A three-phase current automatic balance adjustment method and adjustment system

Abstract

The present invention provides a three-phase current automatic balancing and regulating method and a regulating system, which relate to the technical field of smart grids. The three-phase current of a distribution transformer is collected to make a balance judgment, and when the judgment result is unbalanced, a regulating mechanism is triggered; after the regulating mechanism is triggered, the current characteristics of the three-phase current are input into a pre-trained strategy model for strategy matching, and a switching instruction corresponding to the current three-phase current is obtained; after receiving the switching instruction, the capacitor group dynamically adjusts the reactive compensation amount according to the instruction to compensate for the imbalance of the three-phase current; at the same time, the system continues to monitor the three-phase current of the distribution transformer, and if the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor group are executed again until the three-phase current reaches a balanced state. The present invention realizes automatic identification and regulation of the imbalance of the three-phase current, significantly improves the operation efficiency and stability of the distribution transformer, and reduces the operation and maintenance costs and safety risks.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a three-phase current automatic balancing and regulating method and a regulating system. Background Art

[0002] In the power system, the distribution transformer is a key device for power transmission and distribution. Its operating status is directly related to the stability of the power grid and the quality of power supply. However, in the actual operation process, due to various factors such as load changes, equipment aging, wiring errors, etc., the distribution transformer often has the problem of three-phase current imbalance. The three-phase current imbalance will not only increase the transformer loss and reduce the efficiency, but also may cause safety hazards such as equipment overheating and voltage fluctuations, seriously affecting the stable operation of the power grid and the user's power experience.

[0003] At present, the traditional solution to the problem of three-phase current imbalance in distribution transformers mainly relies on manual inspection and manual adjustment. This method is not only inefficient, but also difficult to accurately determine the specific cause and degree of imbalance. It often requires multiple trials and errors and long power outages to achieve three-phase current balance. In addition, due to the limitations and uncertainties of manual operation, it may also bring additional safety risks and maintenance costs.

[0004] Therefore, it is necessary to provide a three-phase current automatic balancing adjustment method and adjustment system to solve the above technical problems. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a three-phase current automatic balancing adjustment method and adjustment system to achieve the beneficial effect of ...

[0006] The present invention provides a three-phase current automatic balancing and adjusting method, which is applied to a distribution transformer, and the distribution transformer is equipped with a capacitor bank, and the adjusting method comprises:

[0007] Collect the three-phase current of the distribution transformer;

[0008] Performing a balance judgment based on the three-phase current, and triggering a regulation mechanism when the judgment result is unbalanced;

[0009] After the regulation mechanism is triggered, the current characteristics of the three-phase current are input into a pre-trained strategy model for strategy matching to obtain a switching instruction corresponding to the current three-phase current, wherein the strategy model is a neural network model trained by a historical three-phase current data set;

[0010] The capacitor bank responds to the switching instruction and dynamically adjusts the reactive compensation amount according to the switching instruction to compensate for the three-phase current imbalance;

[0011] After the capacitor bank adjusts the reactive compensation amount, the three-phase current of the distribution transformer continues to be monitored. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor bank are executed again until the three-phase current is in a balanced state.

[0012] Preferably, the three-phase current is collected at a preset sampling frequency by current transformers installed on phases A, B and C of the distribution transformer.

[0013] Preferably, the balance judgment is performed based on the three-phase current, and when the judgment result is unbalanced, the adjustment mechanism is triggered, including:

[0014] The unbalance degree is calculated based on the three-phase current, wherein the calculation formula of the unbalance degree is:

[0015]

[0016] Among them, δ is the imbalance degree, I a , I b , I c are the effective current values ​​of phase A, phase B and phase C of the distribution transformer, max(I a , I b , I c ) is the three-phase current I a , I b , I c The maximum value in min(I a , I b , I c ) is the three-phase current I a , I b , I c The minimum value in ;

[0017] The smoothness is calculated based on the three-phase current, wherein the calculation formula of the smoothness is:

[0018]

[0019] Among them, S is the smoothness, I i is the average current value of the same sampling point i in the three-phase current, For all I i The average value of

[0020] Compare the preset imbalance threshold with the imbalance, and the smoothness threshold with the smoothness, and when the imbalance exceeds the imbalance threshold and / or the smoothness exceeds the smoothness threshold, determine that the three-phase current is in an unbalanced state, and trigger the adjustment mechanism.

