A reactive power compensation controller for low voltage management at the end of a power grid and a control method thereof

Abstract

The application discloses a kind of for grid end low voltage management reactive power compensation controller and its control method, belong to power system electric energy quality management technical field.The controller includes plastic shell upper cover, plastic shell lower cover, main control board, power module, sampling module, man-machine interface module, switching control interface and remote communication module;Voltage-reactive power collaborative control algorithm is embedded in main control board, according to end voltage real-time value and preset threshold automatic switching "voltage priority" or "power factor priority" control mode;The sampling module supports three-phase voltage, current synchronous sampling, and automatically identifies wiring mode.The application is through voltage priority control strategy, when end voltage drops, capacitor is quickly put into, effectively lifts voltage, simultaneously has remote operation and maintenance, capacitor life prediction and modularization convenient maintenance function, significantly improves distribution network end power supply quality and operation and maintenance efficiency.

Description

Technical Field

[0001] This invention relates to the field of power system power quality management technology, specifically to a reactive power compensation controller and its control method for low voltage management at the end of the power grid. Background Technology

[0003] Existing low-voltage reactive power compensation controllers are mostly installed on the distribution transformer side and generally adopt a single control strategy of "power factor priority," only activating capacitors when the power factor is below a threshold. In the case of distribution network end scenarios, there is often a situation where "the end voltage is seriously low, but the power factor on the transformer side is still qualified." In this case, traditional controllers will not operate and cannot effectively raise the end voltage, resulting in extremely poor mitigation effect.

[0004] While some existing technologies offer dual-mode control for voltage and power factor, they still suffer from interlocking technical flaws, and localized optimization of a single module cannot fundamentally solve the problem. Since the line impedance cannot be identified online, the compensation capacity can only be set based on experience, which easily leads to overcompensation and undercompensation, causing the voltage to fluctuate repeatedly around the qualified threshold. In order to suppress the fluctuation, the switching interval can only be increased and the hysteresis can be widened, which further reduces the voltage response speed and aggravates the duration of low voltage at the end, forming a vicious cycle of poor accuracy → large fluctuation → slow response.

[0005] The dual-mode switching uses a single fixed threshold and fixed hysteresis, which cannot adapt to the characteristics of severe end load fluctuations. In scenarios with frequent voltage fluctuations, it is easy to switch the working mode frequently, resulting in frequent switching of capacitors. Furthermore, the crude switching sequence strategy cannot balance equipment losses, further accelerating the degradation of capacitor capacity and lifespan.

[0006] The actual capacitance decay and lifespan loss of capacitors cannot be fed back to the control logic, and the control strategy always operates based on ideal parameters. As the equipment ages, the compensation accuracy and treatment effect continue to decline. At the same time, islanded operation cannot use big data of the distribution area to optimize parameters, resulting in poor adaptability and high manual maintenance costs. Summary of the Invention

[0007] The purpose of this invention is to provide a reactive power compensation controller and its control method for low voltage management at the end of a power grid, which solves the problems pointed out in the background art.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: a reactive power compensation controller for low voltage management at the end of the power grid, comprising a modular housing, a main control unit, a sampling unit, a switching control interface, a remote communication unit, and a power supply unit; The main control unit has a built-in online line impedance identification module, a voltage graded priority control module, a power factor optimization control module, and a mode smooth switching module. The line impedance online identification module is used to identify the equivalent impedance parameters of the power supply lines at the end of the distribution network based on real-time collected voltage and current data. The voltage grading priority control module is used to match the corresponding grading compensation strategy according to the drop level of the three-phase lowest voltage. Under the undervoltage state, the power factor is shielded and the capacitor is put into operation first to raise the terminal voltage. The power factor optimization control module is used to switch capacitors on and off according to reactive power deficit under qualified voltage conditions to maintain the power factor in compliance with the standard. The mode smoothing switching module is used to perform hysteresis-free switching between voltage priority control mode and power factor priority control mode, suppressing switching oscillations near the threshold.

[0009] Preferably, the online line impedance identification module incorporates a dual identification mechanism: one is a switching disturbance calibration unit, used to utilize the changes in bus voltage and reactive power before and after capacitor switching, and the system rated voltage satisfies the formula: X=ΔU·U N / ΔQ; Calculate the equivalent reactance X of the line, where ΔU is the voltage change, ΔQ is the reactive power change, and U N The first is the rated phase voltage of the system; the second is the recursive least squares correction unit, which is used to iteratively update the equivalent resistance and equivalent reactance parameters of the line under steady-state conditions based on the discrete model of line voltage drop.

