A transformer area terminal voltage control method and system based on freezing mechanism

Abstract

The application relates to a kind of terminal voltage control method and system based on freezing mechanism of transformer area, which comprises the following steps: real-time acquisition of the voltage of terminal public connection point of transformer area;Real-time identification of the impedance of the terminal public connection point of transformer area based on differential disturbance;Real-time calculation of voltage change rate and judgment whether the voltage change rate exceeds threshold, if exceeds, using the impedance identified last time or impedance history value as perceived impedance, if does not exceed, using the impedance identified this time as perceived impedance;Based on the perceived impedance and the voltage, the compensation current injected into the terminal public connection point of transformer area is calculated.The beneficial effects of the application are that the impedance is identified in real time by injecting differential disturbance into the public connection point to realize the terminal voltage control of transformer area, and the freezing mechanism is innovatively introduced to ensure the robustness of the identification process.

Description

Technical Field

[0001] This invention belongs to the field of power system power quality management technology, specifically relating to a voltage control method applied to the end of a distribution substation, and more particularly to a control method and system that achieves multi-objective compensation by online identification of system impedance based on differential disturbance and recursive least squares (RLS) without the need for load-side current sensors. Background Technology

[0002] The current distribution network's end-point loads (such as charging piles and high-power home appliances) are characterized by diversification and large capacity, leading to increasingly prominent problems such as low voltage, three-phase imbalance, and voltage fluctuations at the distribution network's end. The main technical solutions in the industry for addressing these issues include integrated power quality control (UPQC) devices, energy storage-type voltage management devices, FVR (flexible AC / DC interconnection devices), and phase-switching switches. However, these solutions all require precise control from physical current sensors on the load side, resulting in numerous drawbacks such as high system complexity, low dynamic response accuracy, and poor reliability and adaptability. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a method for controlling the voltage at the end of a transformer substation based on a freezing mechanism, wherein the method includes the following steps:

[0004] Real-time acquisition of voltage at the common connection point at the end of the transformer area;

[0005] The impedance of the common connection point at the end of the transformer area is identified in real time based on differential disturbance;

[0006] The voltage change rate is calculated in real time and it is determined whether the voltage change rate exceeds the threshold. If it exceeds the threshold, the impedance identified in the previous test or the historical impedance value is used as the sensing impedance. If it does not exceed the threshold, the impedance identified in the current test is used as the sensing impedance.

[0007] The compensation current injected into the common connection point at the end of the transformer area is calculated based on the sensed impedance and the voltage.

[0008] Another aspect of the present invention provides a voltage control system for the end of a transformer substation based on a freezing mechanism, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the aforementioned method.

[0009] At least one beneficial effect of the present invention is that it achieves voltage control at the end of the transformer area by injecting differential disturbances into the common connection point to identify impedance in real time, and innovatively introduces a freezing mechanism to ensure the robustness of the identification process. Attached Figure Description

[0010] Figure 1 A flowchart of the freezing mechanism regulation method is introduced.

[0011] Figure 2 A flowchart of the load mutation detection and freezing mechanism regulation method is introduced.

[0012] Figure 3 A schematic diagram of the application scenario of the system of this invention.

[0013] Figure 4 Overall flowchart of the control method of this invention.

[0014] Figure 5 Block diagram of the principle of online system impedance identification.

[0015] Figure 6 Block diagram of compensation command synthesis and tracking control principle.

[0016] Figure 7 Differential disturbances and voltage response.

[0017] Figure 8 Load mutation and RLS freezing mechanism.

[0018] Figure 9 Comparison of compensation effects. Detailed Implementation

[0019] The following embodiments further illustrate the content of the present invention, but should not be construed as limiting the present invention. Any modifications or substitutions made to the methods, steps, or conditions of the present invention without departing from the spirit and essence of the invention are within the scope of the present invention.

