Load adaptability adjusting method and system for dual-voltage conversion dry-type power transformer
By building a load prediction model and emergency protection mechanism, intelligent load adjustment of dual-voltage conversion dry power transformer is solved, and the problem that traditional power transformer regulation cannot cope with grid load fluctuations is improved, and the efficiency and stability of the power system are improved.
Patent Information
- Application Number
- CN202411250217.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional power transformers lack intelligent and real-time response capabilities, and cannot effectively respond to complex load fluctuations in the power grid and rapid changes in power demand, making it difficult to ensure the efficiency and stability of the power system.
The load adaptability adjustment method of a dual-voltage conversion dry power transformer is adopted. By collecting grid status data and operating parameters, a load prediction model is constructed, the adjustment strategy is optimized, and the emergency protection mechanism is combined to achieve intelligent adjustment of the power transformer.
It enhances the adaptability of the power transformer to load changes, improves the operating efficiency and stability of the power system, avoids adjustment deviations caused by a single factor, and achieves early response and risk prediction.
Smart Images

Figure CN119093358A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of load adaptability adjustment, and in particular to a load adaptability adjustment method and system for a dual voltage conversion dry-type power transformer. Background Art
[0002] With the rapid development of renewable energy technology, the application of distributed photovoltaic power generation systems is becoming more and more widespread. Compared with traditional centralized power systems, the power output of distributed power generation is more volatile, especially photovoltaic power generation is affected by factors such as weather and light, resulting in frequent changes in grid load. This puts higher requirements on the transformers used in the power system, especially in the application scenarios of dual voltage conversion dry-type power transformers, which require transformers to be able to flexibly adapt to load changes to ensure the stability and efficiency of power supply quality.
[0003] Traditional power transformer regulation usually relies on static configuration or manual intervention, lacks the ability of intelligence and real-time response, and cannot effectively cope with complex load fluctuations and rapid changes in power demand in the power grid. At the same time, existing transformer load adjustment strategies usually only rely on historical data or fixed rules, lack a comprehensive assessment of the power grid state and accurate prediction of future loads, resulting in the difficulty in optimizing the efficiency and stability of the power system. Therefore, a load adaptive adjustment method for dual voltage conversion dry-type power transformers is proposed, which has important practical significance. Summary of the invention
[0004] The present invention provides a load adaptability adjustment method and system for a dual voltage conversion dry-type power transformer, which can effectively solve the problems in the background technology.
[0005] In order to achieve the above object, the technical solution adopted by the present invention is: A method for load adaptability adjustment of a dual voltage conversion dry-type power transformer, the method comprising: Collecting grid status data and power transformer operating parameter information; Evaluate the state of the power grid according to the state data to obtain a power grid state evaluation result; Constructing a load prediction model and selecting a suitable optimization method to optimize the load prediction model; Taking the operating parameter information as input, predicting the load of the power transformer through the optimized load prediction model to obtain a load prediction result; Formulate a transformer adjustment strategy and adjust the power transformer according to the grid state assessment result and the load prediction result.
[0006] Furthermore, it also includes setting an emergency protection mechanism in the power transformer, setting a multi-level safety threshold for the emergency protection mechanism, and monitoring the power transformer based on the emergency protection mechanism.
[0007] Furthermore, the collected status data and operating parameter information are transmitted to the central control system of the power grid via a communication network, and the central control system stores and analyzes the transmitted data.
[0008] Further, the state of the power grid is evaluated according to the state data to obtain a power grid state evaluation result, including: Preprocessing the status data; Calculating and evaluating key indicators of the power grid based on the preprocessed state data; Based on the calculated key indicators, a comprehensive evaluation is performed on the state of the power grid to obtain a power grid state evaluation result.
[0009] Furthermore, building a load prediction model includes: Collecting historical operating parameter information, historical load conditions and corresponding time characteristics of the power transformer in the central control system; Normalizing the historical operating parameter information and historical load conditions of the power transformer, and merging them with the time features to construct a time series sample data set; Designing the structure of the load prediction model, and training and optimizing the load prediction model through the time series sample data set; The optimized load prediction model is deployed to the central control system of the power grid.