[0010] Preferably, the voltage grading priority control module has two built-in voltage thresholds, dividing the voltage state into three levels: severe undervoltage, mild undervoltage, and qualified voltage. Under severe undervoltage conditions, based on the identified equivalent reactance X, the voltage is controlled using formula Q. need =(U th2 -U min )·U N / X calculates the target compensation capacity Q need Batch deployment of common compensation capacitor banks of corresponding capacity, of which U th2 For the qualified voltage threshold, U min This is the lowest three-phase voltage; under slight undervoltage conditions, step compensation is performed according to the single-group sequential connection strategy until the voltage rises back to the qualified range.

[0011] Preferably, the main control unit also has a built-in capacitor life prediction and equalization switching module, which is used to calculate the remaining life coefficient of each capacitor bank by weighted summation based on three dimensions: the number of capacitor switching operations, the operating environment temperature, and the actual capacity decay rate; when performing switching operations, the combination with the highest average remaining life coefficient is selected first among the capacitor combinations that meet the capacity requirements.

[0012] Preferably, the modular housing includes an upper main control compartment and a lower base compartment, which are detachably connected to the positioning post by a buckle; the lower base compartment has a built-in pluggable functional slot, and the sampling unit, power supply unit and remote communication unit all adopt an independent plug-in structure and are electrically connected to the main control unit through the slot; the back of the lower base compartment integrates a standard DIN rail slot.

[0013] Preferably, the remote communication unit has a built-in Dianhong IoT operating system adaptation layer, supports 4G / 5G / NB-IoT wireless communication, can automatically report the device model to the power distribution IoT cloud platform and receive dynamic control parameters, and realize cloud-edge collaborative control threshold self-tuning; the sampling unit supports automatic identification of three-phase four-wire, three-phase three-wire and single-phase two-wire wiring methods, and uses a 16-bit or higher precision ADC to realize synchronous sampling of three-phase voltage and three-phase current.

[0014] Preferably, a reactive power compensation control method for low voltage mitigation at the end of a power grid includes the following steps: S1. Read the default control parameters and collect three-phase voltage and three-phase current data in real time; S2. Identify the equivalent impedance parameters of the end lines of the distribution network online and calculate the reactive power compensation capacity required for voltage rise under the current operating conditions. S3. Extract the lowest three-phase voltage and determine the voltage drop level; S4. If the voltage is undervoltage, enter the voltage grade priority control mode, match the corresponding grade strategy to put the capacitor into the field, and prioritize the increase of the terminal voltage. S5. If the voltage recovers to the qualified range, the system will smoothly switch to the power factor priority control mode through the hysteresis transition mechanism, and optimize the switching of capacitors according to the reactive power deficit. S6. Periodically update the running data and upload it to the cloud platform, receive the setting parameters issued by the cloud platform, and dynamically update the control threshold.

[0015] Preferably, in step S3, the voltage drop levels are divided into three levels: severe undervoltage, mild undervoltage, and voltage qualified. In step S4, when there is a severe undervoltage, the target compensation capacity is calculated based on the equivalent impedance, and capacitor banks of the corresponding capacity are put into operation in batches; when there is a slight undervoltage, capacitors are put into operation one group at a time according to the single-group cyclic switching strategy, and the voltage is detected again after a fixed interval after each group is put into operation; the power factor judgment logic is shielded throughout the undervoltage state.

[0016] Preferably, the smooth mode switching in step S5 adopts a dynamic hysteresis mechanism. The specific process is as follows: a dynamically adjustable hysteresis width is set based on the qualified voltage threshold to form a dual switching threshold for rising and falling edges; when the voltage rises from the undervoltage range to the hysteresis range, the addition of new capacitors is stopped, and the current compensation capacity is maintained; when the voltage continues to be higher than the qualified threshold and exceeds the set stable time, redundant capacitors are gradually removed, and the power factor detection logic is started to smoothly transition to the power factor priority control mode.

[0017] Preferably, the method further includes a capacitor remaining life calculation and online capacity monitoring step: calculating the remaining life coefficient of each capacitor based on the number of switching operations, operating temperature, and capacity decay rate, and prioritizing the selection of capacitor groups with higher remaining life coefficients during switching operations; detecting the current of each capacitor branch in real time and calculating the actual output reactive power, and generating an alarm and shielding the faulty capacitor group when the capacity decay exceeds a set threshold.

