Transformer neutral point direct current bias suppression heuristic control compensation device and method

Abstract

The invention discloses a transformer neutral point DC bias suppression heuristic control compensation device and method, and relates to the field of power transformer insulation operation reliability. The device comprises a direct current bias detection unit used for collecting a direct current bias voltage of a neutral point of a detected transformer relative to the ground; the polarity discrimination unit is used for receiving the direct current bias voltage and determining the direct current bias polarity of the direct current bias voltage; the control unit is used for performing off-line optimization on the proportional gain Gp and the integral gain Gi by utilizing an artificial bee colony algorithm based on the direct current bias polarity; and the power conversion module is used for outputting direct current based on the Gp and the Gi so as to enable the control unit to determine optimal Gp and Gi, and outputting compensation direct current for direct current bias of the neutral point of the transformer based on the optimal Gp and Gi. According to the invention, the method achieves the quick and stable compensation capability under positive and negative polarity biases, can maintain the consistent dynamic performance in different amplitude intervals, and achieves the control-type neutral point DC compensation with high robustness.

Description

Technical Field

[0001] This invention relates to the field of power transformer insulation operation reliability, specifically to a transformer neutral point DC bias suppression heuristic control compensation device and method. Background Technology

[0002] With the rapid construction of ultra-high voltage and extra-high voltage transmission networks, the use of large-capacity power transformers is becoming increasingly widespread, requiring electrical equipment protection systems to possess higher operational reliability and faster response speeds. As a core piece of equipment in modern industrial systems, the transformer's core is composed of ferromagnetic materials, and its electromagnetic characteristics directly determine its electrical performance. During routine transformer switching operations (such as voltage ratio measurement, DC resistance testing, and no-load testing), residual magnetism is easily generated in the core. If this residual magnetism is not effectively eliminated, it will lead to DC bias in the transformer, causing problems such as increased noise, harmonic distortion, intensified vibration, and increased reactive power demand. In severe cases, it may even cause the transformer to overheat, affecting its operational safety and shortening its lifespan.

[0003] Existing technologies mainly limit DC bias through neutral point series blocking devices, series capacitors, and bypass structures. However, these methods suffer from drawbacks such as sensitivity to system overvoltage, the risk of resonance, inconsistent compensation capabilities in different polarity directions, overcompensation or undercompensation in traditional PI (proportional-integral) control across different bias amplitude ranges, and poor dynamic performance. Therefore, there is an urgent need for a control-type neutral point DC compensation method that possesses fast and stable compensation capabilities under both positive and negative polarity biases, maintains consistent dynamic performance across different amplitude ranges, and exhibits strong robustness. Summary of the Invention

[0004] This invention addresses the DC bias problem that occurs during the daily switching operations of transformers by providing a heuristic control and compensation device and method for suppressing DC bias at the transformer neutral point. It achieves rapid and stable compensation capability under both positive and negative polarity biases, maintains consistent dynamic performance across different amplitude ranges, and exhibits robust control-type neutral point DC compensation.

[0005] This invention is achieved through the following technical solution:

[0006] In a first aspect, a transformer neutral point DC bias suppression heuristic control compensation device is provided, the device comprising:

[0007] The DC bias detection unit is used to acquire the DC bias voltage of the neutral point of the transformer under test relative to ground.

[0008] A polarity determination unit, connected to the DC bias detection unit, is used to receive the DC bias voltage and determine the DC bias polarity of the DC bias voltage;

[0009] The control unit, connected to the polarity discrimination unit, is used to adjust the proportional gain G based on the DC bias polarity using an artificial bee colony algorithm. p Integral gain G i Perform offline optimization;

[0010] The power conversion module, connected to the control unit, is used to convert G... p and G i Output DC current so that the control unit determines the optimal G p and G i And based on the optimal G p and G i Output a compensated DC current for the DC bias of the neutral point of the transformer under test.

[0011] In some embodiments, the polarity discrimination unit includes a voltage comparator, wherein the non-inverting input of the voltage comparator is connected to the output of the DC bias detection unit to receive the DC bias voltage, the inverting input of the voltage comparator is grounded, and the output of the voltage comparator outputs a DC bias polarity discrimination signal.

