Low-voltage power distribution network voltage flexible regulation method and system based on multi-compensation mode determination

Abstract

This invention discloses a method and system for flexible voltage regulation of low-voltage distribution networks based on multi-compensation mode determination, belonging to the field of voltage regulation technology. By real-time acquisition of voltage signals from the low-voltage distribution network and extraction of voltage amplitude, phase, and harmonic component information, long-process compensation gap characteristics are extracted based on a sliding time window. The system determines whether to enter a compensation mode and identifies the compensation type. When compensation is required, a dynamic phase compensation signal or an adaptive harmonic injection signal is generated based on the compensation type, and a composite control command is fused to drive the AC / DC / AC converter unit to output a compensation voltage to the grid. This solution overcomes the limitations of traditional single-amplitude compensation, achieving three-dimensional coordinated flexible regulation of the amplitude, phase, and harmonic components of the low-voltage distribution network voltage, and possesses intelligent sensing and precise compensation capabilities for long-process, complex power quality problems.

Description

Technical Field

[0001] This invention relates to the field of voltage regulation technology, and in particular to a method and system for flexible voltage regulation of low-voltage distribution networks based on multi-compensation mode determination. Background Technology

[0002] With the large-scale integration of intermittent and nonlinear loads such as distributed photovoltaic systems, energy storage systems, and electric vehicle charging stations into low-voltage distribution networks, power quality issues at the grid's end are becoming increasingly severe. Problems such as voltage sags / boosts, three-phase imbalances, harmonic pollution, and phase fluctuations not only affect the normal operation of sensitive electrical equipment but also restrict the absorption capacity of high-proportion renewable energy sources. Therefore, the need for rapid, precise, and multi-dimensional voltage management of low-voltage distribution networks is becoming increasingly urgent.

[0003] Existing low-voltage distribution network voltage regulation technologies mainly focus on amplitude compensation. For example, traditional passive compensation devices (such as parallel capacitors and reactors) have slow response speeds and can only perform coarse reactive power compensation, failing to achieve dynamic and continuous voltage regulation. Active compensation devices based on power electronics technologies, such as Dynamic Voltage Restorers (DVRs) and Static Synchronous Compensators (STATCOMs), can achieve rapid voltage support, but their mainstream solutions still focus on correcting the amplitude by injecting a series voltage in phase with the grid voltage. These methods have limited compensation capabilities for voltage waveform distortion (harmonics) caused by nonlinear loads and the increasingly prominent voltage phase shift problem, or require additional filters, increasing system complexity and cost.

[0004] Furthermore, the control strategies of existing active compensation devices are mostly based on instantaneous voltage deviations, which is a "passive" and "point-to-point" compensation mode. This mode lacks the ability to perceive and analyze the long-term evolution trend of voltage quality. When faced with complex power quality disturbances that last for a long time and have complex characteristics (such as "long-process compensation gaps" that simultaneously include harmonics, phase shifts, and amplitude fluctuations), the adjustment is often lagging and the accuracy is insufficient, making it difficult to achieve globally optimal flexible control.

[0005] In terms of topology, conventional two-level converters suffer from drawbacks such as high switching losses and high harmonic content in the output voltage, which limit the efficiency and performance of compensation devices. While the more advanced three-level topology offers superior performance, its application in comprehensive voltage management of low-voltage distribution networks, particularly its deep integration with a sophisticated control strategy capable of simultaneously and collaboratively addressing voltage amplitude, phase, and harmonic issues, has yet to yield a mature solution.

[0006] Therefore, there is an urgent need for a novel control method and device that can overcome the limitations of traditional single amplitude compensation, achieve three-dimensional coordinated flexible regulation of the amplitude, phase, and harmonic components of low-voltage distribution network voltage, and possess intelligent sensing and precise compensation capabilities for long-term, complex power quality problems. To this end, a flexible voltage control method and system for low-voltage distribution networks based on multi-compensation mode determination is proposed. Summary of the Invention