[0020] The current loads at the end of the distribution network (such as charging piles and high-power home appliances) are characterized by diversification and large capacity, leading to increasingly prominent problems such as low voltage, three-phase imbalance, and voltage fluctuations at the end of the distribution area. The main technical solutions in the industry to solve these problems all have shortcomings, as detailed below:

[0021] Upgraded Power Quality Control (UPQC) devices are advanced power electronic devices that integrate series and parallel compensation functions. They can simultaneously and in real-time compensate for voltage quality issues (such as voltage dips and harmonics) on the power supply side and current quality issues (such as harmonics and reactive power) on the load side, providing a stable voltage to the load and injecting current into the grid. However, UPQC devices are connected in series between the power supply and load lines, requiring a power outage for installation. Furthermore, the load current passes entirely through the voltage regulator, posing a high risk of overload burnout if the load fluctuates significantly. Although most series-connected products are designed with bypass switches, in the event of a transient fault, the bypass switch still requires closing time, potentially causing a short-term power outage that affects the user's electricity supply.

[0022] Energy storage voltage regulation devices are devices that highly integrate energy storage systems (usually batteries, such as lithium-ion batteries) with power electronic converters (PCS) and intelligent control systems. They dynamically regulate the voltage at the connection point by absorbing, storing, or releasing electrical energy in real time, thus solving voltage quality problems in the power grid. However, energy storage devices are generally large and heavy, requiring large supports or foundations for installation, and transportation and installation require specialized large equipment, often necessitating power outages for construction. Furthermore, the response speed of energy storage devices is typically on the order of seconds, mainly relying on discharge to boost voltage, resulting in a single function.

[0023] FVR (Flexible AC / DC Interconnection Device): This is a device that combines power electronics technology and intelligent control, primarily used to achieve flexible conversion and interconnection of AC and DC power. Its core functionality involves real-time adjustment of parameters such as voltage and frequency using power electronic devices (such as IGBTs and PWM controllers). A typical application is converting the AC voltage from a transformer substation to DC voltage, which is then transmitted to the load end via DC lines. However, FVR devices require full load operation, which places requirements on both device power and load power; typically, the device power must be at least 1.1 times the load power.

[0024] A phase-change switch is a device that automatically switches the power supply phase (A, B, or C) to alleviate three-phase load imbalance. However, phase-change switches need to be installed in series between the power supply line and the load, and power must be disconnected during installation. If the device malfunctions, it will directly affect the normal power supply to the load. This switch achieves three-phase current balance control by transferring the load, and typically requires one main unit and multiple slave units working together in a distribution area. However, if the load is a heavy-duty device such as a 7kW home charging station, switching the phase to any phase may cause new imbalance problems, making it difficult to achieve precise load balancing.

[0025] In particular, the aforementioned technical solutions all require precise control of the physical current sensor on the load side, which leads to many defects in the system, such as high complexity, low dynamic response accuracy, poor reliability and adaptability.

[0026] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0027] like Figure 1 As shown, in some embodiments, a method for controlling the voltage at the end of a transformer substation based on a freezing mechanism is involved, the method comprising the following steps:

[0028] P1: Real-time acquisition of voltage at the common connection point at the end of the transformer area;

[0029] P2: Based on differential disturbances, the impedance (also known as system impedance, system complex impedance, Z) of the common connection point (PCC) at the end of the transformer area is identified in real time. sys(This is the equivalent impedance viewed from the PCC point towards the grid side).

[0030] P3: Calculate the voltage change rate in real time and determine whether the voltage change rate exceeds the threshold. If it exceeds the threshold, use the previously identified impedance or the historical impedance value as the sensing impedance. If it does not exceed the threshold, use the current identified impedance as the sensing impedance.

[0031] P5: Calculate the compensation current injected into the common connection point at the end of the transformer area based on the sensed impedance and the voltage.

[0032] In these embodiments, the method for real-time acquisition of the voltage at the end of the transformer substation's common connection point (PCC) in P1 is not specifically limited; for example, the voltage signal acquisition and processing method in S10 can be used. Similarly, the method for real-time identification of the PCC impedance based on differential disturbances in P2 is not specifically limited; for example, the online identification method for system impedance based on differential disturbances in S20 can be used. These embodiments achieve one of the key points of the present invention: based on the identified system impedance and installation point voltage, the system state is reconstructed through algorithms, achieving precise control without the need for physical current sensors on the load side.

[0033] like Figure 2 As shown, in some embodiments, the method further includes the following steps:

[0034] P4: Perform a physical rationality check on the impedance identified this time. If the check fails, use the impedance identified last time or the historical impedance value as the perceived impedance. If the check succeeds, use the impedance identified this time as the perceived impedance.