[0012] In some embodiments, the control unit includes a microprocessor and a pulse width modulation (PWM) generator, wherein the microprocessor is connected to the output of the voltage comparator and is used to determine the G based on the DC bias polarity discrimination signal. p and G i The adjustment direction is to utilize the artificial bee colony algorithm to adjust the G p and G i Adjustments are made so that the PWM generator is based on the adjusted G. p and G i Output PWM signal,

[0013] The artificial bee colony algorithm is used to analyze the G. p and G i Adjustments will be made, including:

[0014] Based on the preset optimization parameter range, several candidate solutions are randomly generated. Each candidate solution includes a set of control parameters (G). p G i The system consists of ( ) components, and sets the maximum number of iterations, population size, and performance evaluation metrics;

[0015] The candidate solutions from the plurality of candidate solutions are input into the PWM generator to generate a PWM signal, and the PWM signal is input into the power conversion module. The square integral of the DC current error output by the power conversion module is calculated as the performance evaluation index.

[0016] Each hired bee searches for new candidate solutions near the current candidate solution and calculates the performance evaluation index of the new candidate solution. If the performance of the new candidate solution is better than that of the original candidate solution, the original candidate solution is updated to the new candidate solution.

[0017] The bees are observed to select a region of superior solutions from the candidate solutions based on probability, and then search for new candidate solutions within the region of superior solutions.

[0018] For candidate solution regions that fail to update several times in a row, the scout bee randomly generates new candidate solution regions;

[0019] Record the optimal parameter combination in the current iteration, and repeat the steps: input the candidate solutions from the candidate solutions into the PWM generator to generate a PWM signal, and input the PWM signal into the power conversion module, calculate the square integral of the DC current error output by the power conversion module, until the step: for the candidate solution region that has failed to update for several consecutive times, the scout bee randomly generates a new candidate solution region, until the maximum number of iterations or the performance evaluation index converges.

[0020] In some embodiments, the power conversion module includes a boost DC-DC converter and a buck-boost DC-DC converter, wherein the boost DC-DC converter and the buck-boost DC-DC converter are respectively connected to the PWM generator for outputting DC current based on the PWM signal.

[0021] In some embodiments, the device further includes a data acquisition unit for acquiring the DC current output by the power conversion module, so that the control unit, based on the DC current output by the power conversion module, uses the integral of the squared DC current error as an evaluation index and employs the artificial bee colony algorithm to determine the minimum value of the integral of the squared DC current error corresponding to G. p and G i As the optimal G p and G i .

[0022] Secondly, a heuristic control compensation method for suppressing DC bias at the transformer neutral point is provided.

[0023] The DC bias voltage of the neutral point of the transformer under test relative to ground is acquired by the DC bias detection unit.

[0024] The polarity determination unit receives the DC bias voltage and determines the DC bias polarity of the DC bias voltage.

[0025] Through the control unit, based on the DC bias polarity, the proportional gain G is adjusted using an artificial bee colony algorithm. p Integral gain G iPerform offline optimization;

[0026] Through the power conversion module, based on the G p and G i Output DC current so that the control unit determines the optimal G based on the DC current. p and G i And based on the optimal G p and G i Output a compensated DC current for the DC bias of the neutral point of the transformer under test.

[0027] In some embodiments, the DC bias voltage is received at the non-inverting input of the voltage comparator of the polarity discrimination unit, the inverting input is connected to ground, and the output output is a DC bias polarity discrimination signal.

[0028] In some embodiments, the microprocessor of the control unit receives the DC bias polarity discrimination signal to determine the G. p and G i The direction of adjustment is then used to apply the artificial bee colony algorithm to the G. p and G i Adjustments are made, and the PWM generator of the control unit is based on the adjusted G. p and G i Output PWM signal, where,

[0029] The artificial bee colony algorithm is used to analyze the G. p and G i Adjustments will be made, including:

[0030] Based on the preset optimization parameter range, several candidate solutions are randomly generated. Each candidate solution includes a set of control parameters (G). p G i The system consists of ( ) components, and sets the maximum number of iterations, population size, and performance evaluation metrics;

[0031] The candidate solutions from the plurality of candidate solutions are input into the PWM generator to generate a PWM signal, and the PWM signal is input into the power conversion module. The square integral of the DC current error output by the power conversion module is calculated as the performance evaluation index.

