A method, system, equipment, and medium for the operation control of a converter transformer.

Abstract

This invention discloses a method, system, equipment, and medium for the operation control of a converter transformer, belonging to the field of operation control. The method comprises: acquiring power operation data and environmental data of the converter transformer, and determining a first carbon emission amount based on the power operation data and the environmental data; correcting the first carbon emission amount based on a preset dynamic response function to obtain a second carbon emission amount, and then filtering and correcting the second carbon emission amount using an adaptive Kalman filter method to obtain a target carbon emission amount; inputting the target carbon emission amount into a preset multi-objective optimization function for solving to obtain the optimal combination of operating parameters for the converter transformer, and controlling the operation of the converter transformer based on the optimal combination of operating parameters. Therefore, by implementing this invention, the operational stability of the converter transformer can be improved.

Description

Technical Field

[0001] This invention relates to the field of power systems, and more particularly to a method, system, equipment, and medium for the operation control of a converter transformer. Background Technology

[0002] Converter substations, especially flexible DC converter stations, are crucial hubs for modern power grids to achieve cross-regional, high-capacity power transmission and interconnection. The operational status of their core equipment, the converter transformer, directly affects the reliability, power quality, and power supply security of the entire transmission system. If a converter transformer fails due to overheating, overload, or insulation aging, it not only leads to its own destruction but can also trigger a chain reaction, causing regional power grid fluctuations or even large-scale blackouts, resulting in enormous economic and social losses. Therefore, ensuring the operational stability of converter substations is their most fundamental and important technical requirement, and the cornerstone of maintaining the safe and stable operation of the power grid.

[0003] Existing technologies typically rely on setting fixed temperature and current protection thresholds and using static models or empirical formulas for load distribution and control to ensure stability. The core flaw of these methods lies in neglecting the complex dynamic coupling between extreme weather conditions and transformer operating states. Drastic changes in ambient temperature, humidity, and wind speed significantly alter a transformer's heat dissipation capacity and loss characteristics, and traditional static models cannot respond to these changes in real time, leading to conservative and inaccurate control strategies. This not only exposes transformers to overheating risks under extreme weather conditions, creating safety hazards, but also, due to its failure to accurately quantify the impact of climate on carbon emissions, results in severely inaccurate carbon emission calculations during low-carbon operation, significantly diminishing optimization effectiveness and making it difficult to achieve efficient and stable operation while ensuring stability. Summary of the Invention

[0004] This invention provides a method, system, equipment, and medium for the operation control of converter transformers, which can improve the operational stability of converter transformers.

[0005] An embodiment of the present invention provides an operation control method for a converter transformer, comprising:

[0006] Acquire power operation data and environmental data of the converter transformer, and determine the first carbon emission based on the power operation data and the environmental data;

[0007] The first carbon emission is corrected based on a preset dynamic response function to obtain a second carbon emission, and an adaptive Kalman filter is used to filter and correct the second carbon emission to obtain a target carbon emission. The dynamic response function is constructed based on the power operation data, the environmental data, and the climate response factor, and the climate response factor is obtained after processing the environmental data.

[0008] The target carbon emission is input into a preset multi-objective optimization function for solution, to obtain the optimal combination of operating parameters for the converter transformer, and the operation of the converter transformer is controlled based on the optimal combination of operating parameters.

[0009] This invention acquires real-time power operation data and environmental data, converting the transformer's operating state into a quantifiable first carbon emission. This first carbon emission accurately identifies potential risk points that could lead to instability. By using a climate response factor to precisely map climate stress into carbon emission calculations, the second carbon emission accurately reflects the transformer's operating load under harsh conditions, providing a precise and environmentally adapted decision-making basis for stability control. Adaptive Kalman filtering filters out interference from sensor drift and sudden environmental changes, correcting unavoidable noise and errors in data acquisition and model calculations in real time. This results in a smoother and more reliable target carbon emission output, significantly improving the reliability of state perception and avoiding instability risks caused by control decisions based on distorted data. By solving a multi-objective optimization problem, the final optimal combination of operating parameters is a globally optimal solution that prioritizes all stability constraints while pursuing low carbon emissions and high efficiency. Controlling the transformer based on this combination proactively maintains its operating state within a safe and stable range, directly achieving the ultimate goal of improving operational stability. Compared to existing technologies, this invention improves the operational stability of converter transformers.

[0010] Further, determining the first carbon emission based on the power operation data and the environmental data includes:

[0011] The power loss of the transformer is calculated based on the power operation data.

[0012] The initial transformer efficiency is corrected using the temperature and humidity data from the environmental data to obtain the target transformer efficiency;

[0013] The power loss, the target transformer efficiency, and the grid carbon factor are input into a preset carbon emission measurement model to calculate the first carbon emission.

[0014] By acquiring real-time power operation data and environmental data, and converting the transformer's operating status into quantifiable primary carbon emissions, potential risk points that may lead to instability can be accurately identified from these primary carbon emissions.

[0015] Furthermore, the step of correcting the first carbon emission amount based on a preset dynamic response function to obtain the second carbon emission amount includes:

[0016] The environmental data and the power operation data are input into the dynamic response function to generate temperature correction coefficient, humidity correction coefficient and load correction coefficient;

[0017] The temperature correction factor, humidity correction factor, and load correction factor are combined to obtain a composite correction factor. The first carbon emission is then multiplied by the composite correction factor to obtain the second carbon emission.

[0018] By using a dynamic response function, the real-time impact of extreme weather on transformer losses and efficiency is quantified. The corrected carbon emissions more accurately reflect the actual thermal stress of the equipment, providing precise input for subsequent multi-objective optimization. This effectively prevents overheating and insulation aging in control decisions and improves operational stability.

[0019] Furthermore, the dynamic response function is constructed by combining the power operation data, the environmental data, and the climate response factor, and includes:

[0020] Based on the environmental data, the extreme climate exceedance threshold and the persistence index are calculated, and based on the extreme climate exceedance threshold and the persistence index, a climate response factor is generated;

[0021] A high-order nonlinear regression analysis was performed on the climate response factors to obtain temperature correction factors, humidity correction factors, and load correction factors.

[0022] Based on the environmental data and power operation data, a fuzzy correction factor is obtained by training an adaptive neural fuzzy inference system, and a temporal correction factor is obtained by training a long short-term memory network.

[0023] The temperature correction factor, humidity correction factor, load correction factor, fuzziness correction factor, and time sequence correction factor are integrated to obtain the dynamic response function.