[0021] Preferably, the training of the strategy model includes:

[0022] Collect historical three-phase current data and extract current features;

[0023] The neural network model is forward propagated using historical three-phase current data, and then the parameters of the neural network model are updated through backpropagation to minimize the loss function, and finally a trained strategy model is obtained.

[0024] Preferably, the step of inputting the current characteristics of the three-phase current into a pre-trained strategy model for strategy matching to obtain a switching instruction corresponding to the current three-phase current includes:

[0025] Extracting current characteristics of the three-phase current, wherein the current characteristics include frequency, peak value and effective value;

[0026] The extracted current characteristics of the three-phase current are used as input of the strategy model, and forward propagation is performed to output a matching adjustment strategy;

[0027] The adjustment strategy output by the strategy model is decoded, and a switching instruction is generated and sent to the capacitor bank, wherein the switching instruction includes an adjustment target, an adjustment time, and an adjustment amount.

[0028] Preferably, the capacitor bank responds to the switching instruction and dynamically adjusts the reactive compensation amount according to the adjustment target, adjustment time and adjustment amount contained in the switching instruction, including:

[0029] The control unit of the capacitor bank receives the switching instruction output from the strategy model and parses out the regulation target, regulation time and regulation amount;

[0030] The reactive power compensation amount is dynamically adjusted according to the adjustment target, adjustment time and adjustment amount in the switching instruction.

[0031] Preferably, the dynamic adjustment of the reactive power compensation amount according to the adjustment target, adjustment time and adjustment amount in the switching instruction includes:

[0032] According to the adjustment target in the switching instruction, the control unit of the capacitor bank determines the capacitor bank unit that needs to be adjusted;

[0033] According to the adjustment time in the switching instruction, the control unit of the capacitor bank sets the adjustment period of the reactive compensation amount;

[0034] According to the adjustment amount in the switching instruction, the control unit of the capacitor bank calculates the reactive compensation amount that needs to be adjusted;

[0035] The capacitor group is controlled to dynamically adjust the reactive compensation amount according to the capacitor group unit that needs to be adjusted, the adjustment period of the reactive compensation amount, and the reactive compensation amount that needs to be adjusted determined by the control unit of the capacitor group.

[0036] The present invention also provides a three-phase current automatic balancing and regulating system, which is used to execute a three-phase current automatic balancing and regulating method, and is applied to a distribution transformer, and the distribution transformer is equipped with a capacitor bank, and the regulating system comprises:

[0037] Current acquisition module, used to collect the three-phase current of the distribution transformer;

[0038] A balance judgment module, used to make a balance judgment based on the three-phase current, and trigger a regulation mechanism when the judgment result is unbalanced;

[0039] A strategy matching module, for inputting the current characteristics of the three-phase current into a pre-trained strategy model for strategy matching after the regulation mechanism is triggered, so as to obtain a switching instruction corresponding to the current three-phase current, wherein the strategy model is a neural network model trained by a historical three-phase current data set;

[0040] A capacitor bank control module, used for the capacitor bank to respond to the switching instruction and dynamically adjust the reactive compensation amount according to the switching instruction to compensate for the three-phase current imbalance;

[0041] The closed-loop monitoring module is used to continue monitoring the three-phase current of the distribution transformer after the capacitor bank adjusts the reactive compensation amount. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor bank are executed again until the three-phase current is in a balanced state.

[0042] Compared with the related art, the three-phase current automatic balancing adjustment method and adjustment system provided by the present invention have the following beneficial effects:

[0043] The present invention collects the three-phase current of the distribution transformer to make a balance judgment, and triggers the adjustment mechanism when the judgment result is unbalanced; after the adjustment mechanism is triggered, the current characteristics of the three-phase current are input into the pre-trained strategy model for strategy matching to obtain the switching instruction corresponding to the current three-phase current; the strategy model is a neural network model trained by the historical three-phase current data set, which can accurately identify the unbalanced characteristics of the three-phase current and quickly generate the optimal switching instruction; after receiving the switching instruction, the capacitor group will dynamically adjust the reactive compensation amount according to the instruction to compensate for the imbalance of the three-phase current; at the same time, the system will continue to monitor the three-phase current of the distribution transformer. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor group will be executed again until the three-phase current reaches a balanced state. The present invention realizes automatic identification and adjustment of the imbalance of the three-phase current, significantly improves the operation efficiency and stability of the distribution transformer, and reduces the operation and maintenance costs and safety risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A flow chart of a three-phase current automatic balancing and adjusting method provided by the present invention;

[0045] Figure 2 A module structure diagram of a three-phase current automatic balancing and regulating system provided by the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only the parts related to the present invention, rather than all structures, are shown in the accompanying drawings. In addition, the embodiments of the present invention and the features in the embodiments may be combined with each other without conflict.

