Primary current rising protection system of high-capacity transformer

Abstract

The invention discloses a primary current-rising protection system for a high-capacity transformer, which relates to the technical field of power transformation equipment and comprises an intelligent fault prediction module, a current-rising trigger monitoring module, a dynamic load adaptation module, an environmental influence evaluation module, a fault isolation and recovery module, a self-adaptive protection strategy module and a multi-dimensional alarm and response module. According to the primary current rising protection system for the high-capacity transformer, by introducing the intelligent fault prediction module, the dynamic load adaptation module, the environmental influence evaluation module, the self-adaptive protection strategy module and the like, comprehensive monitoring and intelligent protection of the high-capacity transformer can be achieved, and protection measures can be adjusted in real time through cooperative work of all the modules; the safety and reliability of the transformer under complex working conditions are remarkably improved, the occurrence rate of equipment faults is reduced, the self-learning capability is achieved, the protection strategy can be continuously optimized according to actual operation data, and it is ensured that the transformer is always kept in the optimal working state in various operation environments.

Description

Technical Field

[0001] The invention relates to the technical field of power transformation equipment, in particular to a primary current boost protection system for a large-capacity transformer. Background Art

[0002] With the continuous growth of electricity demand, large-capacity transformers play a vital role in the power system. However, transformers face various risks during operation, especially the problem of current rise protection. Most existing protection systems rely on traditional fixed threshold settings, lacking flexibility and intelligence. Although current technology has made some progress in data acquisition and fault detection, it is still insufficient in dynamic load adaptation, environmental impact assessment and adaptive protection strategies. Many systems are unable to analyze and respond to load changes, environmental factors and historical fault data in real time, resulting in the risk of overload in extreme cases. Transformers, thus affecting their safety and reliability. In addition, existing monitoring and protection strategies often lack self-learning capabilities and cannot continuously optimize protection measures based on real-time data. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention provides a primary current boost protection system for a large-capacity transformer to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above-mentioned object, the present invention provides the following technical solutions: a large-capacity transformer primary current boost protection system, comprising an intelligent fault prediction module, a current boost trigger monitoring module, a dynamic load adaptation module, an environmental impact assessment module, a fault isolation and recovery module, an adaptive protection strategy module, and a multi-dimensional alarm and response module; Intelligent fault prediction module uses machine learning algorithms to analyze historical operation data and real-time data to identify potential faults. By building a multi-dimensional feature model, it monitors changes in parameters such as temperature, vibration, and oil level, and uses anomaly detection technology to provide early warning of potential faults. The current boost trigger monitoring module monitors the current boost process of the transformer in real time. Through high-frequency sampling technology, it accurately measures the primary current, monitors the current fluctuation in real time, and sets the current boost threshold. The dynamic load adaptation module analyzes the load characteristics of the transformer in real time and dynamically adjusts the current boost protection strategy. When the load changes dramatically, it identifies the instantaneous load changes and intelligently adjusts the protection parameters, such as the current boost protection time and current limit value. Environmental impact assessment module, integrated with meteorological data acquisition equipment, obtains environmental parameters such as temperature, humidity and wind speed in real time, and performs correlation analysis with transformer operation data; Fault isolation and recovery module, when a current rise fault is detected, the fault isolation procedure is automatically started. The system quickly identifies the fault area and quickly cuts off the power supply of the faulty part through the switch device. At the same time, the fault type and occurrence time are recorded to ensure the normal operation of other equipment; The adaptive protection strategy module learns and analyzes the working status and historical fault data of the transformer in real time, automatically optimizes the protection strategy, and simulates the current and temperature parameters under different working scenarios through the model, and intelligently adjusts the triggering conditions of the current rise protection; The multi-dimensional alarm and response module provides multiple alarm methods, including sound alarm, SMS notification, and mobile phone App push.

[0005] To further optimize the technical solution, in the intelligent fault prediction module, the multidimensional feature model is as follows: ; in, Indicates at a point in time Failure risk index; Indicates at a point in time Temperature parameters; Indicates at a point in time Vibration parameters of Indicates at a point in time Oil level parameters; Indicates at a point in time Current parameters; are the weights of each parameter, reflecting their influence on the failure risk; is a bias term used to adjust the flexibility of the model.