[0032] Each hired bee searches for new candidate solutions near the current candidate solution and calculates the performance evaluation index of the new candidate solution. If the performance of the new candidate solution is better than that of the original candidate solution, the original candidate solution is updated to the new candidate solution.

[0033] The bees are observed to select a region of superior solutions from the candidate solutions based on probability, and then search for new candidate solutions within the region of superior solutions.

[0034] For candidate solution regions that fail to update several times in a row, the scout bee randomly generates new candidate solution regions;

[0035] Record the optimal parameter combination in the current iteration, and repeat the steps: input the candidate solutions from the candidate solutions into the PWM generator to generate a PWM signal, and input the PWM signal into the power conversion module, calculate the square integral of the DC current error output by the power conversion module, until the step: for the candidate solution region that has failed to update for several consecutive times, the scout bee randomly generates a new candidate solution region, until the maximum number of iterations or the performance evaluation index converges.

[0036] In some embodiments, the power conversion module outputs DC current based on the PWM signal through its boost DC-DC converter and buck-boost DC-DC converter.

[0037] In some embodiments, the DC current output by the power conversion module is acquired by the acquisition unit, so that the control unit, based on the DC current output by the power conversion module, uses the integral of the squared DC current error as an evaluation index and employs the artificial bee colony algorithm to determine the minimum value of G corresponding to the integral of the squared DC current error. p and G i As the optimal G p and G i .

[0038] Compared with existing technologies, this invention has the following advantages and beneficial effects: By introducing an independent measurement unit and a polarity discrimination unit to automatically identify the polarity of the DC bias voltage at the neutral point, the polarity result is input into the control unit. The PI control parameters are then optimized offline using an artificial bee colony algorithm to obtain the optimal proportional-integral gain combination. The corresponding power module is switched according to the polarity to achieve automatic compensation for positive and negative DC bias. This invention exhibits good adaptability to different power grid operating conditions and can effectively resist external interference. Furthermore, the circuit structure of this technical solution mainly uses conventional power electronic devices, making it easy to inherit and maintain, and possessing good engineering scalability. It provides a new technical approach for managing DC bias at the neutral point in high-voltage direct current transmission systems and has significant engineering application value for ensuring the safe operation of power systems. Attached Figure Description

[0039] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as limiting the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of a transformer neutral point DC bias suppression heuristic control compensation device according to an embodiment of the present invention.

[0041] Figure 2 This is a circuit connection diagram of a transformer neutral point DC bias suppression heuristic control compensation device according to an embodiment of the present invention.

[0042] Figure 3 This is a flowchart of PI control according to an embodiment of the present invention.

[0043] Figure 4 The flowchart illustrates the algorithm for optimizing PI control parameters using an artificial bee colony according to an embodiment of the present invention.

[0044] Figure 5 This is a flowchart of a transformer neutral point DC bias suppression heuristic control compensation method according to an embodiment of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0046] To address the DC bias problem that occurs during the daily switching operations of transformers, this invention provides a neutral point DC compensation method that can achieve fast, stable, and low-oscillation suppression under different polarities and different bias amplitudes. Figure 1 This is a schematic diagram of a transformer neutral point DC bias suppression heuristic control compensation device according to an embodiment of the present invention. (Reference) Figure 1 The transformer neutral point DC bias suppression heuristic control compensation device includes: a DC bias detection unit, a polarity discrimination unit, a control unit, and a power conversion module.

[0047] The DC bias detection unit is used to acquire the DC bias voltage of the neutral point of the transformer under test relative to ground. Specifically, the measuring terminal of the DC bias detection unit is connected to the neutral point of the transformer under test, and the reference terminal is connected to ground to measure the DC bias voltage of the neutral point relative to ground in real time. The output terminal is connected to the polarity discrimination unit to provide the neutral point-to-ground DC bias voltage to the polarity discrimination unit.

[0048] The polarity determination unit, connected to the DC bias detection unit, is used to receive the DC bias voltage and determine its DC bias polarity. Specifically, the input terminal of the polarity determination unit is connected to the DC bias detection unit to detect the polarity of the DC bias voltage, and the output terminal is connected to the control unit to provide a DC bias polarity determination signal to the control unit.