[0047] It should also be noted that, for ease of description, only the part relevant to the present invention but not all content is shown in the accompanying drawings. It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processing or methods depicted as flow charts. Although the flow chart describes each operation (or step) as sequential processing, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. When its operation is completed, the processing can be terminated, but it can also have additional steps not included in the accompanying drawings. The processing can correspond to methods, functions, procedures, subroutines, subprograms, etc.

[0048] Embodiment 1

[0049] The present invention provides a three-phase current automatic balancing and adjusting method, which is applied to a distribution transformer, and the distribution transformer is equipped with a capacitor bank, referring to Figure 1 As shown, the adjustment method includes:

[0050] S1: Collect the three-phase current of the distribution transformer.

[0051] In this embodiment, three-phase current collection is achieved through current transformers installed on the A-phase, B-phase and C-phase lines of the distribution transformer. These current transformers continuously collect current signals at a preset sampling frequency (for example, 100 times per second) and convert them into digital signals, ensuring real-time and accurate current data collection, and providing a solid data foundation for subsequent balance judgment and adjustment mechanisms.

[0052] S2: Perform a balance judgment based on the three-phase current, and trigger a regulation mechanism when the judgment result is unbalanced.

[0053] In this embodiment, two key indicators, imbalance and smoothness, are calculated based on the three-phase current data. The imbalance is measured by calculating the ratio of the difference between the maximum and minimum values ​​of the three-phase current to the average value, while the smoothness is to evaluate the fluctuation of the three-phase current over time. When the imbalance exceeds the preset threshold and / or the smoothness exceeds the preset threshold, it is determined that the three-phase current is in an unbalanced state, triggering the adjustment mechanism, thereby identifying the unbalanced state in the power system, providing a start signal for subsequent automatic adjustment, and avoiding power loss and equipment damage caused by current imbalance.

[0054] S3: After the regulation mechanism is triggered, the current characteristics of the three-phase current are input into a pre-trained strategy model for strategy matching to obtain switching instructions corresponding to the current three-phase current, wherein the strategy model is a neural network model trained using a historical three-phase current data set.

[0055] In this embodiment, a neural network model trained based on a historical three-phase current data set uses a large amount of historical current data and corresponding switching instructions as training samples during the model training process, and continuously optimizes the weight parameters of the neural network through the back propagation algorithm to minimize the error between the predicted instructions and the actual instructions. In practical applications, when a three-phase current imbalance is detected, the model inputs the current current characteristics and outputs the most appropriate switching instructions, including the adjustment target (capacitor bank), adjustment time and adjustment amount, thereby realizing the generation of intelligent adjustment strategies, improving the accuracy and efficiency of adjustment, reducing the need for human intervention, and improving the automation level and response speed of the system.

[0056] S4: The capacitor bank responds to the switching instruction and dynamically adjusts the reactive compensation amount according to the switching instruction to compensate for the three-phase current imbalance.

[0057] In this embodiment, after receiving the switching instruction output by the strategy matching module, the adjustment action specified in the instruction is immediately executed, including increasing or decreasing the number of capacitor groups put into operation, and adjusting the output level of the capacitor to achieve dynamic adjustment of the reactive compensation amount, so as to be able to quickly and accurately adjust the reactive compensation amount and effectively compensate for the imbalance of the three-phase current, thereby improving the stability of the power system and the power supply quality.

[0058] S5: After the capacitor bank adjusts the reactive compensation amount, the three-phase current of the distribution transformer continues to be monitored. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor bank are executed again until the three-phase current is in a balanced state.

