusion analysis as in the above embodiment.

[0048] The storage medium may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The storage medium may also be an external storage device of the terminal device, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc. equipped on the terminal device.

[0049] The above technical features constitute the embodiments of the present invention, which have strong adaptability and implementation effect. Non-essential technical features can be added or reduced according to actual needs to meet the requirements of different situations.

Claims

1. A power supply optimization system for substations based on multivariate data fusion analysis, characterized in that It includes data acquisition module, multivariate data fusion and analysis module, prediction model application module and monitoring and feedback module; The data acquisition module is used to obtain various raw data from the data center, including power load, meteorological data, historical operation data, equipment parameters, user load classification and fault work order data; Among them, the multivariate data fusion and analysis module is used to fuse multivariate data, extract key features and conduct in-depth analysis; Among them, the prediction model application module is used to make predictions and classifications based on the input data and analysis results, and make power supply optimization decisions; Among them, the monitoring and feedback module is used to monitor the power supply status in real time and provide feedback and adjustment suggestions.

2. A method for optimizing power supply in substations based on multivariate data fusion analysis, characterized in that The following steps are involved: Step S1: Obtain various types of original data from the data center, including power load, meteorological data, historical operation data, equipment parameters, user load classification and fault work order data; Step S2: Fusion of multivariate data, extraction of key features and in-depth analysis; Step S3: Based on the input data and analysis results, prediction and classification are performed to make power supply optimization decisions; Step S4: Monitor the power supply status in real time and provide feedback and adjustment suggestions.

3. The method for optimizing power supply in substations based on multivariate data fusion analysis according to claim 2 is characterized in that Step S1 specifically includes the following steps: Step S1.1, interface call: request data from the data center through the API interface; Step S1.2, data integration: parsing the acquired data in various formats; Step S1.3, data storage: storing the integrated data in a local database.

4. The method for optimizing power supply in substations based on multivariate data fusion analysis according to claim 2 or 3 is characterized in that Step S2 specifically includes the following steps: Step S2.1, data cleaning: handle missing values ​​and outliers, and perform cleaning, preprocessing and unified formatting to ensure the consistency of data in time series and structure; Step S2.2, data fusion: fusion of data from different sources to form a comprehensive data set; Step S2.3, feature extraction: extract important features from the data, where important features include equipment operating years, maintenance history, fault work orders, meteorological data, operating data, user load classification, and holiday data; Step S2.4, data analysis: By analyzing the correlation between equipment parameters and meteorological conditions, and the correlation between load types and historical faults, the deep relationship between different data is explored; regression analysis, multivariate regression and interaction effect modeling techniques are used to reveal the complex interactive relationship between multiple factors, providing basic data support for power supply reliability prediction.

5. The method for optimizing power supply in substations based on multivariate data fusion analysis according to claim 2 or 3 is characterized in that Step S3 specifically includes the following steps: Step S3.1, data preparation: Based on the pre-processed data output by the multivariate data fusion and association analysis module, further select key features related to the load forecasting and equipment failure warning of the substation area; perform feature interaction, feature combination and add some derived variables according to specific business needs and model requirements to improve the performance of the model; divide the data into training set, validation set and test set; Step S3.2, model construction: based on the specific forecasting requirements and data characteristics, select the appropriate machine learning algorithm; among them, the long short-term memory network can be selected for the load forecast of the substation area, which is suitable for time series data; among them, the random forest is used for equipment failure warning; Step S3.3, model training and validation: input the training set into the model and use the selected algorithm for training; continuously optimize the model parameters so that the model can capture the laws of load fluctuations and equipment failures; adjust the model's hyperparameters through grid search or random search methods; use the validation set to evaluate the model's performance, measure the model's predictive ability, and prevent overfitting; Step S3.4, model optimization: perform feature importance analysis, retain the features that are most helpful for prediction, and reduce model complexity; Step S3.5, prediction and application: predict the future substation load based on the input characteristics of the future time; monitor and predict potential failure risks in real time through equipment status and load data, and issue early warnings; deploy the trained model to the actual system, combine it with real-time data input, and continuously monitor and predict substation load and failure warnings.

6. The method for optimizing power supply in substations based on multivariate data fusion analysis according to claim 2 or 3, characterized in that Step S4 specifically includes the following steps: Step S4.1, real-time monitoring: observe the operation status of the substation in real time through the monitoring system; Step S4.2, feedback mechanism: Establish a feedback mechanism to regularly evaluate the effectiveness of optimization decisions; Step S4.3, Adjustment Suggestions: As data accumulates, further optimization suggestions are made and the model is retrained regularly to maintain its prediction accuracy under changing environments and load patterns.

7. A terminal device, comprising a memory and a processor, wherein the memory stores a program that can be run on the processor, characterized in that: When the processor executes the program, the method for optimizing power supply in substations based on multivariate data fusion analysis as described in any one of claims 2 to 6 is implemented.

8. A storage medium, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the substation power supply optimization method based on multivariate data fusion analysis as described in any one of claims 2 to 6.

Images