[0049] The control unit, connected to the polarity determination unit, is used to adjust the proportional gain G based on the DC bias polarity using an artificial bee colony algorithm. p Integral gain G i Offline optimization is performed. Specifically, the control unit's input is connected to the polarity determination unit to receive the DC bias polarity determination signal, and its output is connected to the power conversion module to control the power conversion module's output DC current. The control unit uses an artificial bee colony algorithm to optimize the PI control parameter G of the control unit. p G i Offline optimization was performed to obtain the optimal parameter value (G) that minimizes the neutral point current. p *、G i *).

[0050] The power conversion module, connected to the control unit, is used for G-based... p and G i Output DC current to enable the control unit to determine the optimal G p and G i and based on optimal G p and G i Outputs a compensated DC current for DC bias at the transformer neutral point.

[0051] In some embodiments, the polarity discrimination unit includes a voltage comparator, wherein the non-inverting input of the voltage comparator is connected to the output of the DC bias detection unit to receive the DC bias voltage, the inverting input of the voltage comparator is grounded, and the output of the voltage comparator outputs a DC bias polarity discrimination signal. Specifically, the polarity discrimination unit compares the neutral point voltage with the ground potential and outputs a polarity discrimination signal. When the output is high, it indicates that the neutral point voltage is positive, and the control unit drives the boost DC-DC converter module; when the output is low, it indicates that the neutral point voltage is negative, and the control unit drives the Buck-Boost DC-DC converter module.

[0052] In some embodiments, the control unit includes a microprocessor and a pulse width modulation (PWM) generator, wherein the microprocessor is connected to the output of a voltage comparator for determining the polarity discrimination signal G. p and G i The adjustment direction is to utilize the artificial bee colony algorithm for G p and G i Adjustments are made so that the PWM generator is based on the adjusted G.p and G i Output PWM signal.

[0053] The control unit controls the power conversion module. As the core of the entire device, the control unit adjusts and controls the power conversion module (including boost and buck-boost power converters) through PI control.

[0054] The time-domain equations of the control signals output from the control unit to the PWM generator are defined by equations (1) and (2):

[0055] (1)

[0056] (2)

[0057] Where e(t) is the voltage error signal, V n V0(t) is the DC bias voltage measured from the neutral point of the transformer, and G is the output voltage of the power conversion module at the sampling time. p G i These are the proportional gain and integral gain of the control unit, respectively.

[0058] After the time-domain equation is transformed by Laplace, the transfer function in the frequency domain is obtained as (3):

[0059] (3)

[0060] Using bilinear transformation to convert continuous-time control into discrete-time control:

[0061] (4)

[0062] Where W represents the sampling period. Therefore, the discrete form of the control signal output from the PI controller to the PWM generator is: (5)

[0063] Introducing new parameters G0 and G1:

[0064] (6)

[0065] (7)

[0066] Substituting C(z) and performing an inverse transform, we obtain:

[0067] (8)

[0068] Dividing (9) by W yields the control law equation of the discrete PI controller as shown in equation (12):

[0069] (9)

[0070] make:

[0071] (10)

[0072] The PWM signal output by the controller can be simply represented as:

[0073] (11)

[0074] The PWM signal controls the duty cycle of the power conversion and adjusts the compensation current I. n (t). Figure 3 The flow of PI control is shown.

[0075] Specifically, the microprocessor can use the artificial bee colony algorithm to process G. p and G i The adjustments are as follows, including steps S1 to S6.

[0076] In S1, several candidate solutions are randomly generated based on a preset range of optimization parameters. Each candidate solution includes a set of control parameters (G). p G i It consists of ) and sets the maximum number of iterations, population size and performance evaluation indicators.

[0077] In S2, one of the candidate solutions from several options is input to a PWM generator to produce a PWM signal. This PWM signal is then input to the power conversion module, where the integral of the squared error of the DC current output by the power conversion module is calculated as a performance evaluation metric. This invention uses the integral of the squared error of the neutral point DC current (ISE) as the metric:

[0078] (12)

[0079] Among them, I n (t) represents the neutral point DC current, and the goal is to minimize J.

[0080] In S3, each hired bee searches for new candidate solutions near the current candidate solution and calculates the performance evaluation index of the new candidate solution. If the performance of the new candidate solution is better than that of the original candidate solution, the original candidate solution is updated to the new candidate solution (G). p G i ).