[0024] By converting climate stress into quantifiable correction factors, carbon emission calculations become more accurate; based on this, optimized control can proactively prevent transformer overheating and insulation aging, thereby directly improving operational stability.

[0025] Further, the step of using an adaptive Kalman filter to filter and correct the second carbon emission amount to obtain the target carbon emission amount includes:

[0026] Real-time monitoring of the actual carbon emissions of converter transformers;

[0027] The second carbon emission amount and the actual carbon emission value are input into an adaptive Kalman filter algorithm to dynamically calculate the Kalman gain, and the second carbon emission amount and the actual carbon emission value are weighted and fused according to the Kalman gain to obtain the target carbon emission amount.

[0028] In this way, adaptive Kalman filtering can filter out interference caused by sensor drift and sudden environmental changes, and can correct the unavoidable noise and errors in data acquisition and model calculation in real time, thereby outputting a smoother and more reliable target carbon emission, which greatly improves the reliability of state perception and avoids the risk of instability caused by making control decisions based on distorted data.

[0029] Furthermore, the step of inputting the target carbon emissions into a preset multi-objective optimization function for solution to obtain the optimal combination of operating parameters for the converter transformer includes:

[0030] The target carbon emissions are input into a preset multi-objective optimization function, and a multi-objective genetic algorithm is used to solve the multi-objective optimization function to obtain a solution set. The obtained solution set is then stratified by non-dominated sorting to determine the non-dominated level, and the crowding degree of each solution within the same non-dominated level is calculated.

[0031] Based on the non-dominated level and the congestion degree, the optimal solution is selected from the solution set, and the optimal combination of operating parameters for the converter transformer is determined based on the optimal solution. The multi-objective optimization function includes a carbon emission objective function, a thermal stability objective function, and an operating energy efficiency objective function.

[0032] By solving this multi-objective optimization problem, the final optimal combination of operating parameters is a globally optimal solution that prioritizes all stability constraints while pursuing low carbon emissions and high efficiency.

[0033] Furthermore, the control of the converter transformer's operation based on the optimal combination of operating parameters includes:

[0034] The optimal combination of operating parameters is sent to the transformer control system to adjust at least one of the following in real time: load distribution of the converter transformer, operating status of the cooling system, on-load tap position, and magnetic flux density.

[0035] Another embodiment of the present invention provides an operation control system for a converter transformer, comprising: an acquisition module, a processing module, and a control module;

[0036] The acquisition module is used to acquire the power operation data and environmental data of the converter transformer, and determine the first carbon emission based on the power operation data and the environmental data;

[0037] The processing module is used to correct the first carbon emission amount based on a preset dynamic response function to obtain a second carbon emission amount, and to filter and correct the second carbon emission amount using an adaptive Kalman filter method to obtain a target carbon emission amount. The dynamic response function is constructed based on the power operation data, the environmental data, and the climate response factor, and the climate response factor is obtained after processing the environmental data.

[0038] The control module is used to input the target carbon emission into a preset multi-objective optimization function for solution, obtain the optimal combination of operating parameters for the converter transformer, and control the operation of the converter transformer based on the optimal combination of operating parameters.

[0039] Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the steps of the operation control method for the converter transformer of the present invention.

[0040] Another embodiment of the present invention provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform steps such as the operation control method of the converter transformer of the present invention. Attached Figure Description

[0041] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0042] Figure 1 This is a flowchart illustrating one embodiment of the converter transformer operation control method provided in this application;

[0043] Figure 2 This is a flowchart illustrating one embodiment of steps S201 to S203 provided in this application;

[0044] Figure 3 This is a flowchart illustrating one embodiment of steps S301 to S304 provided in this application;

[0045] Figure 4 This is a flowchart illustrating one embodiment of steps S401 to S402 provided in this application;

[0046] Figure 5 This is a flowchart illustrating one embodiment of steps S501 to S502 provided in this application;

[0047] Figure 6 This is a schematic diagram of an embodiment of the operation control system for the converter transformer provided in this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0050] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0051] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0052] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0053] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0054] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0055] Converter substations are critical hubs in the power grid, and the operational stability of their core equipment, the converter transformer, directly affects power supply security and system reliability. Current technologies primarily rely on fixed thresholds and static models for control and protection, failing to fully consider the dynamic impacts of extreme weather conditions (such as drastic changes in temperature, humidity, and wind speed) on transformer heat dissipation and losses. This results in delayed response and insufficient accuracy in control strategies, making it difficult to effectively ensure equipment safety under extreme conditions. Furthermore, inaccurate carbon emission calculations prevent the achievement of low-carbon optimization based on stable operation, thus hindering the overall operational efficiency of the power grid in complex environments.

[0056] See Figure 1 To improve the operational stability of converter transformers, an embodiment of the present invention provides an operation control method for converter transformers, including steps S101 to S103.

[0057] Step S101: Obtain the power operation data and environmental data of the converter transformer, and determine the first carbon emission based on the power operation data and the environmental data;

[0058] In some embodiments, acquiring power operation data and environmental data of the converter transformer specifically involves: real-time collection of power operation data and environmental data through a transformer monitoring system and a sensor network; wherein the power operation data includes: transformer load current. Winding resistance Grid frequency The environmental data includes: ambient temperature. Ambient humidity Wind speed and solar irradiance At the same time, it is also necessary to collect the hysteresis loss coefficient through methods such as transformer design manuals. eddy current loss coefficient Initial transformer efficiency Data such as...

[0059] In some embodiments, after the data is collected, the environmental data needs to be preprocessed to remove noise and outliers. Specifically, moving median filtering and the Hampel criterion are used for outlier detection and correction, as shown in the following formula: ;in, For the collected climate data, To adjust the sliding window size, As a smoothing factor, It is a moving average.

[0060] Please refer to Figure 2 In some embodiments, determining the first carbon emission based on the power operation data and the environmental data includes steps S201 to S203:

[0061] Step S201: Calculate the power loss of the transformer based on the power operation data;

[0062] In some embodiments, the power loss of a transformer is divided into two parts: copper loss and iron loss. Copper loss is the loss caused by the resistance of the transformer windings through which current flows, while iron loss is the loss caused by the transformer core under the influence of a magnetic field. The calculation formula for copper loss, caused by the resistance of the transformer windings through which current flows, is as follows: ;in: Copper loss, measured in watts; This is the transformer load current, measured in amperes (A). The resistance is the winding resistance, measured in ohms (Ω). Iron loss is generated by the magnetic field acting on the iron core, and its calculation formula is as follows: ;in: Iron loss, measured in watts; is the hysteresis loss coefficient, with units of a dimensionless constant, which depends on the material properties; is the eddy current loss coefficient, with units of a dimensionless constant, which depends on the material properties; This is the power grid frequency, measured in Hertz (Hz). The peak value of the magnetic flux density of the iron core is expressed in Tesla (T). This represents the core volume, expressed in cubic meters (m³). After calculating the copper and iron losses, the following steps are performed: The total power loss of the transformer can then be calculated. .