[0059] In this embodiment, after the capacitor bank adjusts the reactive compensation amount, the balance state of the three-phase current is continuously monitored. If the monitoring result shows that the current is still unbalanced, the strategy matching and capacitor bank adjustment will be automatically triggered again, and the generation and execution process of the switching instruction will be repeated until the three-phase current reaches the preset balance standard to ensure continuous monitoring and closed-loop control capabilities. Even in the case of changing load conditions or external interference, the balance state of the three-phase current can be maintained, thereby improving the robustness and adaptability of the system.

[0060] Specifically, the three-phase current is collected by current transformers installed on phases A, B and C of the distribution transformer according to a preset sampling frequency.

[0061] In this embodiment, three-phase current collection is achieved by installing current transformers (CTs) on each phase of the distribution transformer (i.e., phase A, phase B, and phase C). These current transformers are designed in a ring shape, tightly surrounding the distribution line, and can accurately sense the current changes flowing through the wire. In order to ensure the real-time and accuracy of the data, the current transformer works at a preset sampling frequency. For example, the sampling frequency is set to 100 times per second (100Hz), which means that in every second, the current transformer will capture and record 100 instantaneous current values, ensuring the continuity and high density of the current data.

[0062] The collected analog current signal is then converted into a digital signal through a built-in analog-to-digital converter (ADC), which is convenient for subsequent electronic equipment processing and storage.

[0063] Specifically, step S2 includes the following steps:

[0064] S201: Calculate the imbalance degree based on the three-phase current, wherein the calculation formula of the imbalance degree is:

[0065]

[0066] Among them, δ is the imbalance degree, I a , I b , I c are the effective current values ​​of phase A, phase B and phase C of the distribution transformer, max(I a , I b , I c ) is the three-phase current I a , I b , I c The maximum value in min(I a , I b , I c ) is the three-phase current I a , I b , I cThe minimum value in .

[0067] In this embodiment, the degree of imbalance is calculated by obtaining the effective value of the three-phase current of the distribution transformer, that is, the effective value of the current of phase A, phase B, and phase C, and then finding the maximum and minimum values ​​of these three values. The degree of imbalance is quantified by calculating the percentage of the difference between the maximum and minimum values ​​of the three-phase current to the average current value through the above formula, which can intuitively reflect the differences between the three-phase currents.

[0068] S202: Calculate smoothness based on the three-phase current, wherein the calculation formula of the smoothness is:

[0069]

[0070] Among them, S is the smoothness, I i is the average current value of the same sampling point i in the three-phase current, and I is the average current value of all I i The average value of .

[0071] In this embodiment, the smoothness index quantifies the degree of current fluctuation over time, and can reveal possible unstable factors in the power system, such as frequent load changes or equipment failures. At the same time, combined with the calculation of imbalance and smoothness, the system can comprehensively evaluate the operating status of the power system and ensure that the regulation mechanism not only pays attention to the balance of the three-phase current, but also considers the stability of the current, thereby improving the overall stability of the power system.

[0072] S203: Compare the preset imbalance threshold with the imbalance, and the smoothness threshold with the smoothness, and when the imbalance exceeds the imbalance threshold and / or the smoothness exceeds the smoothness threshold, determine that the three-phase current is in an unbalanced state, and trigger the adjustment mechanism.

[0073] In this embodiment, the imbalance threshold and the smoothness threshold are preset. When the calculated imbalance exceeds the imbalance threshold and / or the smoothness exceeds the smoothness threshold, it is determined that the three-phase current is in an unbalanced state and the adjustment mechanism is triggered immediately.

[0074] Among them, the imbalance threshold and smoothness threshold settings can be adjusted according to different power system requirements and environmental conditions, making the regulation strategy more flexible and adaptable, and able to meet the diverse and complex power system management needs.

[0075] Specifically, the training of the strategy model includes:

[0076] Collect historical three-phase current data and extract current features.

[0077] In this embodiment, the training of the strategy model begins with the collection of historical three-phase current data, which usually comes from the daily operation records of the power system, including but not limited to the three-phase current sampling data in the past few months or years. The data collection stage needs to ensure that various operating conditions are covered, such as different time periods (peak hours, off-peak hours), seasonal changes, special climate conditions, etc., to enhance the generalization ability of the model.

[0078] The collected historical three-phase current data needs to be preprocessed, including data cleaning (removing outliers and missing values), data standardization (ensuring that different features are on the same scale) and feature extraction. Feature extraction is a key step, which extracts information from the original current data that is helpful for unbalanced state judgment and regulation strategy generation. Current features include but are not limited to frequency, peak value and effective value.

