A method and system for dynamic current sharing control of modular transformers

Abstract

The application belongs to the technical field of power electronics, and particularly relates to a dynamic current equalization control method and system of a modular transformer. The method adopts offline multi-target pre-optimization and online dynamic correction. In the offline stage, a multi-target optimization function is constructed based on a residual factor matrix, and a Pareto optimal solution set is solved to determine the initial topology and parameters of an adjustable current balancing transformer. In the online stage, through distributed current collection, combined with residual factor sensitivity analysis, dominant inductance parameters are selectively identified, and equivalent inductance is dynamically adjusted to realize real-time current equalization. The application can adapt to system parameter drift and load change in real time, solves the poor engineering practicability problem caused by single target optimization, and is suitable for scenarios such as data centers, electric vehicle super fast charging, distributed energy storage and solid state transformers.

Description

Technical Field

[0001] This invention belongs to the field of power electronics technology, specifically relating to a dynamic current sharing control method and system for a modular transformer. Background Technology

[0002] When modular transformers are operated in parallel, factors such as manufacturing tolerances, wiring differences, temperature changes during operation, and winding aging can cause branch parameter mismatch, which can lead to severe uneven current distribution, causing problems such as local overheating, core saturation, and device damage, severely restricting engineering applications.

[0003] Existing active current sharing technology is costly and complex to control; passive parameter matching technology has high accuracy requirements, poor scalability, and cannot compensate for parameter drift; although current balancing transformer technology has good robustness, it has defects such as offline fixed parameters being unable to adapt to dynamic changes, single current sharing target ignoring loss volume, low integration of discrete structure, and no fault tolerance capability.

[0004] Chinese patent publication number CN122456544A, entitled "A Modular Parallel Voltage Regulating and Stabilizing Device for Distribution Networks and Its Current Sharing Control Method," describes a patent application where a module controller collects its own output current and receives current from other modules for calibration. A monitoring unit calculates the current sharing error and corrects it using topology compensation terms, independently generating voltage regulation commands. The module controller decouples its output based on the corrected current sharing error and the voltage regulation commands. In case of a fault, the faulty module is disconnected and put into bypass mode, and the current sharing error is recalculated based on the remaining modules. However, this patent application fails to address the problem that offline fixed parameters cannot adapt to system drift. Summary of the Invention

[0005] To overcome the problems existing in the prior art, the present invention aims to provide a dynamic current sharing control method and system for modular transformers. Based on residual factor sensitivity analysis, the main conductor inductance parameters are dynamically identified, and the equivalent inductance is continuously adjusted by a magnetically integrated adjustable current balancing transformer to adaptively compensate for the drift of parallel branch parameters of high-frequency transformers, while taking into account the dynamic current sharing control that optimizes multiple objectives such as current sharing error and loss volume.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a dynamic current sharing control method for a modular transformer, the modular transformer comprising N parallel primary windings and secondary windings, where N is an integer greater than or equal to 2, characterized by comprising the following steps: S1: Extract the initial inductance matrix L0 of the transformer system, construct a multi-objective optimization function with current sharing error, total loss, total volume of magnetic components, and hot spot temperature rise based on the cofactor matrix, solve the Pareto optimal solution set, and determine the fixed connection topology, initial installation position and initial equivalent inductance parameters of the adjustable current balance transformer. S2: Collect the instantaneous current values ​​of each parallel winding using distributed current sensors, and calculate the real-time current distribution error ε. real ; S3: When the real-time current distribution error ε real When the preset dynamic threshold is exceeded, based on the recursive least squares method combined with residual factor sensitivity analysis, only the main conductive inductance parameters with a weight ratio of ≥10% that affect current sharing are identified, and the real-time inductance matrix L of the system is obtained. real ; S4: Calculate the real-time inductance matrix L of the system real The corresponding real-time cofactor matrix C real and the real-time current weighting value W of each winding i According to the real-time current weight value W i With the target average current weight W avg The difference is used to calculate the equivalent inductance adjustment ΔL required for each adjustable current balance transformer. j ; S5: Outputs corresponding inductance adjustment control signals to each adjustable current balancing transformer, enabling each winding to adjust its current weight value W in real time. i This continues until the real-time current distribution error converges to the preset steady-state threshold.