[0010] Furthermore, the load prediction model structure constructed includes: Input layer, receiving time series sample data; A GRU layer, used for processing the time series sample data and capturing the time dependency in the time series sample data; A dropout layer is added between the GRU layer and the fully connected layer to randomly drop some neurons during the training process of the load prediction model; The fully connected layer maps the features output by the GRU layer to a low-dimensional space for further feature combination and transformation; The output layer generates a final load prediction value according to the output result of the fully connected layer.
[0011] Furthermore, training and optimizing the load prediction model includes selecting a suitable hyperparameter optimization method to optimize the hyperparameters of the load prediction model, including: Determine the search range of hyperparameters and construct a hyperparameter search space; Setting a random search strategy to perform a random search in the hyperparameter search space to obtain an optimal hyperparameter combination; Determining a fine-grained search range according to the optimal hyperparameter combination; Based on the fine-grained search range, the fine-grained optimization is performed through grid search to obtain an optimized hyperparameter combination.
[0012] Furthermore, a termination condition is set during the optimization process of the random search and the grid search.
[0013] A load adaptive adjustment system for a dual voltage conversion dry-type power transformer, the system comprising: A data collection module collects the status data of the power grid and the operating parameter information of the power transformer; A power grid state evaluation module, which evaluates the state of the power grid according to the state data to obtain a power grid state evaluation result; A load prediction model building module is used to build a load prediction model and select a suitable optimization method to optimize the load prediction model; A load prediction module, taking the operating parameter information as input, predicting the load of the power transformer through the optimized load prediction model, and obtaining a load prediction result; The transformer adjustment module formulates a transformer adjustment strategy and adjusts the power transformer according to the grid state assessment result and the load prediction result.
[0014] Furthermore, the power grid status assessment module includes: A data preprocessing unit, which preprocesses the state data; A key indicator calculation unit, which calculates and evaluates the key indicators of the power grid according to the preprocessed state data; The power grid comprehensive evaluation unit performs a comprehensive evaluation on the state of the power grid based on the calculated key indicators to obtain a power grid state evaluation result.
[0015] The technical solution of the present invention can achieve the following technical effects: It effectively enhances the adaptability of power transformers to load changes, combines grid status and load forecasts, avoids adjustment deviations caused by a single factor, and improves the operating efficiency of the overall power system. At the same time, it can effectively foresee potential power system risks, achieve early response and adjustment, and enhance the stability and security of the power system.
[0016] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0018] Figure 1 A flow chart of a method for load adaptability adjustment of a dual voltage conversion dry-type power transformer; Figure 2 A schematic diagram of the process of building a load prediction model; Figure 3 The schematic diagram of the structure of the load adaptive adjustment system of the dual voltage conversion dry-type power transformer is shown in FIG. DETAILED DESCRIPTION
[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.
[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.
[0021] Embodiment 1 like Figure 1 As shown, a load adaptability adjustment method for a dual voltage conversion dry-type power transformer includes: S1: Collecting the state data of the power grid and the operating parameter information of the power transformer; Specifically, data can be collected through a variety of sensors, monitoring equipment, smart meters and other equipment. The collected power grid status data includes: power grid frequency, voltage fluctuation, load status, etc. The collected power transformer operating parameter information includes: voltage, current, temperature, load and other operating parameters, providing reliable input for subsequent load prediction, status assessment and adjustment strategies.
[0022] S2: Evaluate the state of the power grid according to the state data to obtain a power grid state evaluation result; By analyzing the grid operating parameters, we can determine whether the grid is in normal operation, whether there are problems such as overload, undervoltage, frequency fluctuation, etc. The grid status assessment results will serve as the basis for adjusting the transformer, so that the adjustment strategy can better match the actual situation of the grid.
[0023] S3: Build a load prediction model and select a suitable optimization method to optimize the load prediction model; The goal of building a load prediction model in this step is to predict future load changes based on the historical operating data of the transformer and the power grid. Load prediction can help optimize resource allocation, improve the working efficiency of the transformer, and prevent failures caused by excessive power grid load. The load prediction model can be built based on models such as linear regression models, time series models, and deep learning models. For the built model, the model can be trained. The goal of model training is to enable the model to "learn" the mapping relationship between input data and load changes, and improve the performance of the model through appropriate optimization methods, such as gradient descent, regularization, etc.