[0079] The neural network model is forward propagated using historical three-phase current data, and then the parameters of the neural network model are updated through backpropagation to minimize the loss function, and finally a trained strategy model is obtained.

[0080] In this embodiment, after feature extraction is completed, the extracted features and corresponding switching instructions (as labels) are used to train the neural network model. The training process mainly includes two stages: forward propagation and back propagation:

[0081] Forward propagation: Input the feature data into the neural network, calculate the weighted sum of neurons and the output of the activation function layer by layer, and finally get the predicted adjustment instructions. This process is carried out sequentially, and the output of each layer serves as the input of the next layer.

[0082] Back propagation: After the forward propagation is completed, the difference between the predicted switching instructions and the actual switching instructions (loss function) is calculated, and then the weight parameters of the neural network are adjusted layer by layer through the back propagation algorithm to minimize the loss function.

[0083] This process will be iterated repeatedly until the model converges, that is, the value of the loss function no longer decreases significantly. At this time, the model training is considered to be completed, so that the neural network model can automatically adjust the internal parameters through the back propagation algorithm, and learn the complex mapping relationship from current characteristics to switching instructions. It can make reasonable predictions even when faced with new data, demonstrating strong adaptive adjustment capabilities.

[0084] Specifically, step S3 includes the following steps:

[0085] S301: extracting current characteristics of the three-phase current, wherein the current characteristics include frequency, peak value and effective value.

[0086] In this embodiment, this step focuses on extracting current features from the current three-phase current data. These features include frequency, peak value, and effective value, which are crucial for understanding and characterizing the characteristics of the three-phase current. The following is the specific operation process:

[0087] Frequency characteristics: By performing fast Fourier transform (FFT) on the three-phase current signal, the spectrum of the current can be obtained and the main frequency components can be extracted from it, which helps to identify harmonic interference and fundamental frequency in the system.

[0088] Peak characteristics: Measures the maximum amplitude of three-phase current, which is an important indicator for evaluating the transient performance of the power system. The size of the peak value directly reflects the extreme conditions of the current and is very critical for overload protection and fault detection.

[0089] Effective value characteristics: Calculate the effective value of the three-phase current, that is, the root mean square value (RMS). The effective value can reflect the actual working intensity of the current and is a commonly used parameter in power system design and analysis.

[0090] By extracting frequency, peak and effective value features, the strategy model can more comprehensively understand the dynamic characteristics of the three-phase current and provide a more accurate data basis for subsequent strategy generation. At the same time, feature extraction enables the model to rely not only on the original current waveform, but also on more intuitive physical quantities, which helps to improve the transparency and explainability of model decisions.

[0091] S302: Using the extracted current characteristics of the three-phase current as the input of the strategy model, performing forward propagation, and outputting a matching adjustment strategy.

[0092] In this embodiment, after feature extraction, these current features need to be input into the strategy model for forward propagation to generate a matching adjustment strategy. This process includes:

[0093] The frequency, peak value and RMS current features are fed as input to the trained strategy model.

[0094] The model performs forward propagation and converts the input features into predicted adjustment strategies through calculations of multi-layer neural networks. The adjustment strategies here include a variety of control actions implemented on the capacitor bank to achieve the goal of current balance, achieve rapid response to current current characteristics, and generate adjustment strategies in real time, thereby improving the real-time and flexibility of power system regulation.

[0095] S303: Decode the adjustment strategy output by the strategy model, generate a switching instruction, and send it to the capacitor bank, wherein the switching instruction includes an adjustment target, an adjustment time, and an adjustment amount.

[0096] In this embodiment, the adjustment strategy output by the model needs to be decoded into a specific switching instruction and then sent to the capacitor bank for execution. The specific steps include:

[0097] Decoding adjustment strategy: The abstract strategy generated by the strategy model is decoded into specific capacitor switching instructions. For example, if the strategy recommends increasing capacitor compensation, a corresponding capacitor switching instruction is generated; otherwise, a switching instruction is generated.

[0098] Generate switching instructions: The switching instructions should include details such as the regulation target (such as capacitor bank), regulation time (when to execute the instruction) and regulation amount (how many capacitors to put in or remove).