[0007] Optionally, in step S1, the multi-objective optimization function is:

[0008] in, , , , These are the weighting coefficients. For the flow equalization error, The total loss of the main transformer and the adjustable current balancing transformer. This represents the total volume of the magnetic components. To address the temperature rise of the winding hotspots, the NSGA-III algorithm was used to find the Pareto optimal solution, and the final optimal solution was determined by the entropy weight method.

[0009] Optionally, in step S3, the residual factor sensitivity analysis method is as follows: calculate the current weight value W of each element in the initial residual factor matrix C0. i Sensitivity coefficient Filter out For main conductive inductance parameters ≥0.1, only this part of the parameters is identified online, shortening the identification cycle to less than 1 / 5 of the original full parameter identification.

[0010] Optionally, the inductance adjustment amount ΔL in step S4 j The calculation formula is:

[0011] in, and Let be the real-time current weight values ​​of the two windings connected to the j-th adjustable current balancing transformer. and These are the parameters for the PI controller, and they are dynamically and adaptively adjusted according to the system load rate.

[0012] Optionally, when the modular transformer is a three-phase system, it further includes a phase-to-phase current sharing control step: constructing an extended inductance matrix of the three-phase system, calculating the current weight value of each winding of each phase, and balancing the total current of the three phases through an adjustable phase-to-phase current balancing transformer.

[0013] Optionally, the preset dynamic threshold is 3%, and the preset steady-state threshold is 1.5%.

[0014] Secondly, the present invention provides a dynamic current sharing control system for a modular transformer, used to implement the dynamic current sharing control method for the modular transformer.

[0015] Optionally, the dynamic current sharing control system for the modular transformer includes: The magnetic integrated transformer body has a magnetic core, which includes a central column and side columns disposed on both sides of the central column. A main transformer winding is wound on the central column, and adjustable current balance transformer windings are wound on the two side columns respectively. The main transformer winding and the adjustable current balance transformer winding share the main magnetic circuit of the magnetic core. N adjustable current balancing transformers, each of which is connected in series to a parallel winding branch, and its equivalent inductance can be continuously adjusted by a control signal. The distributed current acquisition module corresponds one-to-one with each parallel winding and is used to acquire the instantaneous current value of each winding. The online parameter identification module is electrically connected to the distributed current acquisition module and is used to identify the real-time inductance matrix L of the system based on real-time current data. real ; The residual factor dynamic correction module is electrically connected to the online parameter identification module. It is used to calculate the real-time residual factor matrix and the real-time current weight value of each winding, and to generate the adjustable current balance transformer inductance adjustment amount. The adjustable current balancing transformer drive module is electrically connected to the residual factor dynamic correction module and each adjustable current balancing transformer, and is used to output inductance adjustment control signals.

[0016] Optionally, the adjustable current balance transformer is a magnetic valve type controllable inductor structure.

[0017] Optionally, the dynamic current sharing control system of the modular transformer also includes a fault protection module, which is electrically connected to the residual factor dynamic correction module and the adjustable current balance transformer drive module. The fault protection module is used to transfer the current of the overcurrent winding to other normal windings by adjusting the inductance of the corresponding adjustable current balance transformer when the current of a certain winding exceeds 1.2 times the rated current, thereby achieving fault-tolerant operation.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention has strong dynamic current sharing capability. Through a two-stage architecture that combines offline pre-optimization and online dynamic correction, it can adapt to system parameter drift and load changes in real time. The current sharing error is controlled within 2% in the 20%-100% load range, which is better than the existing offline fixed parameter scheme.

[0019] This invention, based on a multi-objective optimization function, simultaneously considers current sharing error, loss, volume, and temperature rise, thus solving the problem of poor engineering practicality caused by single-objective optimization. The total volume of the magnetic components in the system is reduced by 30%, and the total loss is reduced by 15%.

[0020] This invention employs a magnetically integrated structure to share the same magnetic core with the main transformer and the adjustable current balance transformer, eliminating the volume and cost of independent magnetic cores, while also simplifying wiring and reducing parasitic parameters.