[0035] In these embodiments, software algorithms replace hardware sensors, eliminating the need for load-side current sensing devices and reducing cost and complexity. More importantly, by introducing multiple safety mechanisms (such as load surge freezing and impedance verification), the system exhibits extremely high reliability when facing disturbances in the real power grid, resolving the vulnerability of the core algorithm. The system can continuously sense the power grid state and automatically distinguish between steady-state and transient states. Under different operating conditions, it can intelligently select the optimal strategy to maintain optimal performance, significantly improving the adaptability and practicality of the governance effect.

[0036] In some implementations, P2: Based on differential disturbances, the impedance of the common connection point at the end of the transformer area is identified in real time, specifically including the following sub-steps:

[0037] P21: Generate the differential disturbance and inject the differential disturbance into the common connection point; the differential disturbance includes a current differential disturbance or a voltage differential disturbance;

[0038] P22: Sample the voltage change generated by the differential perturbation exciting the common connection point;

[0039] P23: The impedance is identified based on an identification algorithm, which includes one or more of the following: recursive least squares method or its improved algorithm, Kalman filtering or its improved algorithm, model reference adaptive system or its improved algorithm, stochastic gradient or its improved algorithm.

[0040] In a preferred embodiment of this invention, a small-amplitude differential current perturbation is injected. The perturbation sequence can be of various forms, such as PRBS (pseudo-random binary sequence), M-sequence, Chirp (linear sweep signal), sinusoidal sweep, or random white noise, as long as its energy spectrum has sufficient excitation energy in the range of 50Hz to 1kHz. Alternatively, a voltage differential perturbation can be injected, and the change in the system current response can be observed; the system impedance can then be solved accordingly. The mathematical essence is reciprocal, both following the fundamental relationship of Ohm's law Z = ΔV / ΔI.

[0041] In these embodiments, the method for P21: generating the differential perturbation and injecting the differential perturbation into the common connection point is not specifically limited; for example, the method for S21: generating and injecting the differential perturbation can be adopted. Similarly, the method for P22: sampling the voltage change generated by the differential perturbation exciting the common connection point is not specifically limited; for example, the method for S22: response detection and signal extraction can be adopted.

[0042] In some embodiments, the identification algorithm is an improved algorithm of recursive least squares, wherein the improved algorithm uses recursive least squares and introduces a forgetting factor, wherein:

[0043] The mathematical expression for the Recursive Least Squares (RLS) method is shown in Equation I:

[0044] y(k)=φ T (k)θ Equation I

[0045] Where y(k) represents the system output; φ T (k) represents the system input; θ represents the parameter to be identified; k represents the sampling time index value, which is a non-negative integer; T represents the transpose sign; the meanings of the above symbols can be found in S23: the meaning of the response conformance in online impedance identification and safety assurance based on RLS.

[0046] The forgetting factor ranges from 0.95 to 0.999.

[0047] In a preferred embodiment of this invention, recursive least squares (RLS) is used for online impedance identification. The RLS forgetting factor λ can be adjusted within the range of 0.95 to 0.999, or improved algorithms such as variable forgetting factor (VFF-RLS) can be used to balance parameter tracking speed and steady-state identification accuracy. The identification algorithm is not limited to this; any algorithm capable of online estimation of system parameters based on input and output data can be used as an alternative. For example:

[0048] Kalman filter: can be used for impedance identification, especially in noisier environments, and can provide better state estimation performance.

[0049] Model Reference Adaptive System (MRAS): A reference model can be constructed, and the impedance parameters can be adjusted through an adaptive law so that the output of the actual system tracks the output of the reference model.

[0050] Other stochastic gradient algorithms, such as LMS (Least Mean Square), may converge slower than RLS, but they require less computation and can be used as alternatives on platforms with limited computing resources.

[0051] In some embodiments, a simplified preset and self-learning mode without impedance identification is also included, which can bypass the online impedance identification process in specific scenarios where the power grid structure is extremely stable and impedance changes are negligible. The system can provide a "preset mode" that allows operators to manually preset a fixed system impedance value Z in the system based on field measurements or historical data. preset The preset impedance value can be automatically compared with the online identification result during stable grid periods such as light loads at night. If the deviation continues to exceed 10%, the user is prompted to update, thus forming a "light load self-calibration" mechanism, making the fixed impedance value gradually approach the true value. Subsequent compensation calculations will be based on this (preset or self-learned) impedance. This mode sacrifices some adaptability under certain operating conditions, but reduces system complexity, while still utilizing the core idea of ​​this invention: "compensation based on impedance model".