[0063] Step S202: Using the temperature and humidity data from the environmental data, the initial transformer efficiency is corrected to obtain the target transformer efficiency;

[0064] In some embodiments, since ambient temperature and humidity affect transformer efficiency and power losses, this model incorporates the impact of ambient temperature and humidity on transformer efficiency to improve the accuracy of carbon emission calculations. The specific correction process consists of two steps. The first step is temperature correction, as temperature changes affect transformer efficiency, which typically decreases under high-temperature conditions. The temperature correction formula is as follows: ;in: The transformer efficiency is after humidity correction, expressed as a dimensionless coefficient. The efficiency is the reference transformer efficiency, expressed as a dimensionless coefficient. This is a temperature correction factor, with units of 1 / °C; This refers to the ambient temperature, expressed in degrees Celsius. The reference temperature is in degrees Celsius, typically taken as 25°C or 40°C. The second step is humidity correction, as humidity also affects transformer cooling efficiency. High humidity reduces the transformer's cooling effect, thus increasing losses. The humidity correction formula is as follows: ;in: The cooling efficiency after temperature correction is expressed as a dimensionless coefficient. Ambient humidity, expressed as a percentage (%). This is a humidity correction factor, expressed as a dimensionless constant, obtained from experimental data or tests. The target transformer efficiency can then be calculated from this factor. .

[0065] Step S203: Input the power loss, the target transformer efficiency, and the grid carbon factor into a preset carbon emission measurement model to calculate the first carbon emission.

[0066] In some embodiments, the calculated total power loss Target transformer efficiency after environmental factor correction and the carbon factor of the power grid and time interval Input the preset carbon emission measurement model together. The core calculation formula of this model is: ;in: The total power loss has been calculated in the preceding calculations; and To take into account operating efficiency after adjustments for temperature and cooling conditions; The grid carbon factor is expressed in kg CO2 / kWh, representing the carbon emissions corresponding to a unit of electrical energy. The time interval is in hours, and the output of this formula is the first carbon emission. .

[0067] By acquiring real-time power operation data and environmental data, and converting the transformer's operating status into quantifiable primary carbon emissions, potential risk points that may lead to instability can be accurately identified from these primary carbon emissions.

[0068] Step S102: The first carbon emission is corrected based on a preset dynamic response function to obtain a second carbon emission, and the second carbon emission is filtered and corrected using an adaptive Kalman filter method to obtain a target carbon emission. The dynamic response function is constructed based on the power operation data, the environmental data, and the climate response factor, and the climate response factor is obtained after processing the environmental data.

[0069] Please refer to Figure 3 In some embodiments, the dynamic response function is constructed by combining the power operation data, the environmental data, and the climate response factor, including steps S301 to S304:

[0070] Step S301: Calculate the extreme climate exceedance threshold and persistence index based on the environmental data, and generate a climate response factor based on the extreme climate exceedance threshold and the persistence index;

[0071] In some embodiments, firstly, after obtaining the preprocessed environmental data, in order to quantify the impact of climatic conditions on transformer operation, an extreme climate exceedance threshold is defined, i.e., the portion of ambient temperature and humidity exceeding a baseline value, and a duration index is introduced to quantify the intensity and duration of extreme climate, wherein the temperature exceedance threshold... Used to indicate that the ambient temperature exceeds the reference temperature. The formula for this part is: ;in, This refers to the temperature threshold. However, humidity exceeding the threshold... Used to indicate that ambient humidity exceeds the reference humidity. The formula for this part is: ;in, The baseline humidity is typically set at 70%. Then, to measure the intensity and duration of extreme conditions, a weighted duration index is introduced, applied to conditions exceeding a temperature threshold. and humidity exceeding the threshold By performing a weighted integral, the temperature duration index is obtained. and humidity persistence index The specific formula is as follows: ;in, and These are the memory time constants for temperature and humidity, respectively, in hours, reflecting the duration and intensity of temperature and humidity exceeding the threshold. Finally, the temperature exceeding the threshold... Humidity exceeds threshold Temperature duration integral index and humidity persistence index The temperature response factor can then be calculated. and humidity response factor The relevant calculation formula is:

[0072] ;

[0073] ;

[0074] In the formula, , and As a preset temperature weighting coefficient, , and This application does not impose any restrictions on the preset humidity weighting coefficient.

[0075] It should be noted that, in order to incorporate the persistent effects of extreme temperature and humidity conditions on the operation of converter transformers into the response function modeling, this paper introduces a temperature persistence index. With humidity persistence index As a characterization variable of external disturbances, by using the persistence index as the core input variable of external disturbance factors, it is possible to dynamically characterize the coupling features in extreme climate events, and significantly improve the accuracy and robustness of thermal disturbance identification.

[0076] Then, since wind speed has a significant impact on transformer cooling capacity, this section introduces a quantitative relationship between wind speed and transformer cooling efficiency, considering the influence of wind speed changes on transformer cooling capacity. The effect of wind speed on transformer cooling efficiency can be expressed using the dimensionless wind Reynolds number. and Nusselt To indicate: , ;in, It is the factor affecting the heat transfer coefficient. For Prandtl numbers, Prandtl number is an empirical parameter calibrated based on cooling structure and experimental data. air density, For air viscosity, For air thermal conductivity, For characteristic length, Let be the constant of the wind speed heat transfer coefficient. Then, the gain factor of wind speed on cooling capacity can be calculated using the following formula: In the formula: For wind speed gain factor, For reference heat transfer coefficient, The gain coefficient represents the effect of airflow on cooling efficiency.

[0077] Finally, based on the temperature response factor Humidity response factor and wind speed response factor Climate response factors can then be constructed. The relevant formula is: .

[0078] Step S302: Perform high-order nonlinear regression analysis on the climate response factors to obtain temperature correction factor, humidity correction factor and load correction factor.

[0079] In some embodiments, a multinomial regression model is used to establish the nonlinear relationships between ambient temperature, ambient humidity, transformer load, and carbon emissions.