[0024] S4: using the operating parameter information as input, predicting the load of the power transformer through the optimized load prediction model to obtain a load prediction result; S5: Formulate a transformer adjustment strategy and adjust the power transformer according to the grid status assessment results and load forecast results.
[0025] Through the present invention, the adaptability of the power transformer to load changes is effectively enhanced, the grid status and load prediction are combined, the adjustment deviation caused by a single factor is avoided, and the operation efficiency of the overall power system is improved. At the same time, potential power system risks can be effectively foreseen, early response and adjustment can be achieved, and the stability and safety of the power system are enhanced.
[0026] In order to cope with emergencies of varying degrees, an emergency protection mechanism is also set up in the power transformer, and a multi-level safety threshold is set for the emergency protection mechanism, and the power transformer is monitored based on the emergency protection mechanism.
[0027] On the basis of the above embodiments, in order to cope with emergencies in the power grid or load, an emergency protection mechanism can be added, and a hierarchical management strategy with multi-level safety thresholds can be adopted; sensors or monitoring systems can be used to monitor the key parameters of the transformer and the power grid in real time, and the real-time data of these parameters are continuously compared with the set thresholds, and different response measures are taken in different degrees of abnormal situations. The emergency protection mechanism can respond to emergencies more proactively, ensure the stable operation of the entire power grid system, and prevent possible chain reactions.
[0028] For the definition of multi-level safety thresholds, it can be divided into three levels, namely, the first level threshold, the second level threshold and the third level threshold. Among them, the first level threshold is defined as the critical value of slight abnormality, which means that the system state has deviated from the normal range, but has not yet reached the dangerous level, for example, the load exceeds the rated value by 10%, or the temperature is 5℃ higher than the normal value; when the first level threshold is triggered, slight adjustment measures are taken, for example, the output voltage or power is moderately reduced to relieve the load pressure. The second level threshold represents a more serious abnormal situation, which may cause potential harm to the equipment or system, for example, the load exceeds the rated value by 20%, or the temperature is 10℃ higher than the normal value; when the second level threshold is triggered, stronger emergency response measures are taken, for example, further reducing the output power, or switching or diverting part of the load when necessary. The third level threshold represents a very serious abnormal situation, which requires immediate and comprehensive emergency measures, for example, the load exceeds the rated value by 30%, or the temperature is 15℃ higher than the normal value; when the third level threshold is triggered, comprehensive emergency measures are taken, for example, immediate power-off protection to prevent equipment damage.
[0029] Furthermore, the collected status data and operating parameter information are transmitted to the central control system of the power grid via the communication network, and the central control system stores and analyzes the transmitted data.
[0030] Specifically, the central control system can store the collected grid status data and transformer operating parameter information, and can transmit the data collected from the grid and power transformer to the central control system through some form of communication network, such as optical fiber, wireless network or wired network. These data can be saved in the database of the central control system. At the same time, the central control system has built-in data processing capabilities, which can directly analyze the collected data, run load prediction models, perform grid status assessments, and generate corresponding adjustment strategies. After data processing, the central control system can immediately feed back the adjustment strategy to the transformer control unit for real-time adjustment. This step uploads the grid status and transformer operating data to the central control system, which can uniformly manage the operating status of power equipment over a larger range and improve the accuracy and flexibility of transformer adjustment.
[0031] On the basis of the above-mentioned embodiment, the state of the power grid is evaluated according to the state data to obtain a power grid state evaluation result, including: A1: Preprocess the status data; Specifically, the raw data collected from the power grid may be affected by noise, measurement errors or sensor failures. Therefore, the raw data can be preprocessed, including data cleaning, normalization and denoising. The preprocessed data can improve the accuracy of the analysis and ensure that the calculation and evaluation process is not interfered by the noise and anomalies in the raw data.
[0032] A2: Calculate and evaluate key indicators of the power grid based on the preprocessed status data; A3: Based on the calculated key indicators, a comprehensive assessment is made on the state of the power grid to obtain the grid state assessment result.
[0033] In order to judge the operation status of the power grid, identify the health status of the power grid and potential problems, some key power grid operation indicators can be extracted and calculated through preprocessed data. The key indicators are as follows: (1) Load rate calculation: Calculate the load rate of each transformer and line in the power grid to determine whether the load distribution is balanced; (2) Power flow analysis: Analyze the power flow of each node in the power grid and evaluate the transmission capacity and efficiency; (3) Voltage and frequency stability: Evaluate voltage and frequency fluctuations to ensure they are within normal ranges; (4) Power factor assessment: Evaluate the power factor of the power grid to understand the efficiency of power transmission.