[0007] The main objective of this invention is to provide a method and system for flexible voltage regulation of low-voltage distribution networks based on multi-compensation mode determination. From feature extraction to closed-loop feedback and predictive learning, it constitutes a self-sensing, self-decision-making, and self-executing intelligent regulation framework, providing core equipment and technical paths for building a future proactive and flexible intelligent distribution network, and can effectively solve the problems in the background technology.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A low-voltage distribution network voltage flexible control system based on multi-compensation mode determination includes a voltage acquisition and synchronization unit, a core controller, and an AC / DC / AC converter unit. The AC / DC / AC converter unit adopts a three-level three-phase half-bridge back-to-back topology, with its grid side connected to the low-voltage distribution network and its load side connected to the controlled load. The voltage acquisition and synchronization unit is used to acquire grid voltage and load voltage signals in real time, and extract voltage amplitude, phase and frequency information; The core controller is connected to the voltage acquisition and synchronization unit and the AC / DC / AC converter unit. It is used to perform long-process compensation gap feature extraction and generate a composite control command that integrates dynamic phase compensation and adaptive harmonic injection based on the extraction result, so as to drive the AC / DC / AC converter unit to output compensation voltage.

[0009] Furthermore, the core controller performs long-process compensation gap feature extraction, including: Set a sliding time window T, and calculate the voltage quality gap index within this window using the following formula. : ,in, for Reference voltage at time, for The measured voltage at that moment; The voltage quality gap index With preset threshold If a comparison is made, If so, then the current period is determined to be a gap period that needs to be compensated; Further identification of the gap type as at least one of amplitude gap, phase gap, or harmonic gap.

[0010] Furthermore, the core controller generates composite control instructions based on the gap type, specifically including: When a phase gap is identified, the dynamic phase compensation mode is activated to calculate the target phase compensation amount. And generate a phase offset Fundamental compensation voltage ,in, , As a reference phase, For the load voltage phase, For voltage phase, The fundamental angular frequency of the grid voltage. To compensate for the voltage amplitude; When a harmonic gap is identified, the adaptive harmonic injection mode is activated, and the main harmonic components are extracted through real-time harmonic analysis. And generate reverse-injected harmonic voltage , For harmonic order, For the first The amplitude of the subharmonic.

[0011] Furthermore, the core controller also includes a modulation module for adjusting the fundamental compensation voltage. and the reverse injected harmonic voltage The signals are superimposed and a segmented space vector modulation strategy is used to generate pulse width modulation signals to drive the power switching transistors of the three-level three-phase half-bridge back-to-back topology.

[0012] A method for flexible voltage regulation of low-voltage distribution networks based on multi-compensation mode determination, comprising: S1: Real-time acquisition of voltage signals from low-voltage distribution networks, and extraction of voltage amplitude, phase, and harmonic component information; S2: Extract long-process compensation gap features based on sliding time window, determine whether to enter compensation mode and determine compensation type; S3: If compensation is required, a dynamic phase compensation signal or an adaptive harmonic injection signal will be generated based on the type of compensation. S4: Based on the dynamic phase compensation signal or adaptive harmonic injection signal, generate a composite control command to control the output of the corresponding compensation voltage to the power grid, thereby realizing three-dimensional flexible control of the amplitude, phase and harmonics of the power grid voltage.

[0013] It also includes the following steps: Monitor the voltage quality of the compensated power grid; Calculate the error between the compensated voltage and the reference voltage. ; According to the error The parameters of the dynamic phase compensation signal or adaptive harmonic injection signal are dynamically adjusted to achieve closed-loop optimization control.

[0014] It also includes the following steps: Record gap characteristic data and control parameters during the historical compensation process; The acquired data is analyzed and trained using machine learning models to predict the trend of voltage quality changes in a specific future period. The compensation strategy is triggered and adjusted in advance before the predicted voltage quality gap occurs.

[0015] Furthermore, determining the compensation type in step S2 includes: If the load voltage phase is detected With reference phase If the deviation exceeds the phase tolerance threshold, the compensation type is determined to be phase compensation; If the total harmonic distortion rate of the voltage is detected to exceed the harmonic tolerance threshold, then the compensation type is determined to be harmonic compensation. If both phase compensation and harmonic compensation conditions are met, then the compensation type is determined to be composite compensation.

[0016] Furthermore, the method for generating the dynamic phase compensation signal is as follows: Calculate the required phase compensation amount ; Based on the phase compensation amount The AC / DC / AC converter unit generates a voltage with the same frequency as the grid voltage but a phase difference of [value missing]. Compensation voltage component .