[0080] In some embodiments, the formula for the temperature correction factor is: ;in: The regression coefficient represents the degree of influence of temperature on carbon emissions; The ambient temperature. For regression terms, the most suitable number of regression terms is usually selected through cross-validation; the humidity correction factor can be represented by the following regression model: ;in: The regression coefficient represents the degree of influence of humidity on carbon emissions. The ambient humidity is used. The load correction factor is represented by the following polynomial regression: ;in: The regression coefficient represents the degree of impact of load on carbon emissions. This represents the actual load on the transformer. These are the regression terms. Together, these sub-functions constitute the initial nonlinear correction framework for carbon emissions.

[0081] Step S303: Based on the environmental data and power operation data, a fuzzy correction factor is obtained by training an adaptive neural fuzzy inference system, and a temporal correction factor is obtained by training a long short-term memory network.

[0082] In some embodiments, to further optimize the carbon emission correction function, an adaptive neurofuzzy inference system is introduced. First, fuzzy rules are defined in the adaptive neurofuzzy inference system regarding the relationship between environmental factors such as temperature, humidity, and load and carbon emissions (if...). High and Lowest carbon emissions result in the highest emissions; if low and (Higher humidity results in lower carbon emissions). These rules describe the membership functions of temperature and humidity as follows: ;in: This is a membership function of temperature, representing the effect of temperature on carbon emissions; The ambient temperature; For reference temperature; The membership function represents the control speed, indicating the response speed to temperature changes. Membership functions for humidity and load can be defined by analogy. ANFIS then trains the neural network using a forward propagation algorithm to adjust the parameters of the fuzzy rules and membership functions. The training process adjusts the premise and conclusion parameters by minimizing the following loss function, where the expression for the loss function is: ;in, This refers to the actual measured carbon emissions; The carbon emissions predicted by ANFIS are represented by N, which is the number of training samples. The output of ANFIS is a carbon emission correction factor, also known as a fuzzy correction factor. This correction factor dynamically adjusts carbon emissions. Subsequently, to more accurately handle the temporal characteristics of environmental data and capture the long-term and short-term effects of factors such as temperature, humidity, and load on carbon emissions, a Long Short-Term Memory (LSTM) network is introduced. The LSTM network minimizes the mean squared error between predicted and actual carbon emissions. The trained LSTM network can directly output a time-series correction factor based on the running data. .

[0083] It should be noted that LSTM can predict carbon emissions over time by memorizing past environmental data. LSTM is a special type of recurrent neural network that uses a gating mechanism to control the learning dependency of long short-term memory. The formula for LSTM is as follows: Input gate: Forgotten Gate: Output gate: Assuming the cell state: Cell status update: Final output: ;in: These are the outputs of the input gate, forget gate, and output gate, respectively. , , and It is a weight matrix; , , and For bias terms; For unit states, long-term dependency information is stored; The output represents the state after passing through the LSTM unit.

[0084] It should be noted that the Adaptive Neural Fuzzy Inference System (ANFIS) combines the advantages of fuzzy inference systems and neural networks, and can dynamically adjust the parameters of fuzzy rules and membership functions through training sets to refine the correction of carbon emissions.

[0085] Step S304: Integrate the temperature correction factor, the humidity correction factor, the load correction factor, the fuzziness correction factor, and the time sequence correction factor to obtain the dynamic response function.

[0086] In some embodiments, the temperature correction factor, the humidity correction factor, the load correction factor, the fuzzy correction factor, and the time-series correction factor are integrated to form a unified composite response function that can comprehensively reflect the complex nonlinear and dynamic effects of the environment and load on carbon emissions. Its final mathematical expression is: This dynamic response function, as the core correction module, can perform high-precision dynamic adjustments to the initially calculated carbon emissions (first carbon emissions) under different environmental and operating conditions, providing an accurate data foundation for subsequent filtering and optimization.

[0087] By converting climate stress into quantifiable correction factors, carbon emission calculations become more accurate; based on this, optimized control can proactively prevent transformer overheating and insulation aging, thereby directly improving operational stability.

[0088] Please refer to Figure 4 In some embodiments, the step of correcting the first carbon emission amount based on a preset dynamic response function to obtain the second carbon emission amount includes steps S401 to S402:

[0089] Step S401: Input the environmental data and the power operation data into the dynamic response function to generate temperature correction coefficient, humidity correction coefficient and load correction coefficient;

[0090] In some embodiments, the environmental data and the power operation data are input into the dynamic response function to generate a temperature correction coefficient, a humidity correction coefficient, and a load correction coefficient. The temperature correction coefficient is calculated using a high-order nonlinear regression model, the humidity correction coefficient is calculated using environmental humidity data using a similar high-order nonlinear regression model, and the load correction coefficient is calculated using load data from the power operation data.

[0091] Step S402: Combine the temperature correction coefficient, the humidity correction coefficient, and the load correction coefficient to obtain a composite correction coefficient, and multiply the first carbon emission amount by the composite correction coefficient to obtain the second carbon emission amount.

[0092] In some embodiments, the temperature correction factor, humidity correction factor, and load correction factor are combined and integrated by multiplication to obtain a composite correction factor. Subsequently, the calculated first carbon emission is multiplied by this composite correction factor to complete the dynamic correction of the carbon emission and obtain a more accurate second carbon emission.

[0093] By using a dynamic response function, the real-time impact of extreme weather on transformer losses and efficiency is quantified. The corrected carbon emissions more accurately reflect the actual thermal stress of the equipment, providing precise input for subsequent multi-objective optimization. This effectively prevents overheating and insulation aging in control decisions and improves operational stability.

[0094] In some embodiments, the step of using an adaptive Kalman filter to filter and correct the second carbon emission amount to obtain the target carbon emission amount includes: real-time monitoring of the actual carbon emission value of the converter transformer; inputting the second carbon emission amount and the actual carbon emission value into an adaptive Kalman filter algorithm to dynamically calculate the Kalman gain; and weighting and fusing the second carbon emission amount and the actual carbon emission value according to the Kalman gain to obtain the target carbon emission amount. Specifically, firstly, the actual carbon emission value is acquired in real time by a carbon emission monitoring sensor installed in the converter transformer system; then, the actual carbon emission value and the second carbon emission amount are input into an adaptive Kalman filter algorithm to calculate the Kalman gain. Among them, Kalman gain This is the core of the algorithm; it dynamically determines whether to place more trust in model predictions or sensor measurements during the correction process. The calculation formula is as follows: ;in: Let be the error covariance matrix of the previous time step, representing the uncertainty in estimating the system state; The observation matrix describes the relationship between state variables (predicted carbon emissions) and observed quantities (measured carbon emissions). The noise matrix represents the uncertainty and noise level of the sensor measurements. Finally, the model predictions and sensor measurements are weighted and fused to calculate the final target carbon emissions. The relevant formula is as follows: ,in: This is the carbon emission forecast value at the previous moment, also known as the second carbon emission amount; This refers to actual carbon emissions; Kalman gain reflects the correction ratio between the measured and predicted values; through this correction process, the accuracy of carbon emission calculations can be continuously optimized, thereby improving the accuracy of predictions.