[0034] Based on the key indicators calculated above, a comprehensive evaluation method can be used to conduct an overall assessment of the power grid to determine the health status or potential problems of the power grid, conduct a quantitative analysis of the power grid status, and provide a basis for subsequent transformer adjustments, such as threshold judgment method, weighted scoring method and multi-level evaluation model.
[0035] Furthermore, if Figure 2 As shown, building a load prediction model includes: B1: Collect historical operating parameter information, historical load conditions and corresponding time characteristics of power transformers in the central control system; In this embodiment, the central control system collects the transformer's past operating data, load conditions and time-related characteristics as the basic data for building a load prediction model. The historical operating parameters include voltage, current, temperature and other data. The historical load conditions refer to the load changes of the transformer at different times. The time characteristics may include timestamps, seasons, working days / non-working days, etc., which can help the prediction model identify the temporal changes in the load.
[0036] B2: Normalize the historical operating parameter information and historical load conditions of the power transformer and combine them with the time features to construct a time series sample data set; In this step, data processing tools can be used to normalize data and construct time series. Normalization is a preprocessing of data to eliminate dimensional differences and make the data of each parameter vary within the same range, which helps the model converge faster during training and avoids the impact of too large or too small values on the model. By combining the normalized operating parameter information, historical load conditions and time characteristics, a time series data set can be constructed to capture the change pattern of load over time. Common normalization methods include minimum-maximum normalization and standardization, and the construction of time series samples can use the sliding window method to move the window according to the time sequence to generate training samples.
[0037] B3: Design the structure of the load forecasting model, and train and optimize the load forecasting model through the time series sample data set; Specifically, the load forecasting model is trained using the time series sample data set constructed in the early stage. The training set, validation set, and test set are separated. The model can be learned on the training set and validation set, and its prediction ability can be evaluated on the test set. The training process includes the adjustment of model parameters so that the model can minimize the prediction error. The model parameters can be adjusted using the gradient descent algorithm, Adam optimizer, etc. In addition, the model's learning rate, number of hidden layers, number of neurons and other hyperparameters can be adjusted through hyperparameter optimization methods.
[0038] B4: Deploy the optimized load forecasting model to the central control system of the power grid.
[0039] As a preferred embodiment of this invention, the load prediction model structure constructed includes: Input layer, receiving time series sample data; GRU layer, used to process time series sample data and capture the time dependency in the time series sample data; GRU can capture long-term and short-term time dependencies in time series data, extract time features from transformer load changes, and improve the model's understanding of load change trends. GRU can be more efficient in maintaining memory of time series and can avoid the problem of gradient disappearance. By updating gates and resetting gates to control the transmission of information, it can better handle long time series dependencies.
[0040] Dropout layer: A dropout layer is added between the GRU layer and the fully connected layer to randomly drop some neurons during the training process of the load prediction model. The dropout layer is a regularization technique that randomly drops some neurons during training to prevent the model from overfitting. By dropping some neurons, the model can avoid relying on certain specific features or paths and improve the generalization ability of the model. The dropout layer randomly disables some neurons with a certain probability during training to make the model more robust. When neurons are dropped, their weight updates are temporarily interrupted, but other neurons continue to update, which can prompt the model to learn more general features.
[0041] The fully connected layer maps the features output by the GRU layer to a low-dimensional space for further feature combination and transformation; The fully connected layer can connect the time series features output by GRU with all its own input neurons, perform a weighted sum operation, and then perform a nonlinear transformation through an activation function, which helps the model further process the deep features in the time series data, thereby improving prediction accuracy.
[0042] The output layer generates the final load prediction value based on the output results of the fully connected layer.