[0081] In S4, the bee observes the region of good solutions among the candidate solutions according to probability, and searches for new candidate solutions within the region of good solutions.

[0082] In S5, for candidate solution regions that fail to update several times in a row, the scout bee randomly generates new candidate solution regions.

[0083] In S6, record the optimal parameter combination in the current iteration, and repeat steps S2 to S5 until the maximum number of iterations is reached or the performance evaluation index converges.

[0084] Finally, the optimal control parameters (G) are output. p * G i * This refers to the PI control gain obtained through offline optimization, which is used consistently during device operation and does not require real-time recalculation. Figure 4 The specific process of the artificial bee colony optimization algorithm is shown.

[0085] In some embodiments, the power conversion module includes a boost DC-DC converter and a buck-boost DC-DC converter, wherein the boost DC-DC converter and the buck-boost DC-DC converter are respectively connected to a PWM generator for outputting DC current based on a PWM signal.

[0086] The power module is a controllable DC voltage source used to generate the DC power required to compensate for the neutral DC. It consists of two DC-DC converters: a boost DC-DC converter and a Buck-Boost DC-DC converter. In this invention, when a negative DC bias voltage is detected, a boost DC-DC converter is used to generate a positive DC voltage; when a positive DC bias voltage is detected, a Buck-Boost DC-DC converter is used to generate a negative DC voltage. The minimum inductance and capacitance required by the DC-DC power converter are calculated according to the commonly used formulas under continuous current mode conditions, as shown in equations (13)-(15):

[0087] (13)

[0088] (14)

[0089] (15)

[0090] in, Minimum inductance; The minimum capacitance is V; the input voltage is V. in The output voltage is V o The switching frequency is f s Inductor current ripple ΔI L Take 5% of the rated current, D as the duty cycle, and R as the equivalent load resistance. The allowable peak-to-peak output voltage ripple.

[0091] In some embodiments, the device further includes a data acquisition unit for acquiring the DC current output by the power conversion module, so that the control unit, based on the DC current output by the power conversion module, uses the integral of the squared DC current error as an evaluation index and employs an artificial bee colony algorithm to determine the minimum value of the G corresponding to the integral of the squared DC current error. p and G i As the optimal G p and G i .

[0092] This invention connects an independent DC bias detection unit, polarity discrimination unit, control unit, and power conversion module to the transformer neutral point, and utilizes an artificial bee colony algorithm to optimize the control parameters of the control unit offline. This enables the transformer to achieve rapid and effective compensation during field operation, stably suppressing the neutral point DC current to below a preset threshold (typical engineering target <1 A), thereby significantly reducing the risk of half-cycle saturation in the transformer core and improving the reliability of transformer operation and the safety of power system operation. Furthermore, the circuit structure primarily uses conventional power electronic devices, making it easy to inherit and maintain, and possessing good engineering scalability. It provides a new technical approach for neutral point DC bias management in high-voltage direct current transmission systems and has significant engineering application value for ensuring the safe operation of power systems.

[0093] On the other hand, the present invention provides a heuristic control compensation method for DC bias suppression of transformer neutral point. Figure 5 This is a flowchart of a heuristic control compensation method for DC bias suppression of the transformer neutral point according to an embodiment of the present invention. (Reference) Figure 5 The transformer neutral point DC bias suppression heuristic control compensation method includes S10 to S40.

[0094] In S10, the DC bias voltage of the neutral point of the transformer under test relative to ground is acquired through the DC bias detection unit.

[0095] In S20, the DC bias voltage is received and the DC bias polarity of the DC bias voltage is determined by the polarity discrimination unit.

[0096] In S30, the proportional gain G is adjusted by the control unit based on the DC bias polarity using an artificial bee colony algorithm. p Integral gain G i Perform offline optimization.

[0097] In S40, through the power conversion module, based on G... p and G i Output DC current so that the control unit can determine the optimal G based on the DC current. p and G i and based on optimal G p and G iOutputs a compensated DC current for DC bias at the transformer neutral point.

[0098] In some embodiments, a DC bias voltage is received at the non-inverting input of the polarity discrimination unit voltage comparator, the inverting input is grounded, and the output output is a DC bias polarity judgment signal.