[0095] It should be noted that a key step in Kalman filtering is updating the error covariance matrix. , representing the magnitude of the estimation error. The update formula is: ;in: It is an identity matrix. For Kalman gain.

[0096] It should be noted that the "adaptive" characteristic is reflected in the system's ability to adjust in real time based on changes in the second carbon emission level and the actual carbon emission value. When sensor data is reliable (actual carbon emissions are low), Increase the value of the measurement and trust it more; when the model prediction is more stable ( When carbon emissions decrease (in smaller amounts), the second carbon emission level decreases, and the forecasts are more trusted.

[0097] In this way, adaptive Kalman filtering can filter out interference caused by sensor drift and sudden environmental changes, and can correct the unavoidable noise and errors in data acquisition and model calculation in real time, thereby outputting a smoother and more reliable target carbon emission, which greatly improves the reliability of state perception and avoids the risk of instability caused by making control decisions based on distorted data.

[0098] Step S103: Input the target carbon emission amount into a preset multi-objective optimization function for solution to obtain the optimal combination of operating parameters for the converter transformer, and control the operation of the converter transformer based on the optimal combination of operating parameters.

[0099] In some embodiments, the construction process of the multi-objective optimization function is as follows: First, in the stage of constructing decision variables and the basic model, the transformer operating parameters that the optimization algorithm can directly control are defined, i.e., the decision vector:

[0100] ;

[0101] ;

[0102] in: For the first The transformer at all times Decision vector; Load rate; For cooling equivalent speed / frequency (relative value or Hz); For on-load tap position (discrete set); The target magnetic flux density (T); Number of units connected in parallel (units); For the first Maximum allowable load rate; This refers to the lower / upper limit of the cooling speed; For the first Available gear settings; Minimum design magnetic flux density; The saturation magnetic flux density of the material; This represents a safety margin for saturation.

[0103] Subsequently, based on the three core objectives of carbon emissions, equipment thermal stability, and operational energy efficiency, corresponding mathematical functions were established. The carbon emissions objective function aims to minimize the total carbon emissions of the system, and its expression is:

[0104] ;

[0105] in: To achieve carbon emission targets; For the sake of emission sovereignty; Model – observation bias weights; Its carbon emissions after Kalman filtering correction;

[0106] Thermal stability is a prerequisite for the safe operation of transformers. In extreme weather conditions, excessively high winding and core temperatures can accelerate insulation aging and even lead to equipment failure. Therefore, the thermal stability objective function penalizes situations where the transformer winding and core temperatures exceed safe limits, ensuring equipment safety. Its expression is:

[0107] ;

[0108] in: For thermal stability penalty; For positive part operators; For the first Taiwan temperature limit; This refers to the core temperature.

[0109] The energy efficiency optimization target aims to minimize transformer body losses and auxiliary equipment energy consumption, improve energy utilization efficiency, and indirectly reduce carbon emissions. The established energy efficiency optimization target is as follows:

[0110] ;

[0111] in: The target is the energy consumption of the main body; Copper loss / Iron loss; Cooling power; Weighted by cooling energy consumption; To switch cost weights; For switching count; Step size;

[0112] Subsequently, while constructing the aforementioned objective function, to ensure that the optimization results meet the requirements of equipment operation safety and system load, multi-dimensional constraints are established. These constraints include:

[0113] Load balance constraints: ;

[0114] in: For the first Base value for unit capacity; For busbar load; For the first Taiwan Current; Rated current; Load rate;

[0115] Magnetic flux density constraint: Magnetic flux density and on-load tap position are key parameters affecting the transformer's operating status. Constraints need to be set to avoid magnetic flux saturation and frequent tap position operations, as shown in the formula:

[0116] ;

[0117] in: Minimum magnetic flux density; It is the saturation magnetic flux density; This is the magnetic flux density margin; Gear; For gear sets;

[0118] On-load tap position constraints: To further limit frequent switching of on-load taps and extend switch life, constraints on tap dwell time and single-step switching count are established. The relevant formulas are as follows:

[0119] ;

[0120] in: This is the last time the gear was shifted; Minimum number of dwell steps; For indicator functions, This refers to the maximum number of stations that can be switched synchronously in a single step.

[0121] Temperature constraints are crucial for the safe operation of equipment. They limit the upper limits of winding and top oil temperatures and the rate of temperature change to prevent sudden temperature rises that could lead to abrupt changes in insulation stress, as shown in the formula:

[0122] ;

[0123] in: This refers to the hot spot temperature of the winding. This refers to the upper limit of the hotspot temperature. This refers to the top oil temperature; This is the upper limit of the top oil temperature.

[0124] Temperature ramping constraints: ;

[0125] in: This represents the upper limit of the temperature ramp-up. This represents the upper limit of load rate variation. This represents the upper limit of the rotational speed variation;

[0126] Robustness Constraints: To address the uncertainty of model parameters under extreme weather conditions, robustness constraints are added. Conditional Risk Values ​​(CVaRs) are used to quantify the risk of carbon emission calculation bias, ensuring the robustness of the optimization results. Conventional absolute value constraints only consider single-point biases, while CVaRs can more comprehensively capture the cumulative risk of extreme biases. Especially in scenarios where extreme weather conditions increase the deviation between the model and the observed values, CVaR constraints can ensure the robustness of the optimization results to uncertainty, avoiding risk exceeding limits. The formula is as follows:

[0127] ;

[0128] in: For confidence level Conditional risk value; Confidence level; This represents the allowable upper limit. Ultimately, these three objective functions, along with all constraints, constitute a complete multi-objective optimization problem.

[0129] It should be noted that, These are the predicted values ​​output by the carbon emission measurement model in steps S101–S102; The carbon emission value is corrected by adaptive Kalman filtering; , These are the winding temperature and core temperature calculated using the thermal model in the previous section; , The thresholds and decision variables are given by the thermal stability constraints and equipment parameters mentioned earlier. The specific calculation method for the thresholds has been given in the model construction section. This step mainly organizes the existing state variables and thresholds into optimization objective functions and constraints, and the derivation process will not be repeated. Here, we can only retain the final form of objective functions such as J1 and J2 and their physical meaning; it should be noted that the thresholds used are all from the previous subsection or equipment technical standards and are used as known parameters in the optimization.