[0099] Send instructions to the capacitor bank: Send the generated instructions to the capacitor control system through the communication interface, which will perform specific capacitor switching operations.

[0100] Specifically, step S4 includes the following steps:

[0101] S401: The control unit of the capacitor bank receives a switching instruction output from the strategy model and parses out a regulation target, a regulation time and a regulation amount.

[0102] In this embodiment, the switching instruction is generated by the strategy model based on the previously extracted current characteristics. The core of this step is to correctly parse the key information in the instruction for subsequent dynamic adjustment. The specific operations include:

[0103] Receiving instructions: The control unit of the capacitor bank receives switching instructions from the strategy model through a dedicated communication channel.

[0104] Parsing instructions: The control unit parses the received instructions and extracts the adjustment target (i.e. the capacitor group that needs to be adjusted), the adjustment time (i.e. the time point when the instruction is executed) and the adjustment amount (i.e. the number or capacity of capacitors put into or removed).

[0105] Ensure that the control unit can accurately identify the intention of the strategy model to avoid improper operations caused by instruction parsing errors; at the same time, parsing the adjustment time information in the instruction helps the control unit prepare in advance or delay execution to ensure that the adjustment action is consistent with the power grid demand.

[0106] S402: Dynamically adjust the reactive power compensation amount according to the adjustment target, adjustment time and adjustment amount in the switching instruction.

[0107] In this embodiment, the specific process is as follows:

[0108] Determine the adjustment plan: Based on the adjustment amount in the instruction, calculate the number and capacity of capacitors that need to be put into or removed to meet the required reactive power compensation changes.

[0109] Execute adjustment action: At the specified adjustment time point, the control unit operates the switching of the capacitor through the control circuit to achieve dynamic increase or decrease of the reactive compensation amount.

[0110] By dynamically adjusting the reactive power compensation amount, it is possible to quickly respond to changes in the reactive power demand of the power grid and improve the stability and efficiency of the power system.

[0111] Specifically, step S402 includes:

[0112] S402a: According to the adjustment target in the switching instruction, the control unit of the capacitor bank determines the capacitor bank unit that needs to be adjusted.

[0113] In this embodiment, this step requires the control unit of the capacitor bank to determine which capacitor bank units need to be adjusted based on the adjustment target in the switching instruction. This step is the basis of the entire dynamic adjustment process, ensuring that subsequent operations are only performed on specific capacitor banks, avoiding unnecessary energy loss and equipment wear.

[0114] By accurately identifying the adjustment target, the control unit can quickly locate the capacitor bank unit that needs to be adjusted, improving the adjustment efficiency and accuracy. By adjusting only the necessary capacitor banks, blind operation of all capacitor banks is avoided, saving energy and extending equipment life.

[0115] S402b: According to the adjustment time in the switching instruction, the control unit of the capacitor bank sets an adjustment period of the reactive compensation amount.

[0116] In this embodiment, this step mainly involves setting the reactive compensation adjustment period by the control unit according to the adjustment time in the switching instruction. This setting determines when to start and complete the reactive compensation adjustment, ensuring that the adjustment action matches the demand cycle of the power grid.

[0117] By setting the adjustment period, the control unit can accurately start reactive power compensation adjustment when grid demand is at its peak or trough, thus enhancing the system's adaptability to load changes. At the same time, setting the adjustment period in advance makes the entire adjustment process more planned, helping to prevent potential grid instability factors and improve the overall safety of the power system.

[0118] S402c: According to the adjustment amount in the switching instruction, the control unit of the capacitor bank calculates the reactive compensation amount that needs to be adjusted.

[0119] In this embodiment, this step requires the control unit to calculate the reactive compensation amount that actually needs to be adjusted based on the adjustment amount in the switching instruction. This calculation is based on the current grid state and the preset target value to ensure that the adjustment of the reactive compensation amount can meet the immediate demand without being excessive or insufficient.

[0120] By accurately calculating the amount of reactive compensation that needs to be adjusted, over-compensation or under-compensation is avoided, and the accuracy and efficiency of reactive compensation are improved; at the same time, reasonable calculation of reactive compensation helps to reduce unnecessary energy consumption, reduce operating costs, and improve economic benefits.