[0006] To further optimize the technical solution, the multidimensional feature model is used to weight the formula and the bias term Determine, including the following process: Data collection: Collect historical data of the transformer under different working conditions, including changes in temperature, vibration, oil level and current, as well as the time and type of fault occurrence; Feature selection and engineering: Analyze the collected data, select the most relevant features using correlation analysis and principal component analysis (PCA), and standardize them to facilitate model training; Model training: Use a labeled data set that contains faults and normal states, and apply machine learning algorithms such as random forests and gradient boosting trees to train the model. Use the back propagation algorithm to continuously optimize weights and biases to minimize the prediction error. Cross-validation: Use k-fold cross-validation to evaluate the accuracy and generalization ability of the model to ensure that the model performs well on unseen data.

[0007] To further optimize the technical solution, the dynamic load adaptation module has a built-in real-time load analysis model, which is based on multiple factors such as load change rate, current limit, and current rise protection time, and is expressed as the following formula: ; in, Indicates at a point in time Dynamic protection parameters such as current rise protection time or current limit value; Indicates at a point in time The actual load value, such as current current; Indicates at a point in time The load change rate is calculated as follows: ; In the above formula, is the time interval; is a coefficient, which indicates the intensity of the load's influence on the protection parameters; is an attenuation factor, reflecting the influence of load change rate on the adjustment of protection parameters; is a bias term used to adjust the baseline value of the model output.

[0008] To further optimize the technical solution, the real-time load analysis model determines the coefficients by the following steps: , and the bias term : Data collection: Collect historical operating data of transformers under different load conditions, including current, load changes and corresponding protection measures; Characteristic analysis: Use statistical analysis methods such as regression analysis to evaluate the relationship between load values ​​and protection parameters to determine the coefficients and ; Model calibration: By simulating different load change scenarios, adjust value, so that the model output meets the actual operation requirements.

[0009] To further optimize the technical solution, the real-time load analysis model includes the following process when used: Real-time load monitoring: The system monitors the load status of the transformer in real time and obtains the current value ; Calculate the load change rate: Use historical data to calculate the load change rate; Dynamically adjust protection parameters: Substitute the current load value and load change rate into the model formula, and , dynamically adjust the current boost protection time and current limit value; Real-time feedback and adjustment: When the load continues to change, repeat the above steps to ensure that the protection parameters always remain at the optimal level to adapt to the current load status; Strategy optimization and learning: Regularly analyze the prediction effect of the model and optimize the model coefficients based on actual fault records and maintenance data , and the bias term , improving the accuracy and response speed of dynamic adjustment.

[0010] To further optimize the technical solution, the environmental impact assessment module establishes an environmental impact model to systematically assess the impact of environmental changes on the upflow protection strategy and automatically adjusts the protection parameters to prevent the risk of equipment overload caused by environmental changes; The environmental impact model combines multiple influencing factors such as temperature, vibration, oil level and current and is constructed using the following formula: ; in, Indicates at a point in time The environmental impact index indicates the degree of environmental impact on upwelling protection strategies; Indicates at a point in time Temperature parameters; Indicates at a point in time Humidity parameters; Indicates at a point in time Wind speed parameters; are the weights of the environmental impact index of temperature, humidity and wind speed, respectively, reflecting the influence of each environmental factor on the protection parameters; is a bias term used to adjust the flexibility of the model; And determine the coefficients by following the steps below and the bias term : Data collection: Collect the operating data of the transformer under different environmental conditions, record the temperature, humidity and wind speed, as well as the corresponding fault conditions and protection parameters; Characteristic analysis: Evaluate the relationship between environmental factors and conservation parameters through statistical analysis methods such as multivariate linear regression to determine the weights ; Model calibration: bias term based on historical data Make adjustments to ensure that the model output meets actual needs.