[0021] This invention is based on selective parameter identification using residual factor sensitivity analysis. It identifies only the dominant parameters that have the greatest impact on current sharing, improving the identification speed by more than 5 times. Combined with the response time of the solenoid valve structure of ≤10μs, it meets the real-time control requirements of high-frequency systems.

[0022] This invention has fault-tolerant operation capability. When a branch experiences overcurrent or a fault, the current can be automatically transferred to other normal branches to avoid system downtime. Compared with the traditional chain topology, the star modular topology reduces the total number of current balancing transformers by 20%, further improving system reliability.

[0023] This invention is applicable to any N-way parallel primary / secondary winding system, and supports various modular connection methods such as Input Parallel Output Parallel (IPOP) and Input Series Output Parallel (ISOP). It can be widely used in data centers, super-fast charging of electric vehicles, distributed energy storage and solid-state transformers. Attached Figure Description

[0024] The accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of the invention in any way. Furthermore, the shapes and proportions of the components in the drawings are merely schematic to aid in understanding the invention and do not specifically limit the shapes and proportions of the components. In the drawings: Figure 1 This is a flowchart of the modular transformer dynamic current sharing control method according to an embodiment of the present invention; Figure 2 This is a block diagram of the overall architecture of the dynamic flow sharing control system according to an embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0026] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0027] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper", "lower", "horizontal", "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the invention is usually placed during use, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0028] When an element is referred to as being "set on" another element, it can be directly on the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only embodiments. The use of the term "horizontal" does not imply that the component is required to be absolutely horizontal, but rather that it may be slightly tilted. "Horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but rather that it may be slightly tilted.

[0029] It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0030] 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 invention pertains. The terminology used herein in the specification of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification and appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0031] The present invention will now be described in detail with reference to the accompanying drawings.

[0032] This invention provides a dynamic current sharing control method for modular transformers, comprising the following steps: Step 1: Offline Multi-Objective Pre-Optimization Stage: Extract the initial inductance matrix L0 of the transformer system, construct a multi-objective optimization function based on the cofactor matrix, including current sharing error, total loss, total volume of magnetic components, and hot spot temperature rise, and use the NSGA-III (Non-dominated Sorting Genetic Algorithm III) algorithm to solve for the Pareto optimal solution set. Determine the final optimal solution through the entropy weight method, and then determine the fixed connection topology, initial installation location, and initial equivalent inductance parameters of the Adjustable Current Balancing Transformer (ACBX).

[0033] Step Two: Online Real-Time Monitoring Stage: Using distributed current sensors corresponding one-to-one with each parallel winding, the instantaneous current value of each winding is collected, and the real-time current distribution error ε is calculated. real .

[0034] Step 3: Dynamic Parameter Identification Stage: When ε real When the preset dynamic threshold is exceeded, based on the recursive least squares method combined with residual factor sensitivity analysis, only the main conductive inductance parameters with a weight ratio of ≥10% that affect current sharing are identified, and the real-time inductance matrix L of the system is obtained. real .

[0035] Step 4: Residual Factor Dynamic Correction Stage: Calculate L real The corresponding real-time cofactor matrix C real and the real-time current weighting value W of each winding i According to the real-time current weight value W i Weighted by the target average current value W avg The difference is used to calculate the equivalent inductance adjustment ΔL required for each adjustable current balancing transformer using a PI controller. j .

[0036] Step 5: Closed-loop execution phase: Output the corresponding inductance adjustment control signal to each adjustable current balancing transformer to make the real-time current weight value of each winding approach W. avg This continues until the real-time current distribution error converges to the preset steady-state threshold.

[0037] The multi-objective optimization function mentioned in step one is:

[0038] In the formula: , , , The weight coefficients for each optimization objective satisfy... , The system current sharing error is expressed as a percentage. The total loss (W) of the main transformer and the adjustable current balancing transformer. The total volume of the magnetic components in the system (cm²) 3 ), Temperature rise of the winding hot spot (°C).