[0121] S402d: controlling the capacitor group to dynamically adjust the reactive compensation amount according to the capacitor group unit that needs to be adjusted, the adjustment period of the reactive compensation amount, and the reactive compensation amount that needs to be adjusted determined by the control unit of the capacitor group.

[0122] In this embodiment, this step integrates the information of the first three steps, and the control unit performs dynamic adjustment of the reactive compensation amount of the capacitor group, specifically: according to the determined capacitor group unit, adjustment period and reactive compensation amount, the dynamic change of reactive compensation is achieved by controlling the switching of the capacitor.

[0123] By comprehensively considering multiple factors, the control unit is able to achieve fine control of the capacitor group, demonstrating the system's coordination and scheduling capabilities in complex environments; at the same time, dynamic adjustment of reactive power compensation helps maintain the voltage stability of the power grid and improves the overall stability of the power system and power supply reliability.

[0124] The working principle of a three-phase current automatic balancing and regulating method provided by the present invention is as follows: by collecting the three-phase current of the distribution transformer, a balance judgment is performed, and when the judgment result is unbalanced, the regulating mechanism is triggered; after the regulating mechanism is triggered, the current characteristics of the three-phase current are input into the pre-trained strategy model for strategy matching to obtain a switching instruction corresponding to the current three-phase current; the strategy model is a neural network model trained by a historical three-phase current data set, which can accurately identify the unbalanced characteristics of the three-phase current and quickly generate the optimal switching instruction; after receiving the switching instruction, the capacitor group will dynamically adjust the reactive compensation amount according to the instruction to compensate for the imbalance of the three-phase current; at the same time, the system will continue to monitor the three-phase current of the distribution transformer. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor group will be executed again until the three-phase current reaches a balanced state. The present invention realizes automatic identification and regulation of three-phase current imbalance, significantly improves the operating efficiency and stability of the distribution transformer, and reduces operation and maintenance costs and safety risks.

[0125] Embodiment 2

[0126] The present invention also provides a three-phase current automatic balancing and regulating system, which is used to execute a three-phase current automatic balancing and regulating method, and is applied to a distribution transformer, and the distribution transformer is equipped with a capacitor bank, referring to Figure 2 As shown, the regulating system includes:

[0127] The current acquisition module 100 is used to acquire the three-phase current of the distribution transformer;

[0128] A balance judgment module 200 is used to make a balance judgment based on the three-phase current and trigger a regulation mechanism when the judgment result is unbalanced;

[0129] A strategy matching module 300 is used to input the current characteristics of the three-phase current into a pre-trained strategy model for strategy matching after the adjustment mechanism is triggered, so as to obtain a switching instruction corresponding to the current three-phase current, wherein the strategy model is a neural network model trained by a historical three-phase current data set;

[0130] A capacitor bank control module 400, configured for causing the capacitor bank to respond to the switching instruction and dynamically adjust the reactive compensation amount according to the switching instruction to compensate for the three-phase current imbalance;

[0131] The closed-loop monitoring module 500 is used to continue monitoring the three-phase current of the distribution transformer after the capacitor bank adjusts the reactive compensation amount. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor bank are executed again until the three-phase current is in a balanced state.

[0132] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0134] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. A three-phase current automatic balancing and regulating method, applied to a distribution transformer, wherein the distribution transformer is provided with a capacitor bank, characterized in that: Adjustment methods include: Collect the three-phase current of the distribution transformer; Performing a balance judgment based on the three-phase current, and triggering a regulation mechanism when the judgment result is unbalanced; After the regulation mechanism is triggered, the current characteristics of the three-phase current are input into a pre-trained strategy model for strategy matching to obtain a switching instruction corresponding to the current three-phase current, wherein the strategy model is a neural network model trained by a historical three-phase current data set; The capacitor bank responds to the switching instruction and dynamically adjusts the reactive compensation amount according to the switching instruction to compensate for the three-phase current imbalance; After the capacitor bank adjusts the reactive compensation amount, the three-phase current of the distribution transformer continues to be monitored. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor bank are executed again until the three-phase current is in a balanced state; The three-phase current is collected at a preset sampling frequency by current transformers installed on phases A, B and C of the distribution transformer; The balance judgment is performed based on the three-phase current, and when the judgment result is unbalanced, the adjustment mechanism is triggered, including: The unbalance degree is calculated based on the three-phase current, wherein the calculation formula of the unbalance degree is: Among them, δ is the imbalance degree, I a , I b , I c are the effective current values ​​of phase A, phase B and phase C of the distribution transformer, max(I a , I b , I c ) is the three-phase current I a , I b , I c The maximum value in min(I a , I b , I c ) is the three-phase current I a , I b , I c The minimum value in ; The smoothness is calculated based on the three-phase current, wherein the calculation formula of the smoothness is: Among them, S is the smoothness, I i is the average current value of the same sampling point i in the three-phase current, For all I i The average value of Compare the preset imbalance threshold with the imbalance, and the smoothness threshold with the smoothness, and when the imbalance exceeds the imbalance threshold and / or the smoothness exceeds the smoothness threshold, determine that the three-phase current is in an unbalanced state, and trigger the adjustment mechanism.