[0130] Please refer to Figure 5 In some embodiments, the step of inputting the target carbon emissions into a preset multi-objective optimization function for solution to obtain the optimal combination of operating parameters for the converter transformer includes steps S501 to S502:

[0131] Step S501: Input the target carbon emission amount into a preset multi-objective optimization function, and use a multi-objective genetic algorithm to solve the multi-objective optimization function to obtain a solution set. Then, perform non-dominated sorting to stratify the obtained solution set to determine the non-dominated level, and calculate the crowding degree of each solution within the same non-dominated level.

[0132] In some embodiments, firstly, the target carbon emissions and other relevant parameters are input into a preset multi-objective optimization function, and then solved using a process based on the Non-Dominated Sorting Genetic Algorithm (NSGA-II). Firstly, in multi-objective optimization, the adjustment cycles for cooling speed, on-load tap position, and magnetic flux density are typically long, while the adjustment cycle for load rate is short. Therefore, under fixed conditions, this invention decomposes the load allocation problem into approximate sub-problems, and quickly solves for the optimal load rate using closed-form solutions, reducing the overall computational complexity of the optimization. The objective function of this sub-problem is: ;in: For each unit's load rate; The coefficient is a quadratic coefficient; The coefficient is a first-order coefficient; Base value capacity; Total load; Let this be the upper bound of the load factor. Then, ignoring the upper and lower bound constraints of the load factor, a closed-form solution to the approximate subproblem of load allocation is obtained using the Kuhn-Tucker conditions. When the closed-form solution exceeds the constraints, the load factor is taken as the constraint boundary value closest to the closed-form solution. This ensures that safety constraints are met while approximating the optimal solution as closely as possible, balancing safety and optimization performance. The formula for the closed-form solution is: ;in: It is a closed-form solution; These are Lagrange multipliers. Based on this, NSGA-II optimizes the complete decision vector, which includes load factor, cooling speed, tap position, and flux density. Then, to solve the aforementioned multi-objective optimization problem, NSGA-II is used, and a hybrid encoding scheme is designed considering the mixed characteristics of transformer decision variables. An adaptive strategy associated with extreme climate indices is introduced to improve the algorithm's convergence and optimization performance. The first step involves feasible projection and tap position mapping to handle variable boundaries and discrete characteristics. ;in: For any continuous variable component; Lower / upper bound; This is an interval projection operator. The on-load tap position is a discrete decision variable, but the crossover and mutation operations of NSGA-II generate continuous prototype integers, which cannot be directly used for equipment control. Therefore, a position mapping rule is designed to transform the continuous prototype integers into practically usable discrete positions, ensuring that the optimization results conform to the mechanical operating characteristics of on-load tap switches. ;in: The original integer; This is the set of gear levels. Then, the Pareto dominance relation is used to perform a non-dominated sorting of the solution set generated by the multi-objective genetic algorithm, thereby dividing the solution set into multiple levels. The Pareto dominance relation is defined as follows: for any two solutions x and y, ,in: This is a Pareto dominance relationship; For the first The objective function values ​​are sorted. Through this sorting, all non-dominated solutions, i.e., solutions not dominated by any other solutions, are assigned to the first non-dominated layer as the optimal candidate solution set. Subsequently, to maintain the diversity of solution distribution in the objective space, the crowding distance for each solution i within the same non-dominated layer is calculated, using the following formula: ;in: For the first Individual crowding distance; To be according to Sort adjacent values ​​in ascending order; The endpoint of the target range is defined as follows: the larger the congestion distance, the more "open" the solution is around, and the better the diversity of solutions.

[0133] In some embodiments, the expressions for the first-order coefficient and the second-order coefficient are:

[0134] ;

[0135] in, For the carbon factor of the power grid; Step size; Copper loss coefficient; For efficiency; This serves as the reference point for linearization.

[0136] In some embodiments, the copper loss coefficient As in the formula: ;in: Copper loss coefficient; Rated current; Used as a reference resistor; Temperature coefficient; For hotspot temperature; This is a reference temperature.

[0137] It's important to note that non-dominated sorting only distinguishes the hierarchical quality of individuals, but cannot measure the dispersion of individuals within the same hierarchical level. If individuals are overly concentrated within the same non-dominated level, it can lead to missing regions in the Pareto optimal front, reducing the selectivity of the optimization results. Therefore, crowding distance is introduced to quantify the spatial distribution density of individuals within the same non-dominated level, prioritizing the retention of individuals with good dispersion to ensure the integrity of the Pareto front. By quantifying the spatial distance between individuals and their neighbors, individuals with larger distances are prioritized, avoiding excessive clustering of individuals within the same target area. This ensures that the final Pareto optimal front covers more optimization scenarios, and in engineering applications, solutions for different regions can be selected according to actual needs.

[0138] By using this approximate subproblem and closed-form solution, the present invention can quickly solve for the load rate under fixed conditions, significantly reducing the variable dimension and computational load of multi-objective optimization, and adapting to online real-time optimization requirements.

[0139] Step S502: Based on the non-dominated level and the congestion degree, select the optimal solution from the solution set, and determine the optimal combination of operating parameters for the converter transformer based on the optimal solution. The multi-objective optimization function includes a carbon emission objective function, a thermal stability objective function, and an operating energy efficiency objective function.

[0140] In some embodiments, to select the optimal solution for final execution from the multi-objective optimization solution set and determine the combination of operating parameters, it is first necessary to construct an evaluation and selection system for individuals in the population. The core of this system is the non-dominated ranking based on Pareto dominance and the calculation of crowding density to measure the distribution density of individuals. The dominance relationship between individuals is constructed according to the following principles: for an individual solution... Dominating another body (recorded as) ), if and only if for all optimization objective functions The relevant formula is: Based on this dominance relationship, the algorithm performs a non-dominated sort on all individuals in the current generation, dividing them into multiple levels. The first non-dominated level contains the set of optimal solutions that are not dominated by any other individual. After completing the non-dominated sort, to further differentiate the quality of individuals within the same non-dominated level, a crowding distance is constructed. As a metric, the calculation formula is: ;in: For the first Individual crowding distance; To be according to Sort adjacent values ​​in ascending order; The endpoint of the target range is defined as follows: a larger crowding distance indicates a lower solution density around the individual, and a higher contribution to its diversity. Subsequently, the population undergoes iterative evolution through genetic operations to generate a new set of candidate solutions. After generating a new population through the aforementioned genetic operations and adaptive adjustment, non-dominated sorting and crowding calculation are performed iteratively. Finally, a selection strategy is implemented to determine the final optimal solution from the first non-dominated layer. This method employs a knee-point selection strategy, aiming to select the solution that achieves the best balance among multiple objectives. This strategy determines the final solution from the first non-dominated layer by solving the following optimization problem. This method employs a knee-point selection strategy, aiming to select a solution that achieves the best balance among multiple objectives. This strategy determines the final solution from the first non-dominated layer by solving the following optimization problem: ;in: To select a solution individual; It is the first non-dominated layer; These are weighting coefficients; Given the endpoint of the target range for this layer, this formula ensures that the selected solution is relatively balanced across all targets and is closest to the ideal point by minimizing the largest weighted normalized value among all targets.