[0017] Furthermore, the method for generating the adaptive harmonic injection signal is as follows: Fast Fourier Transform analysis was performed on the acquired voltage signal to identify the major harmonics. and its corresponding amplitude and phase ; For each major harmonic, a harmonic voltage component with equal amplitude and opposite phase is generated. The superposition of all harmonic voltage components constitutes the adaptive harmonic injection signal. .

[0018] The present invention has the following beneficial effects: Compared with existing technologies, this solution can independently or collaboratively adjust the three core quality dimensions of voltage amplitude, phase and harmonics simultaneously through dynamic phase compensation and adaptive harmonic injection. It not only solves the problem of voltage being too high or too low, but also corrects phase lag / lead caused by nonlinear loads and actively cancels specific subharmonics in the power grid, making the output voltage waveform closer to the ideal sine wave, and fully meeting the demand for high-quality power from precision industrial loads, data centers and other applications.

[0019] Compared with existing technologies, this solution uses a long-process compensation gap feature extraction method to analyze and identify voltage quality problems with long duration and complex characteristics, rather than just reacting to instantaneous deviations. By combining historical data learning, it can further achieve predictive compensation.

[0020] Based on power electronic converters and advanced segmented space vector modulation strategies, this invention can achieve dynamic responses at the millisecond level or even faster. The three-level topology itself can generate more output voltage levels, and combined with fine-grained control, it enables smaller voltage regulation steps and higher precision.

[0021] Compared to existing technologies, this solution employs a three-level back-to-back AC / DC / AC topology, which, compared to the traditional two-level topology, results in lower output voltage harmonics at the same switching frequency, or allows for a reduction in the switching frequency while maintaining the same harmonic requirements. This reduces switching losses and improves overall operating efficiency. Furthermore, the DC bus capacitor provides the device with a certain energy buffering capacity.

[0022] Compared with existing technologies, the adaptive harmonic injection strategy proposed in this solution can achieve harmonic mitigation function using the main power circuit without the need for additional large passive filters or separate active power filters. It achieves "one machine, multiple functions" (voltage regulation + harmonic mitigation), saving equipment investment, installation space and maintenance costs. It is particularly suitable for low-voltage distribution network scenarios with limited space and difficult transformation (such as old residential areas and industrial and commercial parks). When dealing with slow voltage changes or repetitive disturbances caused by photovoltaic power fluctuations, periodic impact loads, etc., it can take compensation measures in advance or more accurately, upgrading from "passive response" to "active defense" and "predictive optimization", enhancing the stability and foresight of regulation. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to the present invention. Figure 2 This is a schematic diagram of the low-voltage distribution network voltage flexible control system based on multi-compensation mode determination according to the present invention. Figure 3This is a schematic diagram of the low-voltage distribution network voltage flexible regulation method and system based on multi-compensation mode determination of the present invention; Figure 4 This is a comparison of the measured waveforms obtained using this method in the embodiments. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] Example 1:

[0026] See Figure 1 The flowchart shown is a method for flexible voltage regulation of low-voltage distribution networks based on multi-compensation mode determination according to the present invention. The specific implementation process of this scheme includes the following steps: Phase 1: System Initialization and Preparation Step S101: Power on and self-test of the device The control power supply is turned on, and the core controller starts up.

[0027] Run the hardware self-test program to check the status of AC / DC / AC converter power modules, voltage / current sensors, drive circuits, etc.

[0028] Initialize the DC bus capacitor pre-charge circuit to safely build up the DC bus voltage to its rated value.

[0029] Step S102: Initialize control system parameters Load default control parameters: including reference voltage amplitude (e.g., 220V), reference phase (Usually synchronized with the power grid), allowable voltage fluctuation range ±ΔV, target value for total harmonic distortion (THD), etc.

[0030] Set the sliding time window length T (e.g., adjustable from 1 to 10 power frequency cycles) and the gap exponent threshold for long-process feature extraction. .

[0031] Set the harmonic compensation range and priority (e.g., prioritize the treatment of the 5th, 7th and 11th harmonics).

[0032] Initialize the Sector-SVPWM parameter table.