[0018] Preferably, the equivalent impedance identification of the line in step S2 adopts a dual-mechanism combination: during initial operation or periodic calibration, switching disturbance calibration is used: a capacitor of known capacity is connected, and the equivalent reactance of the line is calculated based on the voltage change and reactive power change before and after the switching; during continuous daily operation, a recursive least squares method with a forgetting factor is used to iteratively correct the impedance parameters based on the discrete model of line voltage drop, and parameter updates are only performed when the system is in steady state. The switching disturbance calibration is specifically as follows: the equivalent reactance is calculated using the reactive power disturbance generated by the switching of a capacitor of known capacity: the effective value of the steady-state bus phase voltage U0 and the reactive power of the load Q0 are recorded before the switching; a set of capacitors with a rated capacity of ∆Q is connected. C The common capacitor is used to wait for the voltage to stabilize after 5 power frequency cycles; the effective value of the steady-state bus phase voltage U1 and the total reactive power Q1 of the system are recorded after switching; the voltage change ΔU=U1-U0 and the reactive power change ΔQ=Q1-Q0 are calculated; the reactance of the line at the end of the distribution network is much larger than the resistance, and the active voltage drop can be ignored, which can be calculated by ΔU=ΔQ·X / U N Derive the equivalent reactance of the line; X = ΔU·U N / ΔQ; where UN is the system's rated phase voltage. Identify the equivalent resistance R by switching two sets of capacitors with different capacities, and establish a system of two linear equations to solve for R and X. Use the recursive least squares method to establish a discrete model of the line voltage drop: U(k) = U N -(Pk·R+Qk·X) / U N ); R and X are iteratively updated every 1 second using a recursive least squares algorithm with a forgetting factor. Parameter updates are only performed when the mode smoothing switching module determines that the system is in a steady state. Iteration is paused during the transient process to avoid interference from fluctuating data with identification accuracy.

[0019] The hysteresis width is dynamically adjusted: the active power fluctuation rate of the load is calculated in real time, and the hysteresis width and mode stability determination time are adjusted according to the fluctuation rate. The larger the load fluctuation rate, the larger the hysteresis width and stability determination time, thus suppressing frequent switching in the case of severe fluctuations.

[0020] The formula for calculating the remaining life factor of a capacitor is: L = w1·L s +w2·L t +w3·L c Where L is the remaining lifetime coefficient, w1, w2, and w3 are weighting coefficients, w1=0.4, w2=0.3, w3=0.3, and L s For the switching life component, L s =N a / N r， Where N a To accumulate the number of throws, N r The rated switching count of the capacitor, L t For temperature aging components, L c For the capacity decay component, L c =1-δ / δ li δ is the current capacity decay rate, δ li The failure threshold is set to 20% by default. The temperature aging component is calculated based on the Arrhenius aging model and the equivalent aging time is converted from the cumulative operating temperature.

[0021] The online monitoring of capacitor capacity is as follows: after the capacitor is put into stable operation, the actual output reactive capacity is calculated by the effective value of the branch current, and then the capacity decay rate is calculated; when the capacity decay rate exceeds the first threshold, the switching priority of the corresponding capacitor is reduced; when the capacity decay rate exceeds the second threshold, a serious alarm is generated and the capacitor is removed from the switching queue.

[0022] In step S6, the cloud-edge collaborative parameter self-tuning is specifically as follows: the cloud platform, based on the historical operating data of the transformer area, takes maximizing the voltage qualification rate and minimizing the number of switching operations as dual optimization objectives, obtains the optimal control parameters and issues them; after receiving the parameters, the local controller first performs boundary legality verification, and after the verification passes, the updated parameters take effect, while retaining the original parameter backup to support abnormal rollback.

[0023] The beneficial effects of this invention are as follows: The online line impedance identification module improves impedance identification accuracy through a dual-mechanism combination. Compared with empirical fixed capacity switching, it improves the matching degree of compensation capacity, reducing over-compensation and under-compensation problems from the source. The voltage graded priority control module can match the response speed according to the degree of voltage drop. In case of severe undervoltage, it adopts a batch connection strategy, shortening the voltage recovery time.

[0024] Impedance identification improves compensation accuracy from the source, reducing voltage fluctuations caused by over-compensation and under-compensation; graded control ensures response efficiency under different drop levels; and smooth switching further suppresses mode boundary oscillations through dynamic hysteresis. The synergistic effect of these three mechanisms reduces the occurrence of switching oscillations, exceeding the effect of a single smooth switching module. Simultaneously, stable operating data feeds back into impedance identification accuracy, forming a positive cycle and improving the end-voltage qualification rate.