[0011] To further optimize the technical solution, the environmental impact model, when used, includes the following specific processes: Real-time environmental monitoring: Use meteorological data acquisition equipment to monitor the environmental conditions around the transformer in real time and obtain the temperature ,humidity and wind speed data; Calculate the environmental impact index: Substitute the real-time collected environmental data into the model formula; Analyze environmental impact: Based on the calculated environmental impact index , judge the influence of the current environment on the operation of the transformer. If Exceeding a certain preset threshold , it means that the current environmental conditions pose a risk to the transformer's current-boosting protection strategy; Dynamic adjustment of protection parameters: If the environmental impact index reaches the threshold, the system automatically adjusts the current increase protection strategy, including modifying the current limit value and protection time, to adapt to the current environmental conditions and reduce the risk of overload caused by environmental changes; Continuous monitoring and feedback: The system continuously monitors environmental changes and regularly updates the model’s parameters, retraining based on new data to improve the model’s accuracy and adaptability.

[0012] To further optimize the technical solution, the adaptive protection strategy module has a built-in adaptive protection strategy model, which automatically adjusts the triggering conditions of the current boost protection by performing nonlinear mapping on the current and temperature parameters. The model is as follows: ; in, For at time point The protection strategy output indicates the current boost protection triggering condition, such as the current threshold; The input feature vector contains multiple parameters of current, temperature, and load monitored in real time: ; is a weight matrix, which indicates the influence of each input feature on the protection strategy and needs to be obtained through training; is the bias term, which adjusts the output reference value; As the activation function, ReLU or Sigmoid function is selected to introduce nonlinear characteristics.

[0013] To further optimize the technical solution, the steps of optimizing the protection parameters of the adaptive protection strategy model are as follows: Real-time data monitoring: The system monitors the working status of the transformer in real time and collects current ,temperature ,load data, forming the input feature vector ; Calculate protection strategy output: Input real-time data into the model and calculate protection strategy output; Dynamically adjust protection parameters: according to the protection strategy value output by the model , dynamically adjust the triggering conditions of the current boost protection, such as modifying the current threshold or delaying the triggering time, to adapt to the current working environment and status; Real-time feedback and learning: After the protection measures are implemented, the system continues to monitor the operating status of the transformer, collect feedback data, and if a fault occurs, record the event and use it for subsequent model retraining, optimize the weight matrix and bias terms, and enhance the model's adaptive ability; Policy iteration and optimization: Regularly evaluate model performance, combine new fault data and operation data, and iterate the policy to ensure that the protection measures always meet the actual operating environment.

[0014] Compared with the prior art, the present invention provides a large-capacity transformer primary current boost protection system, which has the following beneficial effects: 1. The large-capacity transformer primary current protection system can realize comprehensive monitoring and intelligent protection of large-capacity transformers by introducing modules such as intelligent fault prediction, dynamic load adaptation, environmental impact assessment and adaptive protection strategy. The coordinated work of various modules enables the protection measures to be adjusted in real time, which significantly improves the safety and reliability of the transformer under complex working conditions and reduces the occurrence rate of equipment failures.

[0015] 2. The large-capacity transformer primary current protection system has self-learning capabilities and can continuously optimize protection strategies based on actual operating data to ensure that the transformer always remains in the best working condition in various operating environments. This not only improves the stability of the power system, but also effectively reduces maintenance costs and extends the service life of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1A structural schematic diagram of a primary current boost protection system for a large-capacity transformer proposed by the present invention; Figure 2 A flow chart for determining the weight and bias of a multi-dimensional characteristic model in a primary current boost protection system for a large-capacity transformer proposed by the present invention; Figure 3 A flow chart for determining the weight and bias of a real-time load analysis model in a primary current boost protection system for a large-capacity transformer proposed by the present invention; Figure 4 A flow chart for determining the weight and bias of an environmental impact model in a primary current boost protection system for a large-capacity transformer proposed by the present invention; Figure 5 The invention discloses a flow chart for determining the weight and bias of an adaptive protection strategy model in a primary current boost protection system for a large-capacity transformer. DETAILED DESCRIPTION

[0017] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example: See also Figure 1 , a large-capacity transformer primary current boost protection system, including an intelligent fault prediction module, a current boost trigger monitoring module, a dynamic load adaptation module, an environmental impact assessment module, a fault isolation and recovery module, an adaptive protection strategy module and a multi-dimensional alarm and response module.