[0099] In some embodiments, the polarity discrimination signal is received by the microprocessor of the control unit to determine G. p and G i The direction of adjustment is then used to apply the artificial bee colony algorithm to G. p and G i Adjustments are made, and the PWM generator of the control unit is based on the adjusted G. p and G i Output PWM signal, wherein the artificial bee colony algorithm is used to control G. p and G i Adjustments are made, including: randomly generating several candidate solutions based on a preset range of optimization parameters, each candidate solution including a set of control parameters (G). p G i The system consists of several candidate solutions, and sets the maximum number of iterations, population size, and performance evaluation index. One candidate solution is input into a PWM generator to generate a PWM signal, which is then input into the power conversion module. The square integral of the DC current error output by the power conversion module is calculated as the performance evaluation index. Each hired bee searches for new candidate solutions near the current candidate solution and calculates the performance evaluation index of the new candidate solution. If the performance of the new candidate solution is better than that of the original candidate solution, the original candidate solution is updated to the new candidate solution. Observer bees select a region of superior solutions from the candidate solutions based on probability and search for new candidate solutions within this region. For candidate solution regions that fail to update consecutively for several times, scout bees randomly generate new candidate solution regions. The optimal parameter combination in the current iteration is recorded, and the following steps are repeated: one candidate solution is input into a PWM generator to generate a PWM signal, which is then input into the power conversion module. The square integral of the DC current error output by the power conversion module is calculated, until the maximum number of iterations or the performance evaluation index converges.

[0100] In some embodiments, the DC-DC converter and the buck-boost DC-DC converter of the conversion module output DC current based on the M signal.

[0101] In some embodiments, the DC current output by the conversion module is acquired by the acquisition unit, so that the control unit, based on the DC current output by the conversion module, uses the integral of the squared DC current error as an evaluation index and employs an artificial bee colony algorithm to determine the minimum value of the integral of the squared DC current error corresponding to G. p and G i As the optimal G p and G i .

[0102] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A transformer neutral point DC bias suppression heuristic control compensation device, characterized in that, The device includes: The DC bias detection unit is used to acquire the DC bias voltage of the neutral point of the transformer under test relative to ground. A polarity determination unit, connected to the DC bias detection unit, is used to receive the DC bias voltage and determine the DC bias polarity of the DC bias voltage; The control unit, connected to the polarity discrimination unit, is used to adjust the proportional gain G based on the DC bias polarity using an artificial bee colony algorithm. p Integral gain G i Perform offline optimization; The power conversion module, connected to the control unit, is used to convert G... p and G i Output DC current so that the control unit determines the optimal G p and G i And based on the optimal G p and G i Output a compensated DC current for the DC bias of the neutral point of the transformer under test.

2. The apparatus according to claim 1, characterized in that, The polarity discrimination unit includes a voltage comparator, wherein the non-inverting input of the voltage comparator is connected to the output of the DC bias detection unit to receive the DC bias voltage, the inverting input of the voltage comparator is grounded, and the output of the voltage comparator outputs a DC bias polarity discrimination signal.

3. The apparatus according to claim 2, characterized in that, The control unit includes a microprocessor and a pulse width modulation (PWM) generator, wherein the microprocessor is connected to the output of the voltage comparator and is used to determine the G based on the DC bias polarity discrimination signal. p and G i The adjustment direction is to utilize the artificial bee colony algorithm to adjust the G p and G i Adjustments are made so that the PWM generator is based on the adjusted G. p and G i Output PWM signal, The artificial bee colony algorithm is used to analyze the G. p and G i Adjustments will be made, including: Based on the preset optimization parameter range, several candidate solutions are randomly generated. Each candidate solution includes a set of control parameters (G). p G i The system consists of ( ) components, and sets the maximum number of iterations, population size, and performance evaluation metrics; The candidate solutions from the plurality of candidate solutions are input into the PWM generator to generate a PWM signal, and the PWM signal is input into the power conversion module. The square integral of the DC current error output by the power conversion module is calculated as the performance evaluation index. Each hired bee searches for new candidate solutions near the current candidate solution and calculates the performance evaluation index of the new candidate solution. If the performance of the new candidate solution is better than that of the original candidate solution, the original candidate solution is updated to the new candidate solution. The bees are observed to select a region of superior solutions from the candidate solutions based on probability, and then search for new candidate solutions within the region of superior solutions. For candidate solution regions that fail to update several times in a row, the scout bee randomly generates new candidate solution regions; Record the optimal parameter combination in the current iteration, and repeat the steps: input the candidate solutions from the candidate solutions into the PWM generator to generate a PWM signal, and input the PWM signal into the power conversion module, calculate the square integral of the DC current error output by the power conversion module, until the step: for the candidate solution region that has failed to update for several consecutive times, the scout bee randomly generates a new candidate solution region, until the maximum number of iterations or the performance evaluation index converges.