[0033] Step S103: Grid connection synchronization and silent monitoring If there is a grid-connected contactor on the grid side, close it first.

[0034] The voltage acquisition unit starts up, and the phase-locked loop (PLL) begins to work, accurately tracking the phase of the grid voltage. ,frequency and amplitude .

[0035] At this time, the AC / DC / AC converter outputs zero or very low power, and the system is in a "silent monitoring" state, continuously collecting and analyzing the grid and load side voltages, but not outputting compensation.

[0036] Phase Two: Real-time Monitoring and Intelligent Decision-Making Step S201: Real-time high-speed data acquisition Simultaneously acquire three-phase voltages on the grid side at a sampling rate of no less than 10kHz. and load-side three-phase voltage .

[0037] Synchronously collect the current of key branches for protection judgment and advanced control.

[0038] Step S202: Dynamic Feature Calculation and Analysis The acquired voltage signal is subjected to coordinate transformation (such as abc / dq transformation), and the d-axis component (corresponding to amplitude deviation) and q-axis component (related to phase deviation) of the positive sequence voltage are calculated in real time.

[0039] Perform Fast Fourier Transform (FFT) or wavelet analysis on the load-side voltage to calculate the total harmonic distortion rate and the content of each harmonic in real time. .

[0040] Calculate the current real-time voltage quality index: amplitude deviation Phase deviation THD value.

[0041] Step S203: Extraction and Judgment of Long-Process Compensation Gap Features See Figure 3 The flowchart for feature extraction and judgment of long-process compensation gap is as follows: Calculate the voltage quality gap index within a continuous sliding time window T. As a quantitative indicator to measure the overall "unideal" degree of voltage within this window, where: ; Perform the judgment logic: 1) Should compensation be provided? Voltage quality gap index With preset threshold Comparison like If a significant "compensation gap" is detected in the current window, the compensation mode is triggered. Otherwise, the power quality is considered to be generally acceptable within the current window, compensation is not triggered, and the system returns to monitoring status to maintain silent monitoring.

[0042] 2) What type of compensation? Once compensation is triggered, it's necessary to clarify "what to compensate" and "how much to compensate." This is achieved by decomposing the voltage quality gap index. Implementation, specifically: Step a: Perform gap decomposition: Total Gap Index It can be decomposed into components in three independent directions, corresponding to three basic types of disturbances: Amplitude Gap Index Contributed by voltage amplitude deviation ΔV: ; Phase gap index Contributed by voltage phase deviation Δφ (calculated via q-axis voltage or direct phase difference): ; Harmonic distortion gap index The contribution from harmonic voltage content is usually represented by the square of the THD value within the window or the sum of the squares of the amplitudes of each harmonic. ; In theory that ignores cross terms, .

[0043] Step b: By comparing the amplitude gap index Phase gap index Harmonic distortion gap index Size identification of dominant components: Identification rule: The component with the largest proportion determines the main contradiction of the current gap, that is, the type of compensation.

[0044] For example: like If the largest value is significantly higher than others, it indicates an amplitude-dominant gap (such as a sustained low voltage).

[0045] like If the maximum value is reached, it is a phase-dominant gap.

[0046] like The largest value indicates a harmonic-dominated gap.

[0047] If the two components are equal and both are large, it is a composite gap (such as when voltage drop and harmonic amplification exist simultaneously).

[0048] Specifically: Amplitude / phase gap: If or Continuously exceeding the threshold, and is The main contributor.

[0049] Harmonic gap: If the THD value continues to exceed the standard and a specific harmonic is prominent.

[0050] Composite gap: The above situations occur simultaneously.

[0051] Step c: Quantization of gap parameters: Amplitude: Calculated based on the dominant components (such as average amplitude deviation, average phase deviation, and dominant harmonic amplitude).

[0052] Duration: Once window T has been triggered, the system will trace and record the number of consecutive windows that have triggered compensation, thus determining how long the gap has lasted.

[0053] Trend analysis: By comparing the gap index of the current window with the index of the previous one or several windows, it can be determined whether the gap is widening, converging, or stabilizing.

[0054] Decision output: Generates an instruction package including: [compensation enable flag, compensation type, target compensation amount].