[0019] The intelligent fault prediction module uses machine learning algorithms to analyze historical operating data and real-time data to identify potential faults. By establishing a multi-dimensional feature model, it monitors changes in parameters such as temperature, vibration, and oil level, and uses anomaly detection technology to provide early warning of potential faults.

[0020] In this embodiment, in the intelligent fault prediction module, the multidimensional feature model is as follows: ; in, Indicates at a point in time Failure risk index; Indicates at a point in time Temperature parameters; Indicates at a point in time Vibration parameters of Indicates at a point in time Oil level parameters; Indicates at a point in time Current parameters; are the weights of each parameter, reflecting their influence on the failure risk; is a bias term used to adjust the flexibility of the model.

[0021] like Figure 2 As shown, further, the multidimensional feature model, the weights in the formula and the bias term Determine, including the following process: Data collection: Collect historical data of the transformer under different working conditions, including changes in temperature, vibration, oil level and current, as well as the time and type of fault occurrence; Feature selection and engineering: Analyze the collected data, select the most relevant features using correlation analysis and principal component analysis (PCA), and standardize them to facilitate model training; Model training: Use a labeled data set that contains faults and normal states, and apply machine learning algorithms such as random forests and gradient boosting trees to train the model. Use the back propagation algorithm to continuously optimize weights and biases to minimize the prediction error. Cross-validation: Use k-fold cross-validation to evaluate the accuracy and generalization ability of the model to ensure that the model performs well on unseen data.

[0022] The steps for fault prediction using the above multidimensional feature model are as follows: Real-time data acquisition: The system monitors the transformer’s temperature, vibration, oil level and current data in real time and inputs this data into the model.

[0023] Calculate the failure risk index: Substitute the real-time collected data into the formula.

[0024] Threshold determination: According to historical data and the occurrence of failures, a reasonable failure risk threshold is set. When this threshold is exceeded, the system triggers a fault warning.

[0025] Alerts and responses: Once the threshold is exceeded, the system immediately sends an alert to the operation and maintenance personnel and can automatically adjust the operating parameters of related equipment to prevent failures.

[0026] Continuous learning and optimization: Based on the new data collected subsequently, the model is retrained regularly to optimize weights and biases to improve the accuracy and adaptability of predictions.

[0027] The current boost trigger monitoring module monitors the current boost process of the transformer in real time. Through high-frequency sampling technology, it accurately measures the primary side current, monitors the current fluctuation in real time, and sets the current boost threshold.

[0028] In this embodiment, when the current value exceeds the set threshold, the system immediately triggers an alarm and executes protective measures, which can effectively suppress noise interference and ensure the accuracy of monitoring data.

[0029] The dynamic load adaptation module analyzes the load characteristics of the transformer in real time and dynamically adjusts the current boost protection strategy. When the load changes dramatically, it identifies the instantaneous load changes and intelligently adjusts the protection parameters, such as the current boost protection time and current limit value.

[0030] In this embodiment, the dynamic load adaptation module has a built-in real-time load analysis model, which is based on multiple factors including load change rate, current limit, and current boost protection time, and is expressed as the following formula: ; in, Indicates at a point in time Dynamic protection parameters such as current rise protection time or current limit value; Indicates at a point in time The actual load value, such as current current; Indicates at a point in time The load change rate is calculated as follows: ; In the above formula, is the time interval; is a coefficient, which indicates the intensity of the load's influence on the protection parameters; is an attenuation factor, reflecting the influence of load change rate on the adjustment of protection parameters; is a bias term used to adjust the baseline value of the model output.

[0031] like Figure 3 As shown, further, the real-time load analysis model determines the coefficients by the following steps , and the bias term : Data collection: Collect historical operating data of transformers under different load conditions, including current, load changes and corresponding protection measures; Characteristic analysis: Use statistical analysis methods such as regression analysis to evaluate the relationship between load values ​​and protection parameters to determine the coefficients and ; Model calibration: By simulating different load change scenarios, adjust value, so that the model output meets the actual operation requirements.