[0039] The residual factor sensitivity analysis method described in step S3 is as follows: calculate the sensitivity coefficient of each element in the initial residual factor matrix C0 to the current weight value W_i. Filter out For main conductivity parameters ≥0.1, only this part of the parameters is identified online. Sensitivity coefficient. The calculation formula is:

[0040] In the formula: No. The current weighting value of the circuit winding affects the inductance matrix. Line number Sensitivity coefficients of column elements; For the first The current weighting value of the parallel windings in the circuit; The first inductor in the system inductor matrix L Line number The elements of the column; when hour, For the first The self-inductance of the winding (including equivalent leakage inductance and terminal bus inductance); when hour, For the first The winding and the first Mutual inductance between windings.

[0041] The formula for calculating the inductance adjustment ΔL_j in step S4 is:

[0042] In the formula: The equivalent inductance adjustment (H) required for the j-th adjustable current balance transformer; The proportional gain of the PI controller; The integral coefficient of the PI controller; This represents the real-time current weight value of the m-th winding connected to the j-th adjustable current balancing transformer. The real-time current weight value of the nth winding connected to the j-th adjustable current balancing transformer.

[0043] When the modular transformer is a three-phase system, it also includes a phase-to-phase current sharing control step: constructing an extended inductance matrix of the three-phase system, calculating the current weight value of each winding of each phase, and balancing the total current of the three phases through an adjustable current balancing transformer between phases, so that the three-phase current imbalance is ≤3%.

[0044] The connection topology of the adjustable current balancing transformer in step S1 adopts a star modular topology, and the common terminal of all adjustable current balancing transformers is connected to the same reference node.

[0045] The preset dynamic threshold is 3%, and the preset steady-state threshold is 1.5%; when the system load rate varies within the range of 20%-100%, the current sharing error is always controlled within 2%.

[0046] This invention provides a dynamic current sharing control system for a modular transformer, used to implement the dynamic current sharing control method for the modular transformer, comprising: The magnetic integrated transformer body adopts an EE-type magnetic core structure. The main transformer winding and the adjustable current balance transformer winding are integrated on the same magnetic core. The main transformer winding is wound on the central column of the magnetic core, and the adjustable current balance transformer winding is wound on the two side columns of the magnetic core, sharing the main magnetic circuit flux.

[0047] The magnetic core of the magnetic integrated transformer body adopts a segmented air gap structure. The air gap length of the side column of the adjustable current balance transformer is greater than the air gap length of the middle column of the main transformer, so that the linearity of inductance adjustment of the adjustable current balance transformer is ≥95%.

[0048] N adjustable current balancing transformers: Each adjustable current balancing transformer is connected in series to a parallel winding branch, and its equivalent inductance can be continuously adjusted by a control signal.

[0049] Distributed current acquisition module: The distributed current acquisition module corresponds one-to-one with each parallel winding and is used to acquire the instantaneous current value of each winding.

[0050] Online parameter identification module: The online parameter identification module is electrically connected to the distributed current acquisition module and is used to identify the real-time inductance matrix L_real of the system based on real-time current data.

[0051] Residual Factor Dynamic Correction Module: The residual factor dynamic correction module is electrically connected to the online parameter identification module. It is used to calculate the real-time residual factor matrix and the real-time current weight value of each winding, and to generate the adjustable current balance transformer inductance adjustment amount.

[0052] Adjustable current balancing transformer drive module: electrically connected to the residual factor dynamic correction module and each adjustable current balancing transformer, used to output inductance adjustment control signal.

[0053] The adjustable current balance transformer is a magnetic valve type controllable inductor structure, including a main magnetic core, a working winding, and a DC bias winding. By adjusting the current of the DC bias winding, the saturation of the main magnetic core is changed, thereby realizing continuous adjustment of the equivalent inductance. The adjustment range is 0.5L0-2L0, and the response time is ≤10μs.

[0054] The PCB (Printed Circuit Board) winding board of the magnetic integrated transformer body adopts a four-layer structure, of which two layers are the main transformer windings and the other two layers are adjustable current balancing transformer windings; the DC bias winding wiring and heat dissipation copper foil of the adjustable current balancing transformer are integrated on the PCB winding board, and the heat dissipation copper foil is in direct contact with the water cooling heat dissipation channel of the main transformer.