[0052] These embodiments achieve one of the key points of the present invention, such as injecting differential perturbations and using the RLS algorithm to identify system impedance online, and innovatively introducing a load change detection and freezing mechanism to ensure the robustness of the identification process.

[0053] In some implementations, if the voltage change rate exceeds the threshold, the parameter update of the identification algorithm needs to be frozen immediately until the voltage change rate does not exceed the threshold; the impedance identified this time includes equivalent resistance and equivalent reactance; the impedance identified last time is the effective value of the impedance identified last time; the impedance history value is the impedance history value after smoothing and filtering.

[0054] In these embodiments, the load mutation detection and RLS freezing, impedance rationality verification and smoothing methods of the security assurance strategy of step S23 can be adopted.

[0055] In some implementations, P5: Calculating the compensation current injected into the common connection point at the end of the transformer area based on the sensed impedance and the voltage, specifically includes the following sub-steps:

[0056] P51: Real-time calculation of compensation amount based on impedance model:

[0057] P52: Compensation instruction synthesis and tracking output;

[0058] P53: Closed-loop adaptive operation and safety protection.

[0059] In some implementations, P51: Real-time calculation of compensation based on the impedance model specifically includes one or more of the following sub-steps:

[0060] a. The total current is calculated based on the identified impedance and the acquired voltage to calculate the total current vector;

[0061] b. Unbalanced component extraction: Apply the symmetric component method to the calculated total current vector to solve and separate the negative sequence current component, and calculate the unbalanced compensation target based on the negative sequence current component.

[0062] c. Zero-sequence compensation enable judgment: When it is determined to be a three-phase four-wire system and the neutral line is reliably connected, the zero-sequence current component is calculated, and the zero-sequence compensation target is calculated based on the zero-sequence current component.

[0063] d. Reactive current component extraction: Perform synchronous rotation coordinate transformation on the total current vector to extract the reactive current component, and calculate the reactive current compensation target based on the reactive current component;

[0064] e. Voltage regulation compensation calculation: The voltage deviation at the common connection point is input into the proportional-integral controller to calculate the voltage regulation current component used to compensate for the voltage deviation. The parameters of the proportional-integral controller are adaptively adjusted according to the impedance to achieve the voltage regulation compensation target.

[0065] The preferred embodiment of this invention achieves multi-objective coordinated control, including voltage stabilization, reactive power compensation, and imbalance mitigation. In practical applications, the compensation functions can be reduced or combined according to specific engineering needs and cost constraints. Furthermore, the system can operate at a reduced speed based on the actual load conditions of the transformer substation: for example, only phase A imbalance can be compensated, or only loads within the 0-50A current range can be mitigated, while other phases or loads exceeding this range enter standby energy-saving mode. Other configurable options include:

[0066] It can be configured to perform only the three-phase imbalance control function (output only -I2, -I0).

[0067] It can be configured to perform only reactive power compensation (output -Iq only) function and be used as an advanced static var generator (SVG).

[0068] It can be configured to perform only the end voltage regulation (output Iv only) function.

[0069] Any combination of the above functions falls within the protection scope of this invention.

[0070] In these embodiments, S30: the method of real-time calculation of compensation amount based on impedance model can be adopted, which realizes one of the key points of the present invention, namely, according to the reconstructed system state, synchronously calculates the commands for voltage stabilization, balancing and reactive power compensation, and can adaptively adjust the control parameters according to the system impedance characteristics, and has command coordination function to realize collaborative optimization control.

[0071] In some implementations, P52: Compensation instruction synthesis and tracking output specifically includes the following steps:

[0072] The instruction synthesis generates a total compensation current instruction based on the negative sequence current component, zero sequence current component, reactive current component, or voltage stabilizing current component.

[0073] The tracking control performs closed-loop tracking control on the command current and generates a PWM wave to drive the IGBT;

[0074] The driver executes the IGBT to output the compensation current based on the PWM wave.

[0075] In these embodiments, the method of S40: compensating instruction synthesis and tracking output can be adopted.