[0032] When the real-time load analysis model is used, it includes the following process: Real-time load monitoring: The system monitors the load status of the transformer in real time and obtains the current value ; Calculate the load change rate: Use historical data to calculate the load change rate; Dynamically adjust protection parameters: Substitute the current load value and load change rate into the model formula, and , dynamically adjust the current boost protection time and current limit value; Real-time feedback and adjustment: When the load continues to change, repeat the above steps to ensure that the protection parameters always remain at the optimal level to adapt to the current load status; Strategy optimization and learning: Regularly analyze the prediction effect of the model and optimize the model coefficients based on actual fault records and maintenance data , and the bias term , improving the accuracy and response speed of dynamic adjustment.

[0033] The environmental impact assessment module integrates meteorological data acquisition equipment to obtain environmental parameters such as temperature, humidity, and wind speed in real time, and conducts correlation analysis with the transformer's operating data.

[0034] In this embodiment, the environmental impact assessment module, by establishing an environmental impact model, systematically assesses the impact of environmental changes on the upflow protection strategy and automatically adjusts the protection parameters to prevent the risk of equipment overload caused by environmental changes; The environmental impact model combines multiple influencing factors such as temperature, vibration, oil level and current and is constructed using the following formula: ; in, Indicates at a point in time The environmental impact index indicates the degree of environmental impact on upwelling protection strategies; Indicates at a point in time Temperature parameters; Indicates at a point in time Humidity parameters; Indicates at a point in time Wind speed parameters; are the weights of the environmental impact index of temperature, humidity and wind speed, respectively, reflecting the influence of each environmental factor on the protection parameters; is a bias term used to adjust the flexibility of the model; like Figure 4 As shown, the coefficients are determined by the following steps and the bias term : Data collection: Collect the operating data of the transformer under different environmental conditions, record the temperature, humidity and wind speed, as well as the corresponding fault conditions and protection parameters; Characteristic analysis: Evaluate the relationship between environmental factors and conservation parameters through statistical analysis methods such as multivariate linear regression to determine the weights ; Model calibration: bias term based on historical data Make adjustments to ensure that the model output meets actual needs.

[0035] Furthermore, when the environmental impact model is used, it includes the following specific processes: Real-time environmental monitoring: Use meteorological data acquisition equipment to monitor the environmental conditions around the transformer in real time and obtain the temperature ,humidity and wind speed data; Calculate the environmental impact index: Substitute the real-time collected environmental data into the model formula; Analyze environmental impact: Based on the calculated environmental impact index , judge the influence of the current environment on the operation of the transformer. If Exceeding a certain preset threshold , it means that the current environmental conditions pose a risk to the transformer's current-boosting protection strategy; Dynamic adjustment of protection parameters: If the environmental impact index reaches the threshold, the system automatically adjusts the current increase protection strategy, including modifying the current limit value and protection time, to adapt to the current environmental conditions and reduce the risk of overload caused by environmental changes; Continuous monitoring and feedback: The system continuously monitors environmental changes and regularly updates the model’s parameters, retraining based on new data to improve the model’s accuracy and adaptability.

[0036] The fault isolation and recovery module automatically starts the fault isolation procedure when a current rise fault is detected. The system quickly identifies the fault area and quickly cuts off the power supply of the faulty part through the switch device. At the same time, it records the fault type and occurrence time to ensure the normal operation of other equipment.

[0037] In this embodiment, after the fault is eliminated, the module can also automatically restore power supply, analyze the cause of the fault, and generate a recovery report for subsequent maintenance.

[0038] The adaptive protection strategy module learns and analyzes the working status and historical fault data of the transformer in real time, and automatically optimizes the protection strategy. The system simulates the current and temperature parameters under different working scenarios through the model, and intelligently adjusts the triggering conditions of the current boost protection.