[0055] Phase Three: Implementation of Multi-Mode Flexible Regulation Step S301: Dynamic Phase Compensation (DPC) Mode Calculation If the decision involves phase compensation, then: calculate the precise amount of phase compensation: .

[0056] Generate the fundamental compensation voltage command in a rotating coordinate system.

[0057] The fundamental compensation voltage reference signal in the three-phase stationary coordinate system is obtained by inverse Park transform. .

[0058] Step S302: Adaptive Harmonic Injection (AHI) Mode Calculation If the decision includes harmonic compensation, then: Based on the FFT analysis results, the target amplitude of the kth harmonic to be compensated is extracted. (Usually set as a negative detection value) and phase .

[0059] Adaptive algorithms such as LMS are used for real-time fine-tuning. and To minimize harmonic residuals on the load side.

[0060] Synthetic harmonic voltage injection command: .

[0061] Step S303: Composite instruction synthesis and modulation fundamental frequency compensation command Harmonic injection command Vector superposition is performed to obtain the final three-phase voltage compensation command. ; Segmented SVPWM modulator receives and DC bus voltage value; Based on the characteristics of the three-level topology, determine the large sector and small triangular region where the reference voltage vector is located; Calculate the optimal action time series of three adjacent vectors to minimize the number of switching operations and output harmonics; The corresponding PWM drive pulse signal is generated and sent to the gate driver of each IGBT / MOSFET in the AC / DC / AC converter unit.

[0062] Step S304: Power stage execution and voltage injection The converter unit operates according to the PWM drive signal, draws energy from the DC bus, and accurately generates power. Consistent compensation voltage ; The compensation voltage is injected into the target node of the low-voltage distribution network through a series or parallel connection of a transformer or reactor. The compensation voltage is superimposed on the original grid voltage, thereby achieving real-time and dynamic correction of the load-side voltage.

[0063] Example 2:

[0064] In this embodiment, the solution is further explained in conjunction with a specific application scenario: it is applied to a typical low-voltage distribution network in an urban commercial park. This network is connected to a large number of LED lights, variable frequency air conditioners, server clusters (non-linear loads), and a 500kW rooftop photovoltaic power station. During the midday peak photovoltaic power generation and evening peak load periods, the end of the network frequently experiences complex power quality problems such as high voltage, three-phase imbalance, and excessive 5th and 7th harmonics.

[0065] See Figure 2 The present invention discloses an overall structural diagram of a low-voltage distribution network voltage flexible control system based on multi-compensation mode determination, comprising: AC / DC / AC converter unit: A back-to-back converter with an ANPC (Active Neutral Point Clamping) three-level topology, rated power 150kVA, DC bus voltage support 800V. Its grid side is connected to the 0.4kV bus via a filter reactor (Lg) and contactor (KM1); the load side is connected to the target feeder via a series injection transformer (Tr, turns ratio 1:0.2).

[0066] Voltage and current acquisition unit: High-precision Hall sensors are used to acquire the three-phase voltage (Ua,b,c) and current (Ia,b,c) on the grid side (PCC point) and the load side (target feeder) respectively, with a sampling frequency of 20kHz.

[0067] Core Controller 3: Adopts a dual-core architecture of "DSP+FPGA". The DSP is responsible for core algorithm calculations, while the FPGA is responsible for high-speed data acquisition, SVPWM pulse generation, and protection logic.

[0068] Human-computer interaction interface: 7-inch touch screen, used for parameter setting, status monitoring and event logging.

[0069] The following describes the execution process of the control method using specific data: Step 1: Initialization and Monitoring (Default state after system startup) Controller loads preset parameters: V ref =220V (effective value of phase voltage), f ref =50Hz, sliding window T=10 power frequency cycles (0.2s), global trigger threshold G th =100(V 2 ).

[0070] The phase-locked loop locks the grid phase, and the system operates silently, continuously calculating G(T) for each window. Under steady-state conditions, the measured G(T) is approximately 10⁻²⁰, far below the threshold.

[0071] Step 2: Long-term compensation gap feature extraction and decision triggering Time point: One afternoon, the photovoltaic output suddenly increased, and at the same time, the air conditioning load in the park was activated.