[0076] In some implementations, P53: Closed-loop adaptive operation and safety protection specifically includes one or more of the following sub-steps:

[0077] The continuous updating ensures that the sensed impedance is continuously updated.

[0078] The disturbance management system evaluates the signal-to-noise ratio of the impedance identification in real time. If the confidence level is too low, it automatically increases the amplitude of the differential disturbance slightly and restores it after the measurement is completed.

[0079] In these embodiments, the S50 method of closed-loop adaptive operation and safety protection can be adopted.

[0080] In other embodiments, a voltage control system for the end of a transformer substation based on a freezing mechanism is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the aforementioned method.

[0081] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0082] Application scenarios of the voltage control system at the end of the transformer substation based on the freezing mechanism of this invention include: Figure 3 As shown. The system is connected in parallel to the point of common coupling (PCC) of the low-voltage distribution line at the end of the distribution area, and is flexibly connected to the power grid. Its core function is to comprehensively address power quality problems such as low voltage at the end, voltage fluctuations, and three-phase imbalance caused by factors such as nonlinear loads, unbalanced loads, and reactive power loss caused by real-time high-precision monitoring of the PCC voltage and dynamic injection of controllable compensation current.

[0083] This control method is executed by an embedded DSP processor as the core control unit, and its overall control flow is as follows: Figure 4 As shown, the specific steps include:

[0084] S10: Voltage Signal Acquisition and Processing

[0085] The three-phase voltage instantaneous signals Ua, Ub, and Uc at the installation point (coinciding with the PCC location) are acquired in real time and synchronously using a voltage sensor and a high-precision analog-to-digital converter (ADC). After preprocessing by digital filtering (such as low-pass filtering or notch filtering), the fundamental voltage positive sequence component and other required components that can be used for subsequent high-precision calculations are obtained.

[0086] S20: Online Identification of System Impedance Based on Differential Perturbation

[0087] This step injects a specific current differential disturbance into the power grid and detects its voltage response. Based on the recursive least squares (RLS) method, it achieves high-precision online sensing of the system's impedance characteristics. Its principle block diagram is shown below. Figure 5 As shown.

[0088] S21: Differential perturbation generation and injection.

[0089] The controller (hereinafter referred to as the controller) of the transformer terminal voltage control system based on the freezing mechanism is based on the compensation current command I of the previous control cycle. ref (k−1), actively superimposed with a small differential perturbation current ΔI comp The amplitude of this disturbance is 0.1% to 1% of the device's rated current Id, and its phase or sequence can be varied to provide sufficient excitation. For example... Figure 7 As shown, at t=3ms, the device injects a differential current pulse with an amplitude of 0.5A into the PCC, and the PCC voltage generates an instantaneous response of about 2.1mV.

[0090] S22: Response Detection and Signal Extraction

[0091] High-precision sampling is achieved by the differential perturbation ΔIcomp The change in grid voltage ΔU caused by the excitation.

[0092] S23: Online Impedance Identification and Safety Assurance Based on RLS

[0093] Establish a system model: The system at the point of common connection (PCC) is equivalent to a Thevenin circuit, with the following equations:

[0094] U(k) = Un - Z sys ·I total (k)

[0095] Where U(k) is the voltage measured at the PCC point at the k-th sampling time, Un is the ideal grid voltage, and Z... sys The complex impedance of the system to be determined is I. total (k) is the total current I at time k. total (k)=I load (k)+I comp (k), I comp (k) is the compensation current. This is to eliminate the unobservable quantity U. n The difference between two adjacent sampling periods is calculated, and U is assumed to be... n With Z sys Short-term invariance, I load The change is slow, so we obtain the voltage change ΔU at time k:

[0096] ΔU(k)≈-Z sys ·ΔI comp (k)

[0097] Discretize the above equation and construct the standard least squares scheme y(k)=φ T (k)θ:

[0098] System output: y(k) = ΔU α (k) or ΔU β (k), that is, the voltage difference in the αβ coordinate system, ΔU α (k) represents the change in voltage along the α-axis at time k (real part); ΔU β (k) represents the change in voltage along the β axis at time k (the imaginary part).