[0039] In this embodiment, the adaptive protection strategy module has a built-in adaptive protection strategy model, which automatically adjusts the triggering conditions of the current boost protection by performing nonlinear mapping on the current and temperature parameters. The model is as follows: ; in, For at time point The protection strategy output indicates the current boost protection triggering condition, such as the current threshold; The input feature vector contains multiple parameters of current, temperature, and load monitored in real time: ; is a weight matrix, which indicates the influence of each input feature on the protection strategy and needs to be obtained through training; is the bias term, which adjusts the output reference value; As the activation function, ReLU or Sigmoid function is selected to introduce nonlinear characteristics.

[0040] like Figure 5 As shown, the weight matrix is ​​determined by the following steps and the bias term : Data collection: Collect historical operating data of the transformer, including characteristics such as current, temperature, load, as well as fault conditions and corresponding protection measures.

[0041] Feature selection and preprocessing: clean and standardize the data, select features related to the fault, and construct the input feature vector .

[0042] Model training: Use a deep learning framework (such as TensorFlow or PyTorch) to build a neural network model, train it with historical data, and use the back-propagation algorithm to continuously adjust the weight matrix WWW and bias terms. , to minimize the loss function (such as mean squared error).

[0043] Cross-validation: Use the cross-validation method to evaluate the accuracy and generalization ability of the model to ensure good performance on unseen data.

[0044] Furthermore, the steps of optimizing protection parameters in the adaptive protection strategy model are as follows: Real-time data monitoring: The system monitors the working status of the transformer in real time and collects current ,temperature ,load data, forming the input feature vector ; Calculate protection strategy output: Input real-time data into the model and calculate protection strategy output; Dynamically adjust protection parameters: according to the protection strategy value output by the model , dynamically adjust the triggering conditions of the current boost protection, such as modifying the current threshold or delaying the triggering time, to adapt to the current working environment and status; Real-time feedback and learning: After the protection measures are implemented, the system continues to monitor the operating status of the transformer, collect feedback data, and if a fault occurs, record the event and use it for subsequent model retraining, optimize the weight matrix and bias terms, and enhance the model's adaptive ability; Policy iteration and optimization: Regularly evaluate model performance, combine new fault data and operation data, and iterate the policy to ensure that the protection measures always meet the actual operating environment.

[0045] The multi-dimensional alarm and response module provides multiple alarm methods, including sound alarm, SMS notification, and mobile phone App push.

[0046] In this embodiment, personalized response strategies can be formulated for different fault types. For example, in the case of a high-temperature current rise fault, the system can give priority to notifying the on-duty personnel and automatically adjust the working state of the cooling system; in the case of a transient overload fault, the system will give priority to cutting off the power supply. The module also supports linkage with other systems to achieve a multi-level response mechanism and improve emergency handling efficiency.

[0047] The beneficial effects of the present invention are: 1. The large-capacity transformer primary current protection system can realize comprehensive monitoring and intelligent protection of large-capacity transformers by introducing modules such as intelligent fault prediction, dynamic load adaptation, environmental impact assessment and adaptive protection strategy. The coordinated work of various modules enables the protection measures to be adjusted in real time, which significantly improves the safety and reliability of the transformer under complex working conditions and reduces the occurrence rate of equipment failures.

[0048] 2. The large-capacity transformer primary current protection system has self-learning capabilities and can continuously optimize protection strategies based on actual operating data to ensure that the transformer always remains in the best working condition in various operating environments. This not only improves the stability of the power system, but also effectively reduces maintenance costs and extends the service life of the equipment.