[0072] Data: Voltage anomalies were detected in three consecutive sliding windows (within 0.6 seconds).

[0073] Window 1: V actual The amplitude increases to 235V, the 5th harmonic content increases to 8%, and G(T) is calculated to be 150.

[0074] Window 2: G(T) rises to 180.

[0075] Window 3: G(T) is 210 and the trend is upward.

[0076] Triggering and Decomposition: Because G(T) > G th The system triggers compensation.

[0077] Decompose G(T) of the third window: G v =120 (contributed by +15V amplitude deviation), G h =85 (mainly contributed by the 5th harmonic), G φ =5 (can be ignored).

[0078] Dominant Identification: G v With G h All are significant, and G vSlightly larger, classified as "amplitude-harmonic composite gap".

[0079] Output decision instructions: {Compensation enable: Yes, Type: Composite compensation, Target: ΔV=-10V, 5th harmonic suppressed to below 3%}.

[0080] Step 3: Multi-mode control execution Dynamic phase compensation module: Since the primary objective is to reduce the voltage drop, the dynamic phase compensation module calculates and generates a fundamental compensation voltage component that is out of phase with the grid voltage. V is set. c The initial amplitude is 8V (RMS), and the phase φ c It is 180° out of phase with the power grid. That is, the target generates a "negative" voltage to offset the overvoltage.

[0081] Adaptive Harmonic Injection Module: FFT analysis confirms the 5th harmonic is 125A@250Hz. The adaptive harmonic injection module starts and uses the least mean square algorithm to adaptively adjust and generate a harmonic current command with an amplitude of 125A and a phase difference of 180° from the detected 5th harmonic (achieved through equivalent control of a voltage source).

[0082] Command synthesis and modulation: The fundamental voltage command (8V, ∠180°) output by dynamic phase compensation and the harmonic current (equivalent voltage) command output by adaptive harmonic injection are synthesized in the controller. The segmented SVPWM module in the FPGA calculates the optimal switching sequence (e.g., the vector action sequence is V0(000)->V1(100)->V2(110)->…) based on the synthesized reference voltage vector and the three-level space vector diagram, and generates PWM pulses to drive the ANPC converter.

[0083] Step 4: Closed-loop feedback and optimization After the converter outputs the compensation voltage / current, the load-side voltage drops from 235V to 222V within approximately 20ms, and the 5th harmonic content decreases to 2.5%.

[0084] The voltage acquisition unit will acquire the new V out Feedback is sent back to the controller. The closed-loop PI regulator fine-tunes V based on the residual error (ΔV = 220 - 222 = -2V). c The amplitude command is set to 9V.

[0085] Meanwhile, the least mean square algorithm continues to fine-tune the phase of the harmonic injection, further reducing the 5th harmonic to 2.0%.

[0086] Protection and monitoring: Throughout the entire process, the DC bus voltage remained stable at 800±10V, the bridge arm current did not exceed the limit, and the system operated normally.

[0087] Step 5: Application of Predictive Compensation After the system had been running for a week, historical data showed that during the period from 13:30 to 14:00 each day, due to the pattern of photovoltaic output and load, G(T) had an 80% probability of exceeding G. th And the gaps are mainly of excessively high amplitude.

[0088] The controller's lightweight learning module (such as one based on simple linear regression) identifies the pattern.

[0089] Predictive action: At 13:25 the following day, the system entered the "preparatory compensation" mode 5 minutes in advance, preset a small negative bias in the PI controller output of the dynamic phase compensation module. When the voltage began to show signs of rising at 13:28, the system began to output the compensation voltage with almost no delay, achieving "disturbance prevention" and making the voltage curve exceptionally smooth during that period, avoiding any obvious voltage overshoot.

[0090] See Figure 4 The measured waveform comparison diagram of this embodiment is as follows: The test conditions are as follows: Scenario: Simulating the midday operating conditions of a commercial park, with fluctuating photovoltaic output and nonlinear loads (frequency converters, rectifiers) being put into operation simultaneously.

[0091] Monitoring point: Load-side bus (target feeder).

[0092] Triggering condition: The system detects an "amplitude-harmonic composite gap" and automatically triggers compensation.