[0055] Fault Protection Module: The fault protection module is electrically connected to the residual factor dynamic correction module and the adjustable current balance transformer drive module. When the current of a certain winding exceeds 1.2 times the rated current, the fault protection module transfers the current of the overcurrent winding to other normal windings by adjusting the inductance of the corresponding adjustable current balance transformer, thereby achieving fault-tolerant operation.

[0056] Example 1 This embodiment provides a modular transformer dynamic current sharing control method, such as... Figure 1 As shown, the specific implementation steps are as follows: Step 1: Offline multi-objective pre-optimization stage: A three-dimensional electromagnetic field simulation of the modular transformer was performed using ANSYS Maxwell electromagnetic simulation software, or a hardware test was conducted on the transformer prototype using an impedance analyzer. The initial inductance matrix L0 (8×8 matrix) of the modular transformer with 4 parallel primary windings and 4 parallel secondary windings was extracted. The diagonal elements of the matrix contain the equivalent leakage inductance of each winding and the terminal bus inductance, while the off-diagonal elements contain the mutual inductance between any two windings.

[0057] Construct a multi-objective optimization function and determine the weight coefficients of each optimization objective using the entropy weight method: w 1 = 0.4 (flow sharing error) w 2 = 0.3 (total loss) w 3=0.2 (magnetic component volume), w4=0.1 (hot spot temperature rise), prioritizing the current sharing performance of the system, and secondarily considering efficiency and power density.

[0058] The NSGA-III algorithm was used to find the Pareto optimal solution set, with 100 iterations and a population size of 50. Among the obtained Pareto fronts, the solution with the minimum current sharing error ≤1.5% and the minimum total loss was selected as the final optimal solution. A star-shaped modular topology was adopted, connecting four adjustable current balancing transformers, with an initial equivalent inductance of 20μH for each transformer.

[0059] Step 2: Online Real-time Monitoring Phase A Rogowski coil current sensor is connected in series in each branch of the primary and secondary windings. The current sampling frequency is set to 1MHz to collect the instantaneous current value of each winding in real time. Based on the collected current data, the real-time current distribution error of the system is calculated.

[0060] In the formula: —Average current of all parallel windings, in amperes (A). —Real-time current distribution error, in percentage (%) ——No. i The instantaneous current value of the parallel windings, in amperes (A). Step 3: Dynamic Parameter Identification Stage When real-time current distribution error When the value is >3%, the dynamic parameter identification process is initiated: Calculate the sensitivity coefficient of each element in the initial cofactor matrix C0 to the winding current weight value Wi:

[0061] In the formula: —The sensitivity coefficient of the current weight value of the i-th winding to the element in the i-th row and j-th column of the inductance matrix.

[0062] —The current weight value of the i-th parallel winding.

[0063] —The element in the i-th row and j-th column of the system inductance matrix L Filter out The main conductive inductance parameter is ≥0.1. In this embodiment, eight main parameters are selected, mainly the equivalent leakage inductance of the primary winding and the terminal bus inductance.

[0064] The eight selected main inductance parameters were identified online using the recursive least squares method, with an identification period of 20 μs, to obtain the real-time inductance matrix L of the system. real .

[0065] Step 4: Residual Factor Dynamic Correction Stage According to the real-time inductance matrix L real Calculate the corresponding real-time cofactor matrix C real Furthermore, the real-time current weight value W of each winding is calculated. i .

[0066] Calculate the target average current weight value: , The target average current weight value, For the first i The real-time current weighting value of the parallel windings, where N is the total number of parallel windings.

[0067] For each adjustable current balancing transformer connected to two windings, calculate the difference in their current weight values, and then use an adaptive PI controller to calculate the required equivalent inductance adjustment.

[0068] In the formula: ——No. j The equivalent inductance adjustment required for an adjustable current balance transformer, in Henrys; —The proportional gain of the PI controller; —The integral coefficient of the PI controller; , ——No. j The real-time current weighting value of the two windings connected by an adjustable current balancing transformer.

[0069] The PI controller parameters are dynamically and adaptively adjusted according to the system load rate: when the load rate is <50%, =0.5、 =100; when the load rate is ≥50%, =0.8、 =200.