[0141] Ultimately, based on the selected optimal solution Determine the optimal combination of operating parameters for the converter transformer. This optimal solution... The gene encoding directly corresponds to the decision variable, namely the load rate of each transformer at time t. Cooling equivalent speed On-load tap position and target magnetic flux density This combination of parameters constitutes the optimal operating parameter combination for directly controlling the operation of the converter transformer.

[0142] In some embodiments, population iterative evolution is performed through genetic operations to generate a new set of candidate solutions. Specifically, firstly, a binary tournament selection strategy is used to select parent individuals from the current population, with the selection criterion being the priority to select non-dominant hierarchical individuals. Lower (better) and crowded distance For larger individuals, the selection process can be represented as: ;in: For non-dominated levels (integer); For crowded distances; The superior parent individuals are indexed. Next, a simulated binary crossover (SBX) operation is performed on the selected parent individuals to generate new offspring. By simulating the crossover characteristics of binary encoding, efficient recombination of parent genes is achieved, generating offspring that possess both superior parent traits and a certain degree of variability. Specifically, this operation uses a random expansion factor. For two parent individuals and The genes were mixed, and the offspring were produced from the mixture. The calculation formula is: ;in: For expansion factor; The numbers are uniformly random. Cross-index; For the parent generation of real number components; This is the offspring component. Furthermore, a multinomial mutation operation is introduced for continuous decision variables. Through low-probability, controllable-amplitude mutations, new genes are injected into the population to maintain diversity. This low-probability, controllable-amplitude perturbation allows individuals to escape their current local optima and explore new feasible regions, while avoiding a sharp decline in individual quality due to excessive mutation amplitude. This ensures that the population maintains diversity while continuously converging towards the global optimum frontier, i.e., through a random perturbation... For individual components To maintain population diversity, small perturbations are made, resulting in mutated components. The calculation formula is: ;in: For disturbance quantity; The variation index; Lower / upper bound; For the parent generation; This refers to the mutated component.

[0143] It should be noted that extreme weather conditions can alter the operating characteristics of transformers, and fixed optimization weights and algorithm parameters may lead to optimization failure. Therefore, to improve the algorithm's adaptability under extreme weather conditions, a climate / ontology adaptive strategy is constructed to dynamically adjust the optimization parameters based on the real-time calculated average hotspot temperature. and extreme climate index Dynamically adjust the weights of carbon emission targets and variability The relevant formula is: ;in: For adaptive emission weights; Basic weights; The average temperature of the hot spots in parallel; This is the temperature limit / reference temperature. For continuous variables, the rate of variation; Lower / upper limit of the mutation rate; The slope of the Sigmoid function; This is the inflection point; This is an extreme climate index. Meanwhile, to uphold the principle of safety first, individuals operating near the safety boundary are penalized through a reduction in weighting coefficient. and Reduce its crowded distance This reduces the probability of it being retained to the next generation; the relevant formula is: ;in: To adjust the crowding distance; The weighting coefficient; The indicator function is (0 / 1); For the first Individual average hotspot temperature; This refers to the temperature margin; For the first Individual average cooling efficiency; This represents the lower limit of efficiency.

[0144] By solving this multi-objective optimization problem, the final optimal combination of operating parameters is a globally optimal solution that prioritizes all stability constraints while pursuing low carbon emissions and high efficiency.

[0145] In some embodiments, controlling the operation of the converter transformer based on the optimal combination of operating parameters includes: sending the optimal combination of operating parameters to the transformer control system to adjust at least one of the following in real time: load distribution, cooling system operating status, on-load tap position, and magnetic flux density of the converter transformer. Specifically, firstly, the optimal combination of operating parameters obtained in each optimization cycle is... ;in: The optimal parameters to be implemented in the current cycle; " indicates the optimized value; The current control time is sent to the transformer control system. Based on these parameters, the control system adjusts the actions of the corresponding actuators in real time, thereby achieving precise control over the load distribution of the converter transformer, the operating status of the cooling system, the on-load tap position and the magnetic flux density, so that it always dynamically maintains or approaches the current optimal operating point.

[0146] It should be noted that, to achieve closed-loop optimization, the optimal combination of operating parameters is written back to the basic carbon emission measurement model, and key parameters in the model are updated synchronously. Specifically, this includes updating the calculations of copper loss, iron loss, cooling efficiency, and carbon emissions using the set of optimal load rate, optimal cooling speed, optimal on-load tap position, and optimal target magnetic flux density. The update formula is as follows: .

[0147] Through the aforementioned rolling execution and closed-loop write-back mechanism, not only are optimization instructions transformed into control actions, but the dynamic synchronization between the basic model and the actual operating status of the equipment is also ensured. This effectively avoids the attenuation of optimization accuracy caused by model staticization, thereby ensuring that the converter transformer achieves continuous, accurate, low-carbon, safe, and efficient operation throughout its entire life cycle.

[0148] This invention acquires real-time power operation data and environmental data, converting the transformer's operating state into a quantifiable first carbon emission. This first carbon emission accurately identifies potential risk points that could lead to instability. By using a climate response factor to precisely map climate stress into carbon emission calculations, the second carbon emission accurately reflects the transformer's operating load under harsh conditions, providing a precise and environmentally adapted decision-making basis for stability control. Adaptive Kalman filtering filters out interference from sensor drift and sudden environmental changes, correcting unavoidable noise and errors in data acquisition and model calculations in real time. This results in a smoother and more reliable target carbon emission output, significantly improving the reliability of state perception and avoiding instability risks caused by control decisions based on distorted data. By solving a multi-objective optimization problem, the final optimal combination of operating parameters is a globally optimal solution that prioritizes all stability constraints while pursuing low carbon emissions and high efficiency. Controlling the transformer based on this combination proactively maintains its operating state within a safe and stable range, directly achieving the ultimate goal of improving operational stability. Compared to existing technologies, this invention improves the operational stability of converter transformers.