[0043] Based on the above embodiment, training and optimizing the load prediction model includes selecting a suitable hyperparameter optimization method to optimize the hyperparameters of the load prediction model, including: C1: Determine the search range of hyperparameters and build a hyperparameter search space; C2: Set a random search strategy and perform random search in the hyperparameter search space to obtain the optimal hyperparameter combination; Specifically, there are some key hyperparameters that need to be optimized when building a model, such as learning rate, batch size, number of neural network layers, etc. Therefore, the upper and lower limits of each hyperparameter can be reasonably determined, which should include possible optimal values while avoiding increasing the amount of calculation due to excessive range. After defining the possible value range and distribution of the parameters, a certain number of parameter combinations are randomly selected for training and evaluation, and the parameter combination with the best performance on the validation set is selected. The random search strategy can be used to perform random search in a wide search space to quickly find a better parameter combination.
[0044] C3: Determine the fine-grained search range based on the optimal hyperparameter combination; C4: Based on the fine-grained search range, fine-grained optimization is performed through grid search to obtain the optimized hyperparameter combination.
[0045] Furthermore, after obtaining the optimal hyperparameter combination, we can use this as a benchmark to make fine-grained adjustments to the hyperparameters, narrow the search range of each hyperparameter, and explore more accurate optimal solutions in a smaller range; in the narrowed hyperparameter search space, use the grid search method to train and evaluate all parameter combinations, and finally select the parameter combination that performs best on the validation set. This step can obtain a more accurate hyperparameter combination through grid search, so that the performance of the model is optimized; in this fine-grained search stage, the grid search can conduct a comprehensive search of a smaller hyperparameter range to ensure that no possible optimization combination is missed.
[0046] In order to ensure that the optimization process of hyperparameters is efficient and reliable, termination conditions are set during the optimization process of random search and grid search.
[0047] In this embodiment, the termination condition is a stopping criterion set during the optimization process, which means that once a specific condition is met, the algorithm will stop searching and will no longer try new hyperparameter combinations. The termination condition includes: iteration limit, time limit, performance improvement threshold, loss function or evaluation indicator convergence, etc. In the load adaptive adjustment method of the dual-voltage conversion dry-type power transformer, the construction and optimization process of the load prediction model requires effective hyperparameter selection. Setting the termination condition can improve the efficiency of the load prediction model optimization process and prevent the model from becoming too complex during the hyperparameter adjustment process, thereby affecting the generalization ability of the model in actual load prediction.
[0048] like Figure 3 As shown, a load adaptive adjustment system for a dual voltage conversion dry-type power transformer includes: A data collection module collects the status data of the power grid and the operating parameter information of the power transformer; A power grid status assessment module assesses the status of the power grid according to the status data and obtains a power grid status assessment result; A load prediction model building module builds a load prediction model and selects a suitable optimization method to optimize the load prediction model; The load prediction module takes the operating parameter information as input, predicts the load of the power transformer through the optimized load prediction model, and obtains the load prediction result; The transformer adjustment module formulates transformer adjustment strategies and adjusts power transformers according to grid status assessment results and load forecast results.
[0049] The above-mentioned adjustment system in the present invention can effectively implement the load adaptability adjustment method of the dual-voltage conversion dry-type power transformer, and the technical effects that can be achieved are as described in the above-mentioned embodiments, which will not be repeated here.
[0050] Furthermore, the power grid status assessment module includes: A data preprocessing unit, which preprocesses the status data; A key indicator calculation unit calculates and evaluates key indicators of the power grid based on the preprocessed status data; The power grid comprehensive evaluation unit conducts a comprehensive evaluation of the power grid status based on the calculated key indicators to obtain the power grid status evaluation results.
[0051] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.
[0052] Although the present application has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined herein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.
Claims
1. A method for load adaptability adjustment of a dual voltage conversion dry-type power transformer, characterized in that: The method comprises: Collecting grid status data and power transformer operating parameter information; Evaluate the state of the power grid according to the state data to obtain a power grid state evaluation result; Constructing a load prediction model and selecting a suitable optimization method to optimize the load prediction model; Taking the operating parameter information as input, predicting the load of the power transformer through the optimized load prediction model to obtain a load prediction result; Formulate a transformer adjustment strategy and adjust the power transformer according to the grid state assessment result and the load prediction result.
2. The load adaptability adjustment method of the dual voltage conversion dry-type power transformer according to claim 1 is characterized in that: It also includes setting an emergency protection mechanism in the power transformer, setting a multi-level safety threshold for the emergency protection mechanism, and monitoring the power transformer based on the emergency protection mechanism.