[0093] See Figure 4 a. Instantaneous voltage waveform before compensation: Before compensation: The voltage amplitude is high and fluctuates, with a peak value exceeding 300V and an effective value of approximately 235V.

[0094] The waveform is severely distorted, with a flat top and obvious "cropping" phenomenon, which is a typical characteristic of odd harmonics (especially the 3rd, 5th, and 7th harmonics) generated by nonlinear loads.

[0095] The waveform deviates significantly from the reference sine wave (blue dashed line); See Figure 4 Harmonic spectrum comparison sub-plots: Before compensation: The fifth harmonic content was the highest, reaching 8.2%.

[0096] The 7th harmonic is the second most common, at approximately 4.5%.

[0097] The third harmonic is also present, at approximately 2.1%.

[0098] The total harmonic distortion (THD) was calculated to be 9.3%, far exceeding the national standard (5%) limit.

[0099] See Figure 4 b. Instantaneous voltage waveform after compensation: After compensation: The voltage amplitude stabilized within the range of 220V±2%, with a peak value of approximately 311V, which is consistent with the theoretical value.

[0100] The waveform is smooth, a perfect sine wave, and almost completely overlaps with the reference sine wave (blue dashed line).

[0101] The original "crown" distortion has completely disappeared.

[0102] See Figure 4 Harmonic spectrum comparison sub-plots: After compensation: The 5th harmonic was suppressed to below 0.8%.

[0103] The 7th harmonic was suppressed to below 0.5%.

[0104] The content of other harmonics is all below 0.5%.

[0105] The total harmonic distortion (THD) was reduced to 1.2%, which is far superior to the national standard (GB / T12325).

[0106] See Figure 4 The dynamic response process sub-diagram (during voltage surge) is shown below: Dynamic process description: At time t0: A sudden voltage surge occurs in the grid (such as a sudden increase in photovoltaic reverse power), and the load-side voltage rises from 220V to 235V within one cycle.

[0107] Within t0+20ms: The system of this invention completes the entire process of detection, gap feature extraction, and decision-making.

[0108] Within t0+50ms: the compensation voltage is fully established, and the load-side voltage is pulled back to the allowable band of 220V±2%.

[0109] Steady-state stability: The voltage remains highly stable during the duration of the disturbance.

[0110] After the disturbance disappears: the system smoothly exits the compensation without overshoot or oscillation.

[0111] See the following key performance indicator comparison table:

[0112] The above demonstrates the effectiveness and superiority of the technical solution of this invention, mainly reflected in: Comprehensive governance capabilities: It can simultaneously and quickly resolve complex issues such as voltage amplitude deviation and waveform distortion (harmonics).

[0113] Excellent dynamic performance: fast response time (<50ms), smooth adjustment process, and no overshoot.

[0114] High precision in power management: It can improve key power quality indicators (such as THD and voltage deviation) from "borderline acceptable" to "excellent" levels, significantly exceeding national standards.

[0115] Practical value of the project: This solution achieves "one machine with multiple functions", replacing the traditional "voltage regulator + filter" combination with a single device, which reduces investment and operation and maintenance costs while ensuring power supply quality.

[0116] To further verify the effectiveness of this embodiment, a 24-hour comparative test was conducted under the same operating conditions with a traditional DVR device based on amplitude adjustment. The test results are shown in the table below:

[0117] The above results show that the solution of the present invention is significantly better than the traditional solution in terms of overall voltage quality improvement, response speed and operating efficiency.

[0118] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for flexible voltage regulation of low-voltage distribution networks based on multi-compensation mode determination, characterized in that, include: S1: Real-time acquisition of voltage signals from low-voltage distribution networks, and extraction of voltage amplitude, phase, and harmonic component information; S2: Extract long-process compensation gap features based on sliding time window, determine whether to enter compensation mode and determine compensation type; S3: If compensation is required, a dynamic phase compensation signal or an adaptive harmonic injection signal will be generated based on the type of compensation. S4: Based on the dynamic phase compensation signal or adaptive harmonic injection signal, generate a composite control command to control the output of the corresponding compensation voltage to the power grid, thereby realizing three-dimensional flexible control of the amplitude, phase and harmonics of the power grid voltage.