[0099] System input: φ T (k)=[−ΔI comp,α (k),−ΔI comp,β (k)],−ΔI comp,α (k) represents the negative α-axis compensation current differential disturbance component at time k; −ΔI comp,β (k) represents the negative β-axis compensation current differential disturbance component at time k;

[0100] Parameter to be identified: θ=[R sys ,X sys ] T R sys Represents the system's equivalent resistance (real part, reflecting active power loss), X sys This represents the system's equivalent reactance (imaginary part, reflecting reactive power characteristics).

[0101] By employing the recursive least squares (RLS) method and introducing a forgetting factor λ (typically ranging from 0.95 to 0.999), the parameter θ can be identified online in real time, thus yielding the system's complex impedance Z. sys .

[0102] Security strategy:

[0103] Load surge detection and RLS freeze: Real-time calculation of the voltage change rate |dU / dt|. If its value exceeds a preset threshold, a severe disturbance in the power grid is determined, and the parameter updates of the RLS algorithm are immediately frozen to prevent erroneous data from contaminating the impedance identification results until the system returns to stability. Figure 8 As shown, when the load current step causes the voltage at point PCC to drop by 25V at 40ms, the system immediately sets the RLS freeze signal (from 0→1) and suspends the impedance parameter update.

[0104] Impedance rationality verification and smoothing: for the identified R sys and X sys Make a physical plausibility judgment (e.g., R) sys >0, |Z sys | Within the typical distribution network impedance range; where R sys Represents the system's equivalent reactance; |Z sys | represents the impedance magnitude, reflecting the system's total resistance to current (including the combined effect of resistance and reactance). If the verification fails, this value is discarded, and the previous valid value or a smoothed and filtered historical value is used.

[0105] S30: Real-time calculation of compensation amount based on impedance model:

[0106] a. Total Current Calculation

[0107] Based on the identified system impedance Z sys Calculate the total current vector based on the measured voltage U:

[0108]

[0109] U n This represents the ideal harmonic-free voltage of the system, that is, the theoretical reference voltage value under conditions of no harmonic distortion or other interference. It should be noted that U here... nUsed only as a mathematical symbol in theoretical derivations, its specific value does not need to be calculated in actual compensation processes. The target current component for compensation can be directly derived from the total current vector I. total Extracted from U without calculating it separately. n and I total The specific value is the compensation current I output by the device itself. comp The quantity is known.

[0110] b. Imbalanced component extraction

[0111] The calculated total current vector I total The negative sequence current component I2 is calculated and separated by applying the symmetrical component method, and the compensation target is −I2.

[0112] c. Zero-sequence compensation enable judgment

[0113] The zero-sequence current component I0 is calculated only when the system is determined to be a three-phase four-wire system and the neutral line is reliably connected, with the compensation target being −I0. Otherwise, the zero-sequence compensation component is set to zero.

[0114] d. Reactive component extraction

[0115] For the total current vector I total Perform synchronous rotating coordinate transformation (dq transformation) to extract its reactive current component I. q The compensation target is −I q .

[0116] e. Calculation of voltage stabilization compensation

[0117] Define the voltage deviation at the installation point as ΔU=U ref −U, where U ref The reference value is the rated voltage. This deviation ΔU is input into a proportional-integral (PI) controller to calculate the current component I used to compensate for the voltage deviation. v :

[0118] I v =K p ·ΔU+K i ·∫ΔUdt

[0119] Wherein, controller parameter K p K i Based on the system impedance Z identified online sys The modulus value is adaptively adjusted: when |Z sys When |Z is large (weak grid), appropriately reduce the control gain to ensure system stability; when |Z sys | When the voltage is relatively low (strong power grid), appropriately increase the control gain to accelerate the response.

[0120] Compensation instruction coordination strategy: for I v(Voltage regulation requirements) and −I q Coordinate (reactive power compensation requirements). Set the priority of voltage deviation; when |ΔU| exceeds a certain limit, prioritize ensuring the voltage regulation command I. v The output.

[0121] S40: Compensation Instruction Synthesis and Tracking Output

[0122] This step completes the transition from decision-making to execution, and its principle block diagram is as follows: Figure 6 As shown.

[0123] S41: Instruction Synthesis

[0124] The calculated compensation components (−I2, −I0, −I) are then... q I v Vector synthesis is performed in the dq coordinate system, and after the above coordination strategy and output limiting processing, the total compensation current command I is generated. ref .