[0070] Step 5: Closed-loop execution phase: The adjustable current balancing transformer drive module adjusts the inductance based on the calculated amount. A DC bias current is output to the corresponding adjustable current balancing transformer to adjust its equivalent inductance. The closed-loop adjustment period is set to 50μs, and steps 2 to 4 are repeated until the real-time current distribution error is correct. ≤1.5%.

[0071] Example 2 like Figure 2 As shown, the modular transformer dynamic current sharing control system of this embodiment includes: The magnetically integrated transformer body adopts an EE-type ferrite core structure, with the main transformer winding and the adjustable current balancing transformer winding integrated on the same core. Specifically, the primary and secondary windings of the main transformer are alternately wound on the central column of the core, while the adjustable current balancing transformer windings are wound on the two side columns of the core, sharing the main magnetic flux. In this embodiment, the magnetically integrated transformer body includes four parallel primary windings and four parallel secondary windings, integrating four adjustable current balancing transformers.

[0072] The distributed current acquisition module consists of eight Rogowski coil current sensors, each connected in series with one of the four primary windings and one of the four secondary windings, to acquire the instantaneous current values ​​of each winding. The current sensors have a bandwidth of 5MHz and an accuracy of ±0.5%, meeting the requirements for high-frequency current measurement.

[0073] Online parameter identification module: Implemented on a Xilinx Artix-7 FPGA (Field-Programmable Gate Array), it is electrically connected to the distributed current acquisition module via a high-speed AD converter. This module receives current sampling data from each winding, executes a recursive least squares algorithm to identify the main conductor inductance parameters online, and outputs the system's real-time inductance matrix L. real .

[0074] The residual factor dynamic correction module is implemented based on the TI TMS320F28379D DSP (Digital Signal Processor) and is electrically connected to the online parameter identification module via a parallel bus. This module receives the real-time inductance matrix, calculates the real-time residual factor matrix and the current weight values ​​of each winding, executes an adaptive PI control algorithm, and generates the inductance adjustment amount for the adjustable current balancing transformer.

[0075] Adjustable current balancing transformer drive module: Composed of 4 independent high-precision DC current sources. The input terminal is electrically connected to the residual factor dynamic correction module, and the output terminal is electrically connected to the DC bias windings of the 4 adjustable current balancing transformers respectively. The output range of the DC current source is 0-1A, the accuracy is ±1mA, and the response time is ≤5μs, which can quickly and accurately output the bias current required for inductor adjustment.

[0076] Fault protection module: Integrated within the DSP, it is electrically connected to the residual factor dynamic correction module and the adjustable current balancing transformer drive module. This module monitors the current value of each winding in real time. When it detects that the current of a certain winding exceeds 1.2 times the rated current, it immediately triggers the fault-tolerant control logic. By adjusting the inductance of the corresponding adjustable current balancing transformer, the current of the overcurrent winding is transferred to other normal windings, avoiding device damage and system shutdown.

[0077] Unless otherwise specified, the equipment components involved in the above embodiments are all conventional equipment components, and the structural settings, working methods or control methods involved are all conventional settings, working methods or control methods in the art unless otherwise specified.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solutions of the present invention, as long as they do not depart from the spirit and scope of the technical solutions of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A dynamic current sharing control method for a modular transformer, wherein the modular transformer comprises N parallel primary windings and secondary windings, where N is an integer greater than or equal to 2, characterized in that, Includes the following steps: S1: Extract the initial inductance matrix L0 of the transformer system, construct a multi-objective optimization function with current sharing error, total loss, total volume of magnetic components, and hot spot temperature rise based on the cofactor matrix, solve the Pareto optimal solution set, and determine the fixed connection topology, initial installation position and initial equivalent inductance parameters of the adjustable current balance transformer. S2: Collect the instantaneous current values ​​of each parallel winding using distributed current sensors, and calculate the real-time current distribution error ε. real ; S3: When the real-time current distribution error ε real When the preset dynamic threshold is exceeded, based on the recursive least squares method combined with residual factor sensitivity analysis, only the main conductive inductance parameters with a weight ratio of ≥10% that affect current sharing are identified, and the real-time inductance matrix L of the system is obtained. real ; S4: Calculate the real-time inductance matrix L of the system real The corresponding real-time cofactor matrix C real and the real-time current weighting value W of each winding i According to the real-time current weight value W i Weighted by the target average current W avg The difference is used to calculate the equivalent inductance adjustment ΔL required for each adjustable current balance transformer. j ; S5: Outputs corresponding inductance adjustment control signals to each adjustable current balancing transformer, enabling each winding to adjust its current weight value W in real time. i This continues until the real-time current distribution error converges to the preset steady-state threshold.