2. The low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to claim 1, characterized in that, In step S2, the extraction of long-process compensation gap features includes: Set sliding time window The voltage quality gap index within this window is calculated using the following formula. : ,in, for Reference voltage at time, for The measured voltage at that moment; The voltage quality gap index With preset threshold If a comparison is made, If so, then the current period is determined to be a gap period that needs to be compensated; Further identification of the gap type as at least one of amplitude gap, phase gap, or harmonic gap.

3. The low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to claim 1 or 2, characterized in that, The method for determining the compensation type includes: If the load voltage phase is detected With reference phase If the deviation exceeds the phase tolerance threshold, the compensation type is determined to be phase compensation; If the total harmonic distortion rate of the voltage is detected to exceed the harmonic tolerance threshold, then the compensation type is determined to be harmonic compensation. If both phase compensation and harmonic compensation conditions are met, then the compensation type is determined to be composite compensation.

4. The low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to claim 1, characterized in that, In step S3, the method for generating the dynamic phase compensation signal is as follows: Calculate the required phase compensation amount ; Based on the phase compensation amount The AC / DC / AC converter unit generates a voltage with the same frequency as the grid voltage but a phase difference of [value missing]. Compensation voltage component .

5. The low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to claim 1, characterized in that, In step S3, the method for generating the adaptive harmonic injection signal is as follows: Fast Fourier Transform analysis was performed on the acquired voltage signal to identify the major harmonics. and its corresponding amplitude and phase ; For each major harmonic, a harmonic voltage component with equal amplitude and opposite phase is generated. The superposition of all harmonic voltage components constitutes the adaptive harmonic injection signal. .

6. The low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to claim 1, characterized in that, In step S4, the generation of the composite control command specifically includes: When a phase gap is identified, the dynamic phase compensation mode is activated to calculate the target phase compensation amount. And generate a phase offset Fundamental compensation voltage ,in, , As a reference phase, For the load voltage phase, For voltage phase, The fundamental angular frequency of the grid voltage. To compensate for the voltage amplitude; When a harmonic gap is identified, the adaptive harmonic injection mode is activated, and the main harmonic components are extracted through real-time harmonic analysis. And generate reverse-injected harmonic voltage , For harmonic order, For the first The amplitude of the subharmonic.

7. The low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to claim 1, characterized in that, The method further includes the following steps: Monitor the voltage quality of the compensated power grid; Calculate the error between the compensated voltage and the reference voltage. ,in The compensated grid voltage; According to the error The parameters of the dynamic phase compensation signal or adaptive harmonic injection signal are dynamically adjusted to achieve closed-loop optimization control.

8. The low-voltage distribution network voltage flexible control method based on multi-compensation mode determination according to claim 1, characterized in that, The method further includes the following steps: Record gap characteristic data and control parameters during the historical compensation process; The acquired data is analyzed and trained using machine learning models to predict the trend of voltage quality changes in a specific future period. The compensation strategy is triggered and adjusted in advance before the predicted voltage quality gap occurs.

9. A low-voltage distribution network voltage flexible control system based on multi-compensation mode determination, used to implement the method described in any one of claims 1-8, characterized in that, Includes a voltage acquisition and synchronization unit, a core controller, and an AC / DC / AC converter unit; The AC / DC / AC converter unit adopts a three-level three-phase half-bridge back-to-back topology, with its grid side connected to the low-voltage distribution network and its load side connected to the controlled load. The voltage acquisition and synchronization unit is used to acquire grid voltage and load voltage signals in real time, and extract voltage amplitude, phase and frequency information; The core controller is connected to the voltage acquisition and synchronization unit and the AC / DC / AC converter unit. It is used to perform long-process compensation gap feature extraction and generate a composite control command that integrates dynamic phase compensation and adaptive harmonic injection based on the extraction result, so as to drive the AC / DC / AC converter unit to output compensation voltage.

10. The low-voltage distribution network voltage flexible control system based on multi-compensation mode determination according to claim 9, characterized in that, The core controller also includes a modulation module for adjusting the fundamental compensation voltage. and the reverse injected harmonic voltage The signals are superimposed and a segmented space vector modulation strategy is used to generate pulse width modulation signals to drive the power switching transistors of the three-level three-phase half-bridge back-to-back topology.

Images