[0125] S42: Tracking Control

[0126] A frequency adaptive proportional-resonant (PR) control strategy based on SOGI-FLL (Second-Order Generalized Integrator - Frequency-Locked Loop) is adopted to control the command current I. ref Closed-loop tracking control is implemented to ensure accurate tracking even when the grid frequency fluctuates, generating PWM waves to drive the IGBT.

[0127] S43: Driver Execution

[0128] The PWM wave is sent to the drive circuit to drive the IGBT converter bridge arm to operate, causing it to output a high-precision compensation current I that matches the command. comp This current value will be stored and used for impedance identification calculations in the next cycle.

[0129] S50: Closed-loop adaptive operation and safety protection

[0130] Repeat steps S10 to S40 to form a high-speed closed-loop control system with a millisecond-level timeframe.

[0131] S51: Continuously Updated

[0132] Make the system impedance Z sys Continuously updated, the compensation strategy can be adaptively adjusted to achieve dynamic management of power quality. Its application scenarios include... Figure 1 As shown, the compensation effect is compared to, for example Figure 9 As shown.

[0133] S52: Safety Protection and Fault Tolerance

[0134] Fault ride-through: Real-time monitoring of grid voltage. If a fault state such as severe overvoltage, undervoltage, or frequency exceeding the limit is detected, compensation is immediately stopped and PWM output is locked, entering protection standby mode. The initialization process needs to be re-executed after the grid returns to normal.

[0135] Disturbance Management: Real-time assessment of the signal-to-noise ratio of impedance identification; if the confidence level is too low, the differential disturbance ΔI can be automatically increased slightly. comp The amplitude is measured and then restored after the measurement is completed.

[0136] The preferred embodiment of this invention uses a DSP as the core processor. However, its hardware implementation platform is not limited to this. Any computing platform with sufficient computing power, ADC sampling capability, and PWM output capability can be used as an alternative, for example:

[0137] FPGA (Field Programmable Gate Array): It can be used to realize ultra-high-speed parallel computing, and is especially suitable for algorithms that require multiple parallel resonators, such as PR control.

[0138] ARM Cortex-M series / RISC-V multi-core microcontrollers: can serve as a cost-effective alternative in applications with slightly lower performance requirements.

[0139] SoC (System-on-a-Chip, such as DSP+ARM heterogeneous chip) or FPGA-DSP heterogeneous chip: can achieve higher integration or domestic substitution, and assign tasks such as impedance identification, compensation calculation, and PWM generation to different cores to improve the overall performance and reliability of the system.

[0140] Each individual power management unit of this invention can operate independently. Furthermore, in large distribution areas, multiple power management units of this invention can form a distributed collaborative control system via communication networks (such as HPLC, RS485, and wireless private networks). After power-on, each unit can automatically broadcast its address information via communication methods such as HPLC, and automatically elect a master station based on a preset algorithm (such as based on received signal strength RSSI or equipment performance indicators), achieving a "zero-configuration" self-organizing network. In the event of a master station failure, the system can automatically re-elect a master station, ensuring high system availability. The master station assigns compensation targets (such as total reactive power and balance) to other slave units based on global information (such as transformer outlet voltage and total load). Each slave unit still performs precise control based on local impedance identification, thereby achieving hierarchical and zoned collaborative optimization of power quality across the entire distribution area.

[0141] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. The above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling the voltage at the end of a transformer substation based on a freezing mechanism, characterized in that, The method includes the following steps: Real-time acquisition of voltage at the common connection point at the end of the transformer area; The impedance at the common connection point at the end of the transformer area, viewed from the common connection point to the grid side, is identified in real time based on the differential disturbance. The impedance includes equivalent resistance and equivalent reactance. Specifically, it is identified based on the Thevenin equivalent circuit model and the negative correlation between the injected compensation current differential disturbance component and the generated voltage change. The identification employs a recursive least squares method that incorporates a forgetting factor, and the mathematical expression of the recursive least squares method is shown in Equation I: y(k)=φ T (k)θ Equation I; where the system output y(k) is the voltage change along the α-axis or β-axis in the αβ coordinate system, and the system input φ T (k) represents the corresponding negative α-axis or negative β-axis compensation current differential disturbance component, and the parameter to be identified is θ=[R]. sys X sys ] T Includes the system equivalent resistance R sys and system equivalent reactance X sys k represents the sampling time index value, which is a non-negative integer, and T represents the transpose sign; the forgetting factor ranges from 0.95 to 0.999; The voltage change rate is calculated in real time and it is determined whether the voltage change rate exceeds a threshold. If it does, the previously identified impedance or the historical impedance value is used as the sensing impedance, and the parameter update of the recursive least squares method used for real-time identification of the impedance is immediately frozen until the voltage change rate does not exceed the threshold. If it does not exceed the threshold, the impedance identified in this instance is used as the sensing impedance. The physical rationality of the impedance identified this time is verified. If the verification fails, the impedance identified last time or the historical impedance value is used as the sensed impedance. If the verification succeeds, the impedance identified this time is used as the sensed impedance. The compensation current injected into the common connection point at the end of the transformer area is calculated based on the sensed impedance and the voltage.