[0149] like Figure 6 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided;

[0150] An embodiment of the present invention provides an operation control system for a converter transformer, comprising: an acquisition module 100, a processing module 200, and a control module 300;

[0151] The acquisition module 100 is used to acquire the power operation data and environmental data of the converter transformer, and determine the first carbon emission based on the power operation data and the environmental data.

[0152] The processing module 200 is used to correct the first carbon emission amount based on a preset dynamic response function to obtain a second carbon emission amount, and to filter and correct the second carbon emission amount using an adaptive Kalman filter method to obtain a target carbon emission amount. The dynamic response function is constructed based on the power operation data, the environmental data, and the climate response factor, and the climate response factor is obtained after processing the environmental data.

[0153] The control module 300 is used to input the target carbon emission into a preset multi-objective optimization function for solving, to obtain the optimal combination of operating parameters for the converter transformer, and to control the operation of the converter transformer based on the optimal combination of operating parameters.

[0154] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the operation control method for converter transformers provided by any of the above-described method embodiments of the present invention.

[0155] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0156] Based on the above embodiments of the converter transformer operation control method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the converter transformer operation control method of any embodiment of the present invention.

[0157] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0158] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0159] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0160] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the operation control method for the converter transformer described in any of the above-described method embodiments of the present invention.

[0161] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0162] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method of operating control of a converter transformer, characterized by, The method comprises the following steps: acquiring power operation data and environmental data of a converter transformer, and determining a first carbon emission based on the power operation data and the environmental data; correcting the first carbon emission based on a preset dynamic response function to obtain a second carbon emission, and filtering and correcting the second carbon emission by using an adaptive Kalman filtering method to obtain a target carbon emission, wherein the dynamic response function is obtained by compounding the power operation data, the environmental data and a climate response factor, and the climate response factor is calculated based on the environmental data to obtain an extreme climate threshold and a duration index; inputting the target carbon emission into a preset multi-objective optimization function to obtain an optimal operation parameter combination of the converter transformer, and controlling the operation of the converter transformer based on the optimal operation parameter combination.

2. The operation control method of a converter transformer according to claim 1, characterized by, The method of determining the first carbon emission based on the power operation data and the environmental data comprises the following steps: calculating power loss of the transformer based on the power operation data; correcting an initial transformer efficiency by using temperature and humidity in the environmental data to obtain a target transformer efficiency; inputting the power loss, the target transformer efficiency and a power grid carbon factor into a preset carbon emission metering model to calculate the first carbon emission.

3. The operation control method of a converter transformer according to claim 1, characterized by, The method of correcting the first carbon emission based on the preset dynamic response function to obtain the second carbon emission comprises the following steps: inputting the environmental data and the power operation data into the dynamic response function to generate a temperature correction coefficient, a humidity correction coefficient and a load correction coefficient; combining the temperature correction coefficient, the humidity correction coefficient and the load correction coefficient to obtain a composite correction coefficient, and multiplying the first carbon emission by the composite correction coefficient to obtain the second carbon emission.

4. The operation control method of a converter transformer according to claim 1, characterized by, The method of obtaining the dynamic response function based on the power operation data, the environmental data and the climate response factor comprises the following steps: calculating an extreme climate threshold and a duration index based on the environmental data, and generating a climate response factor based on the extreme climate threshold and the duration index; performing high-order nonlinear regression analysis on the climate response factor to obtain a temperature correction factor, a humidity correction factor and a load correction factor; training a fuzzy correction factor by using an adaptive neural fuzzy inference system based on the environmental data and the power operation data, and training a time sequence correction factor by using a long short-term memory network; integrating the temperature correction factor, the humidity correction factor, the load correction factor, the fuzzy correction factor and the time sequence correction factor to obtain the dynamic response function.

5. The operation control method of a converter transformer according to claim 1, characterized by, The method of filtering and correcting the second carbon emission by using the adaptive Kalman filtering method to obtain the target carbon emission comprises the following steps: real-time monitoring an actual carbon emission value of the converter transformer; input the second carbon emission and the actual carbon emission value into an adaptive Kalman filtering algorithm to dynamically calculate a Kalman gain, and perform weighted fusion on the second carbon emission and the actual carbon emission value according to the Kalman gain to obtain the target carbon emission.

6. The operation control method of a converter transformer according to claim 1, characterized by, The target carbon emission is input into a preset multi-objective optimization function to obtain an optimal operation parameter combination of the converter transformer. The target carbon emission is input into a preset multi-objective optimization function, and a multi-objective genetic algorithm is used to solve the multi-objective optimization function to obtain a solution set. The obtained solution set is stratified by non-dominated sorting to determine a non-dominated level, and the crowding degree of each solution in the same non-dominated level is calculated. Based on the non-dominated level and the crowding degree, an optimal solution is selected from the solution set, and the optimal operation parameter combination of the converter transformer is determined based on the optimal solution. The multi-objective optimization function includes a carbon emission objective function, a thermal stability objective function, and an operation energy efficiency objective function.

7. The operation control method of a converter transformer according to any one of claims 1 to 6, characterized by, The optimal operation parameter combination is used to control the operation of the converter transformer, including: The optimal operation parameter combination is sent to a transformer control system to adjust at least one of the load distribution, the cooling system operation state, the on-load tap changer gear position, and the magnetic flux density of the converter transformer in real time.

8. An operating control system for a converter transformer, characterized by It includes: An acquisition module, a processing module, and a control module; The acquisition module is configured to acquire power operation data and environmental data of the converter transformer, and determine a first carbon emission based on the power operation data and the environmental data; The processing module is configured to correct the first carbon emission based on a preset dynamic response function to obtain a second carbon emission, and filter and correct the second carbon emission using an adaptive Kalman filtering method to obtain a target carbon emission. The dynamic response function is obtained by combining the power operation data, the environmental data, and a climate response factor. The climate response factor is obtained based on an extreme climate threshold value and a duration index calculated based on the environmental data. The control module is configured to input the target carbon emission into a preset multi-objective optimization function to obtain an optimal operation parameter combination of the converter transformer, and control the operation of the converter transformer based on the optimal operation parameter combination.

9. A terminal device, comprising: It includes: One or more processors; A memory coupled to the processor for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the steps of the converter transformer operation control method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It includes: A stored computer program, wherein when the computer program is running, the device where the computer readable storage medium is located performs the steps of the converter transformer operation control method according to any one of claims 1-7.

Images