[0049] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0050] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A large-capacity transformer primary current boost protection system, characterized in that: It includes intelligent fault prediction module, current rise trigger monitoring module, dynamic load adaptation module, environmental impact assessment module, fault isolation and recovery module, adaptive protection strategy module and multi-dimensional alarm and response module; Intelligent fault prediction module uses machine learning algorithms to analyze historical operation data and real-time data to identify potential faults. By building a multi-dimensional feature model, it monitors changes in parameters such as temperature, vibration, and oil level, and uses anomaly detection technology to provide early warning of potential faults. The current boost trigger monitoring module monitors the current boost process of the transformer in real time. Through high-frequency sampling technology, it accurately measures the primary current, monitors the current fluctuation in real time, and sets the current boost threshold. The dynamic load adaptation module analyzes the load characteristics of the transformer in real time and dynamically adjusts the current boost protection strategy. When the load changes dramatically, it identifies the instantaneous load changes and intelligently adjusts the protection parameters, such as the current boost protection time and current limit value. Environmental impact assessment module, integrated with meteorological data acquisition equipment, obtains environmental parameters such as temperature, humidity and wind speed in real time, and performs correlation analysis with transformer operation data; Fault isolation and recovery module, when a current rise fault is detected, the fault isolation procedure is automatically started. The system quickly identifies the fault area and quickly cuts off the power supply of the faulty part through the switch device. At the same time, the fault type and occurrence time are recorded to ensure the normal operation of other equipment; The adaptive protection strategy module learns and analyzes the working status and historical fault data of the transformer in real time, automatically optimizes the protection strategy, and simulates the current and temperature parameters under different working scenarios through the model, and intelligently adjusts the triggering conditions of the current rise protection; The multi-dimensional alarm and response module provides multiple alarm methods, including sound alarm, SMS notification, and mobile phone App push.

2. A large-capacity transformer primary current rising protection system according to claim 1, characterized in that: In the intelligent fault prediction module, the multidimensional feature model is as follows: ； in, Indicates at a point in time Failure risk index; Indicates at a point in time Temperature parameters; Indicates at a point in time Vibration parameters of Indicates at a point in time Oil level parameters; Indicates at a point in time Current parameters; are the weights of each parameter, reflecting their influence on the failure risk; is a bias term used to adjust the flexibility of the model.

3. A large-capacity transformer primary current boost protection system according to claim 2, characterized in that: The multidimensional feature model, the weights in the formula and the bias term Determine, including the following process: Data collection: Collect historical data of the transformer under different working conditions, including changes in temperature, vibration, oil level and current, as well as the time and type of fault occurrence; Feature selection and engineering: Analyze the collected data, select the most relevant features using correlation analysis and principal component analysis (PCA), and standardize them to facilitate model training; Model training: Use a labeled data set that contains faults and normal states, and apply machine learning algorithms such as random forests and gradient boosting trees to train the model. Use the back propagation algorithm to continuously optimize weights and biases to minimize the prediction error. Cross-validation: Use k-fold cross-validation to evaluate the accuracy and generalization ability of the model to ensure that the model performs well on unseen data.

4. A large-capacity transformer primary current boost protection system according to claim 1, characterized in that: The dynamic load adaptation module has a built-in real-time load analysis model, which is based on multiple factors including load change rate, current limit, and current boost protection time, and is expressed as the following formula: ； in, Indicates at a point in time Dynamic protection parameters such as current rise protection time or current limit value; Indicates at a point in time The actual load value, such as current current; Indicates at a point in time The load change rate is calculated as follows: ； In the above formula, is the time interval; is a coefficient, which indicates the intensity of the load's influence on the protection parameters; is an attenuation factor, reflecting the influence of load change rate on the adjustment of protection parameters; is a bias term used to adjust the baseline value of the model output.

5. A large-capacity transformer primary current boost protection system according to claim 4, characterized in that: The real-time load analysis model determines the coefficients by the following steps , and the bias term : Data collection: Collect historical operating data of transformers under different load conditions, including current, load changes and corresponding protection measures; Characteristic analysis: Use statistical analysis methods such as regression analysis to evaluate the relationship between load values ​​and protection parameters to determine the coefficients and ; Model calibration: By simulating different load change scenarios, adjust value, so that the model output meets the actual operation requirements.

6. A large-capacity transformer primary current boost protection system according to claim 4, characterized in that: When the real-time load analysis model is used, it includes the following process: Real-time load monitoring: The system monitors the load status of the transformer in real time and obtains the current value ; Calculate the load change rate: Use historical data to calculate the load change rate; Dynamically adjust protection parameters: Substitute the current load value and load change rate into the model formula, and , dynamically adjust the current boost protection time and current limit value; Real-time feedback and adjustment: When the load continues to change, repeat the above steps to ensure that the protection parameters always remain at the optimal level to adapt to the current load status; Strategy optimization and learning: Regularly analyze the prediction effect of the model and optimize the model coefficients based on actual fault records and maintenance data , and the bias term , improving the accuracy and response speed of dynamic adjustment.