4. The apparatus according to claim 3, characterized in that, The power conversion module includes a boost DC-DC converter and a buck-boost DC-DC converter. The boost DC-DC converter and the buck-boost DC-DC converter are respectively connected to the PWM generator and are used to output DC current based on the PWM signal.

5. The apparatus according to claim 1, characterized in that, The device further includes a data acquisition unit for acquiring the DC current output by the power conversion module, so that the control unit, based on the DC current output by the power conversion module, uses the integral of the squared DC current error as an evaluation index and employs the artificial bee colony algorithm to determine the minimum value of G corresponding to the integral of the squared DC current error. p and G i As the optimal G p and G i .

6. A heuristic control compensation method for DC bias suppression at the neutral point of a transformer, characterized in that, The DC bias voltage of the neutral point of the transformer under test relative to ground is acquired by the DC bias detection unit. The polarity determination unit receives the DC bias voltage and determines the DC bias polarity of the DC bias voltage. Through the control unit, based on the DC bias polarity, the proportional gain G is adjusted using an artificial bee colony algorithm. p Integral gain G i Perform offline optimization; Through the power conversion module, based on the G p and G i Output DC current so that the control unit determines the optimal G based on the DC current. p and G i And based on the optimal G p and G i Output a compensated DC current for the DC bias of the neutral point of the transformer under test.

7. The method according to claim 6, characterized in that, The DC bias voltage is received at the non-inverting input of the voltage comparator of the polarity discrimination unit, the inverting input is grounded, and the output terminal outputs a DC bias polarity discrimination signal.

8. The method according to claim 7, characterized in that, The microprocessor of the control unit receives the DC bias polarity discrimination signal to determine the G. p and G i The direction of adjustment is then used to apply the artificial bee colony algorithm to the G. p and G i Adjustments are made, and the PWM generator of the control unit is based on the adjusted G. p and G i Output PWM signal, where, The artificial bee colony algorithm is used to analyze the G. p and G i Adjustments will be made, including: Based on the preset optimization parameter range, several candidate solutions are randomly generated. Each candidate solution includes a set of control parameters (G). p G i The system consists of ( ) components, and sets the maximum number of iterations, population size, and performance evaluation metrics; The candidate solutions from the plurality of candidate solutions are input into the PWM generator to generate a PWM signal, and the PWM signal is input into the power conversion module. The square integral of the DC current error output by the power conversion module is calculated as the performance evaluation index. Each hired bee searches for new candidate solutions near the current candidate solution and calculates the performance evaluation index of the new candidate solution. If the performance of the new candidate solution is better than that of the original candidate solution, the original candidate solution is updated to the new candidate solution. The bees are observed to select a region of superior solutions from the candidate solutions based on probability, and then search for new candidate solutions within the region of superior solutions. For candidate solution regions that fail to update several times in a row, the scout bee randomly generates new candidate solution regions; Record the optimal parameter combination in the current iteration, and repeat the steps: input the candidate solutions from the candidate solutions into the PWM generator to generate a PWM signal, and input the PWM signal into the power conversion module, calculate the square integral of the DC current error output by the power conversion module, until the step: for the candidate solution region that has failed to update for several consecutive times, the scout bee randomly generates a new candidate solution region, until the maximum number of iterations or the performance evaluation index converges.

9. The method according to claim 8, characterized in that, The power conversion module outputs DC current based on the PWM signal through its boost DC-DC converter and buck-boost DC-DC converter.

10. The method according to claim 6, characterized in that, The acquisition unit collects the DC current output by the power conversion module, enabling the control unit to determine the minimum value of the DC current error square integral corresponding to G based on the DC current output by the power conversion module, using the DC current error square integral as an evaluation index and employing the artificial bee colony algorithm. p and G i As the optimal G p and G i .

Images