3. The load adaptability adjustment method of the dual voltage conversion dry-type power transformer according to claim 1 is characterized in that: The collected status data and operating parameter information are transmitted to the central control system of the power grid via a communication network, and the central control system stores and analyzes the transmitted data.
4. The load adaptability adjustment method of the dual voltage conversion dry-type power transformer according to claim 1 is characterized in that: The state of the power grid is evaluated according to the state data to obtain a power grid state evaluation result, including: Preprocessing the status data; Calculating and evaluating key indicators of the power grid based on the preprocessed state data; Based on the calculated key indicators, a comprehensive evaluation is performed on the state of the power grid to obtain a power grid state evaluation result.
5. The load adaptability adjustment method of the dual voltage conversion dry-type power transformer according to claim 3 is characterized in that: Building a load forecasting model includes: Collecting historical operating parameter information, historical load conditions and corresponding time characteristics of the power transformer in the central control system; Normalizing the historical operating parameter information and historical load conditions of the power transformer, and merging them with the time features to construct a time series sample data set; Designing the structure of the load prediction model, and training and optimizing the load prediction model through the time series sample data set; The optimized load prediction model is deployed to the central control system of the power grid.
6. The load adaptability adjustment method of the dual voltage conversion dry-type power transformer according to claim 5, characterized in that: The load prediction model structure constructed includes: Input layer, receiving time series sample data; A GRU layer, used for processing the time series sample data and capturing the time dependency in the time series sample data; A dropout layer is added between the GRU layer and the fully connected layer to randomly drop some neurons during the training process of the load prediction model; The fully connected layer maps the features output by the GRU layer to a low-dimensional space for further feature combination and transformation; The output layer generates a final load prediction value according to the output result of the fully connected layer.
7. The load adaptability adjustment method of the dual voltage conversion dry-type power transformer according to claim 1, characterized in that: Training and optimizing the load prediction model includes selecting a suitable hyperparameter optimization method to optimize the hyperparameters of the load prediction model, including: Determine the search range of hyperparameters and construct a hyperparameter search space; Setting a random search strategy to perform a random search in the hyperparameter search space to obtain an optimal hyperparameter combination; Determining a fine-grained search range according to the optimal hyperparameter combination; Based on the fine-grained search range, the fine-grained optimization is performed through grid search to obtain an optimized hyperparameter combination.
8. The load adaptability adjustment method of the dual voltage conversion dry-type power transformer according to claim 7, characterized in that: A termination condition is set during the optimization process of the random search and the grid search.
9. A load adaptive adjustment system for a dual voltage conversion dry-type power transformer, characterized in that: The system comprises: A data collection module collects the status data of the power grid and the operating parameter information of the power transformer; A power grid state evaluation module, which evaluates the state of the power grid according to the state data to obtain a power grid state evaluation result; A load prediction model building module is used to build a load prediction model and select a suitable optimization method to optimize the load prediction model; A load prediction module, taking the operating parameter information as input, predicting the load of the power transformer through the optimized load prediction model, and obtaining a load prediction result; The transformer adjustment module formulates a transformer adjustment strategy and adjusts the power transformer according to the grid state assessment result and the load prediction result.
10. The load adaptability adjustment system of the dual voltage conversion dry-type power transformer according to claim 9, characterized in that: The power grid status assessment module comprises: A data preprocessing unit, which preprocesses the state data; A key indicator calculation unit, which calculates and evaluates the key indicators of the power grid according to the preprocessed state data; The power grid comprehensive evaluation unit performs a comprehensive evaluation on the state of the power grid based on the calculated key indicators to obtain a power grid state evaluation result.
Citation Information
Patent Citations
Remote intelligent regulation and control system and method for distribution transformer
CN111769548A
Double-layer optimization scheduling decision-making method for regional integrated energy system
CN115374692A
Business prediction method
CN116797273A
Value determination method for hyper-parameter in target prediction model associated with target business
CN116992253A
Method and system for automatically checking safety boundary of main transformer protection setting value
CN118157084A
Cited By
Electric reactor load adjusting method and system based on electrical performance optimization
CN119787385A
Reactor load regulation method and system based on electrical performance optimization
CN119787385B
Amorphous alloy dry type power transformer control method for renewable energy access
CN119813854A
Control Method for Amorphous Alloy Dry-Type Power Transformer for Renewable Energy Access
CN119813854B