2. The dynamic current sharing control method for modular transformers according to claim 1, characterized in that, In step S1, the multi-objective optimization function is: in, , , , These are the weighting coefficients. For the flow equalization error, The total loss of the main transformer and the adjustable current balancing transformer. This represents the total volume of the magnetic components. To address the temperature rise of the winding hotspots, the NSGA-III algorithm was used to find the Pareto optimal solution, and the final optimal solution was determined by the entropy weight method.

3. The dynamic current sharing control method for modular transformers according to claim 1, characterized in that, In step S3, the residual factor sensitivity analysis method is as follows: calculate the current weight value W of each element in the initial residual factor matrix C0. i Sensitivity coefficient Filter out For main conductive inductance parameters ≥0.1, only this part of the parameters is identified online, shortening the identification cycle to less than 1 / 5 of the original full parameter identification.

4. The dynamic current sharing control method for modular transformers according to claim 1, characterized in that, The inductance adjustment amount ΔL mentioned in step S4 j The calculation formula is: in, and Let be the real-time current weight values ​​of the two windings connected to the j-th adjustable current balancing transformer. and These are the parameters for the PI controller, and they are dynamically and adaptively adjusted according to the system load rate.

5. The dynamic current sharing control method for modular transformers according to claim 1, characterized in that, When the modular transformer is a three-phase system, it also includes a phase-to-phase current sharing control step: constructing an extended inductance matrix of the three-phase system, calculating the current weight value of each winding of each phase, and balancing the total current of the three phases through an adjustable phase-to-phase current balancing transformer.

6. The dynamic current sharing control method for modular transformers according to claim 1, characterized in that, The preset dynamic threshold is 3%, and the preset steady-state threshold is 1.5%.

7. A dynamic current sharing control system for a modular transformer, characterized in that, A dynamic current sharing control method for implementing the modular transformer according to any one of claims 1 to 7.

8. The dynamic current sharing control system for a modular transformer according to claim 7, characterized in that, The dynamic current sharing control system for modular transformers includes: The magnetic integrated transformer body has a magnetic core, which includes a central column and side columns disposed on both sides of the central column. A main transformer winding is wound on the central column, and adjustable current balance transformer windings are wound on the two side columns respectively. The main transformer winding and the adjustable current balance transformer winding share the main magnetic circuit of the magnetic core. N adjustable current balancing transformers, each of which is connected in series to a parallel winding branch, and its equivalent inductance can be continuously adjusted by a control signal. The distributed current acquisition module corresponds one-to-one with each parallel winding and is used to acquire the instantaneous current value of each winding. The online parameter identification module is electrically connected to the distributed current acquisition module and is used to identify the real-time inductance matrix L of the system based on real-time current data. real ; The residual factor dynamic correction module is electrically connected to the online parameter identification module. It is used to calculate the real-time residual factor matrix and the real-time current weight value of each winding, and to generate the adjustable current balance transformer inductance adjustment amount. The adjustable current balancing transformer drive module is electrically connected to the residual factor dynamic correction module and each adjustable current balancing transformer, and is used to output inductance adjustment control signals.

9. The dynamic current sharing control system for a modular transformer according to claim 8, characterized in that, The adjustable current balance transformer is a magnetic valve type controllable inductor structure.

10. The dynamic current sharing control system for a modular transformer according to claim 8, characterized in that, The dynamic current sharing control system of the modular transformer also includes a fault protection module, which is electrically connected to the residual factor dynamic correction module and the adjustable current balance transformer drive module. The fault protection module is used to transfer the current of the overcurrent winding to other normal windings by adjusting the inductance of the corresponding adjustable current balance transformer when the current of a certain winding exceeds 1.2 times the rated current, thereby achieving fault-tolerant operation.

Images