7. A large-capacity transformer primary current boost protection system according to claim 1, characterized in that: The environmental impact assessment module, by establishing an environmental impact model, systematically assesses the impact of environmental changes on the upflow protection strategy and automatically adjusts the protection parameters to prevent the risk of equipment overload caused by environmental changes; The environmental impact model combines multiple influencing factors such as temperature, vibration, oil level and current and is constructed using the following formula: ； in, Indicates at a point in time The environmental impact index indicates the degree of environmental impact on upwelling protection strategies; Indicates at a point in time Temperature parameters; Indicates at a point in time Humidity parameters; Indicates at a point in time Wind speed parameters; are the weights of the environmental impact index of temperature, humidity and wind speed, respectively, reflecting the influence of each environmental factor on the protection parameters; is a bias term used to adjust the flexibility of the model; And determine the coefficients by following the steps below and the bias term : Data collection: Collect the operating data of the transformer under different environmental conditions, record the temperature, humidity and wind speed, as well as the corresponding fault conditions and protection parameters; Characteristic analysis: Evaluate the relationship between environmental factors and conservation parameters through statistical analysis methods such as multivariate linear regression to determine the weights ; Model calibration: bias term based on historical data Make adjustments to ensure that the model output meets actual needs.

8. A large-capacity transformer primary current boost protection system according to claim 7, characterized in that: When the environmental impact model is used, it includes the following specific processes: Real-time environmental monitoring: Use meteorological data acquisition equipment to monitor the environmental conditions around the transformer in real time and obtain the temperature ,humidity and wind speed data; Calculate the environmental impact index: Substitute the real-time collected environmental data into the model formula; Analyze environmental impact: Based on the calculated environmental impact index , judge the influence of the current environment on the operation of the transformer. If Exceeding a certain preset threshold , it means that the current environmental conditions pose a risk to the transformer's current-boosting protection strategy; Dynamic adjustment of protection parameters: If the environmental impact index reaches the threshold, the system automatically adjusts the current increase protection strategy, including modifying the current limit value and protection time, to adapt to the current environmental conditions and reduce the risk of overload caused by environmental changes; Continuous monitoring and feedback: The system continuously monitors environmental changes and regularly updates the model’s parameters, retraining based on new data to improve the model’s accuracy and adaptability.

9. A large-capacity transformer primary current boost protection system according to claim 1, characterized in that: The adaptive protection strategy module has a built-in adaptive protection strategy model, which automatically adjusts the triggering conditions of the current boost protection by performing nonlinear mapping on the current and temperature parameters. The model is as follows: ； in, For at time point The protection strategy output indicates the current boost protection triggering condition, such as the current threshold; The input feature vector contains multiple parameters of current, temperature, and load monitored in real time: ； is a weight matrix, which indicates the influence of each input feature on the protection strategy and needs to be obtained through training; is the bias term, which adjusts the output reference value; As the activation function, ReLU or Sigmoid function is selected to introduce nonlinear characteristics.

10. A large capacity transformer primary current boost protection system according to claim 9, characterized in that: The steps of optimizing protection parameters in the adaptive protection strategy model are as follows: Real-time data monitoring: The system monitors the working status of the transformer in real time and collects current ,temperature ,load data, forming the input feature vector ; Calculate protection strategy output: Input real-time data into the model and calculate protection strategy output; Dynamically adjust protection parameters: according to the protection strategy value output by the model , dynamically adjust the triggering conditions of the current boost protection, such as modifying the current threshold or delaying the triggering time, to adapt to the current working environment and status; Real-time feedback and learning: After the protection measures are implemented, the system continues to monitor the operating status of the transformer, collect feedback data, and if a fault occurs, record the event and use it for subsequent model retraining, optimize the weight matrix and bias terms, and enhance the model's adaptive ability; Policy iteration and optimization: Regularly evaluate model performance, combine new fault data and operation data, and iterate the policy to ensure that the protection measures always meet the actual operating environment.

Images