2. A three-phase current automatic balancing and adjusting method according to claim 1, characterized in that: The training of the strategy model includes: Collect historical three-phase current data and extract current features; The neural network model is forward propagated using historical three-phase current data, and then the parameters of the neural network model are updated through backpropagation to minimize the loss function, and finally a trained strategy model is obtained.

3. A three-phase current automatic balancing and adjusting method according to claim 2, characterized in that: The inputting the current characteristics of the three-phase current into the pre-trained strategy model for strategy matching to obtain the switching instruction corresponding to the current three-phase current includes: Extracting current characteristics of the three-phase current, wherein the current characteristics include frequency, peak value and effective value; The extracted current characteristics of the three-phase current are used as input of the strategy model, and forward propagation is performed to output a matching adjustment strategy; The adjustment strategy output by the strategy model is decoded, and a switching instruction is generated and sent to the capacitor bank, wherein the switching instruction includes an adjustment target, an adjustment time, and an adjustment amount.

4. A three-phase current automatic balancing and adjusting method according to claim 3, characterized in that: The capacitor bank responds to the switching instruction and dynamically adjusts the reactive compensation amount according to the adjustment target, adjustment time and adjustment amount contained in the switching instruction, including: The control unit of the capacitor bank receives the switching instruction output from the strategy model and parses out the regulation target, regulation time and regulation amount; The reactive power compensation amount is dynamically adjusted according to the adjustment target, adjustment time and adjustment amount in the switching instruction.

5. A three-phase current automatic balancing and adjusting method according to claim 4, characterized in that: The dynamic adjustment of reactive power compensation according to the adjustment target, adjustment time and adjustment amount in the switching instruction includes: According to the adjustment target in the switching instruction, the control unit of the capacitor bank determines the capacitor bank unit that needs to be adjusted; According to the adjustment time in the switching instruction, the control unit of the capacitor bank sets the adjustment period of the reactive compensation amount; According to the adjustment amount in the switching instruction, the control unit of the capacitor bank calculates the reactive compensation amount that needs to be adjusted; The capacitor group is controlled to dynamically adjust the reactive compensation amount according to the capacitor group unit that needs to be adjusted, the adjustment period of the reactive compensation amount, and the reactive compensation amount that needs to be adjusted determined by the control unit of the capacitor group.

6. A three-phase current automatic balancing and regulating system, which executes a three-phase current automatic balancing and regulating method as described in any one of claims 1 to 5, and is applied to a distribution transformer, and the distribution transformer is equipped with a capacitor bank, characterized in that: The regulation system includes: Current acquisition module, used to collect the three-phase current of the distribution transformer; A balance judgment module, used to make a balance judgment based on the three-phase current, and trigger a regulation mechanism when the judgment result is unbalanced; A strategy matching module, for inputting the current characteristics of the three-phase current into a pre-trained strategy model for strategy matching after the regulation mechanism is triggered, so as to obtain a switching instruction corresponding to the current three-phase current, wherein the strategy model is a neural network model trained by a historical three-phase current data set; A capacitor bank control module, used for the capacitor bank to respond to the switching instruction and dynamically adjust the reactive compensation amount according to the switching instruction to compensate for the three-phase current imbalance; The closed-loop monitoring module is used to continue monitoring the three-phase current of the distribution transformer after the capacitor bank adjusts the reactive compensation amount. If the judgment result is still unbalanced, the output of the switching instruction and the adjustment of the capacitor bank are executed again until the three-phase current is in a balanced state.

Images