2. The method for controlling the voltage at the end of a transformer substation based on a freezing mechanism according to claim 1, characterized in that, The impedance at the common coupling point at the end of the transformer substation, viewed from the common coupling point towards the grid side, is identified in real time based on differential disturbances. This process includes the following sub-steps: The differential disturbance is generated and injected into the common connection point at the end of the transformer area; the differential disturbance includes a current differential disturbance or a voltage differential disturbance. The voltage change generated by the differential disturbance excites the common connection point at the end of the transformer area; The impedance is identified based on an identification algorithm, which includes one or more of the following: recursive least squares method or its improved algorithm, Kalman filtering or its improved algorithm, model reference adaptive system or its improved algorithm, stochastic gradient or its improved algorithm.

3. The method for controlling the voltage at the end of a transformer substation based on a freezing mechanism according to claim 1, characterized in that, The previously identified impedance is the effective value of the previously identified impedance; the historical impedance value is the historical impedance value after smoothing and filtering.

4. The method for controlling the voltage at the end of a transformer substation based on a freezing mechanism according to claim 3, characterized in that, The calculation of the compensation current injected into the common connection point at the end of the transformer area based on the sensed impedance and the voltage specifically includes the following sub-steps: Real-time calculation of compensation amount based on impedance model; Compensation instruction synthesis and tracking output; Closed-loop adaptive operation and safety protection.

5. The method for controlling the voltage at the end of a transformer substation based on a freezing mechanism according to claim 4, characterized in that, Real-time calculation of compensation based on impedance model includes one or more of the following steps: a. The total current is calculated based on the identified impedance and the acquired voltage to calculate the total current vector; b. Unbalanced component extraction: Apply the symmetric component method to the calculated total current vector to solve and separate the negative sequence current component, and calculate the unbalanced compensation target based on the negative sequence current component. c. Zero-sequence compensation enable judgment: When it is determined to be a three-phase four-wire system and the neutral line is reliably connected, the zero-sequence current component is calculated, and the zero-sequence compensation target is calculated based on the zero-sequence current component. d. Reactive current component extraction: Perform synchronous rotation coordinate transformation on the total current vector to extract the reactive current component, and calculate the reactive current compensation target based on the reactive current component; e. Voltage regulation compensation calculation: The voltage deviation at the common connection point is input into the proportional-integral controller to calculate the voltage regulation current component used to compensate for the voltage deviation. The parameters of the proportional-integral controller are adaptively adjusted according to the impedance to achieve the voltage regulation compensation target.

6. The method for controlling the voltage at the end of a transformer substation based on a freezing mechanism according to claim 5, characterized in that, The compensation instruction synthesis and tracking output specifically includes the following steps: The instruction synthesis generates a total compensation current instruction based on the negative sequence current component, zero sequence current component, reactive current component, or voltage stabilizing current component. The tracking control performs closed-loop tracking control on the command current and generates a PWM wave to drive the IGBT; The driver executes the IGBT to output the compensation current based on the PWM wave.

7. The method for controlling the voltage at the end of a transformer substation based on a freezing mechanism according to claim 6, characterized in that, Closed-loop adaptive operation and safety protection specifically include one or more of the following steps: The continuous updating ensures that the sensed impedance is continuously updated. The disturbance management system assesses the signal-to-noise ratio of impedance identification in real time. If the confidence level is too low, it automatically increases the amplitude of the differential disturbance slightly and restores it after the measurement is completed.

8. A voltage control system for the end of a transformer substation based on a freezing mechanism, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

Images