[0025] In scenarios requiring rapid batch switching due to severe undervoltage, prioritizing capacitor banks based on lifespan ensures both a fast voltage rise response and avoids excessive wear on individual capacitor banks caused by fixed-sequence switching. Without sacrificing voltage regulation speed, the overall lifespan of the capacitors is extended, achieving a simultaneous improvement in both governance effectiveness and equipment economy. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the overall structure of the reactive power compensator of the present invention; Figure 2 This is an overall flowchart of the reactive power compensation control method described in this invention; Figure 3 This is a logical diagram illustrating the voltage grading and smooth mode switching of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0028] Embodiments of the present invention: like Figure 1As shown, this invention provides a reactive power compensation controller for low voltage management at the end of a power grid. It adopts a modular, split-type housing, including an upper main control compartment 1 and a lower base compartment 2, which are detachably connected via positioning posts and a snap-fit ​​structure. Internally, they are electrically interconnected through board-to-board connectors, supporting rapid on-site disassembly and maintenance. The modular hardware architecture, in conjunction with the lifespan monitoring and fault alarm algorithms of the main control unit 3, enables rapid location and replacement of faulty modules without requiring the entire unit to be scrapped. The main control unit 3 and the human-machine interface module 6 are fixed inside the upper main control compartment 1. The main control unit 3 uses an AT32F455VE 32-bit ARM Cortex-M4 microprocessor with a main frequency of 240MHz, built-in 512KB Flash and 128KB SRAM, and runs on the FreeRTOS embedded real-time operating system, integrating all core control algorithms and inter-module collaborative logic. The human-machine interaction module 6 uses a 128×64 dot matrix LCD display and four waterproof buttons. It can display information such as three-phase voltage, current, power factor, switching status, control mode, voltage qualification rate statistics, and capacitor remaining life in real time, and supports local parameter settings.

[0029] The lower base compartment 2 serves as the functional support base, with three sets of standardized slots on its inner wall, corresponding to the power supply unit 4, sampling unit 5, and remote communication unit 8, respectively. Each unit is an independent plug-in design, supporting hot-swappable replacement, allowing module replacement without disassembling the main control unit. Power supply unit 4: Employs an isolated switching power supply solution, with a wide input voltage range of AC85-265V and outputs +5V, +3.3V, and multiple isolated power supplies, providing power to the main control, sampling, and communication modules. It features overvoltage, overcurrent, and reverse connection protection. Sampling unit 5: Built-in TV1013 voltage transformer and TA1225 current transformer, paired with an AD7606 16-bit 8-channel synchronous ADC, communicating with the main control unit 3 via SPI bus; supports automatic identification of three-phase four-wire and single-phase two-wire wiring methods, with a sampling frequency of 10kHz and voltage measurement accuracy ≤0.2%, providing a unified high-precision data source for all modules such as impedance identification, voltage classification, and lifespan monitoring. Remote communication unit 8: Utilizing the Hezhou Air724UG4GCat.1 module, it incorporates a lightweight adaptation layer for the Dianhong IoT operating system and an MQTT protocol stack. It connects bidirectionally to the main control unit 3 via a UART interface; it also integrates a Bluetooth 5.0 module, supporting near-field debugging. Remote communication unit 8 and all control modules within the main control unit 3 work in tandem, uploading full operational data and receiving cloud-based optimization parameters, synchronizing them to each control module. The bottom of the lower cover base compartment 2 features a pluggable terminal block 9 with spring terminals for external voltage, current signals, and power supply input. The back of the lower cover integrates a 35mm standard DIN rail slot 10 and wall-mounting holes, allowing direct mounting on distribution box rails, adapting to various distribution box installation scenarios at the end of rural power grids. The switching control interface 7 is located on the side of the housing, including two independent RS485 interfaces using RJ45 connectors. A single bus can connect up to 32 smart capacitors, supporting mixed networking of common and separate compensation, and issuing switching commands via the Modbus-RTU protocol.

[0030] like Figure 2 , Figure 3 As shown, a reactive power compensation control method for low voltage mitigation at the end of a power grid includes the following steps: S1. Read the default control parameters and collect three-phase voltage and three-phase current data in real time; S2. Identify the equivalent impedance parameters of the end lines of the distribution network online and calculate the reactive power compensation capacity required for voltage rise under the current operating conditions. S3. Extract the lowest three-phase voltage and determine the voltage drop level; S4. If the voltage is undervoltage, enter the voltage grade priority control mode, match the corresponding grade strategy to put the capacitor into the field, and prioritize the increase of the terminal voltage. S5. If the voltage recovers to the qualified range, the system will smoothly switch to the power factor priority control mode through the hysteresis transition mechanism, and optimize the switching of capacitors according to the reactive power deficit. S6. Periodically update the running data and upload it to the cloud platform, receive the setting parameters issued by the cloud platform, and dynamically update the control threshold.

[0031] In step S3, the voltage drop levels are divided into three levels: severe undervoltage, mild undervoltage, and voltage qualified.

[0032] In step S4, when there is a severe undervoltage, the target compensation capacity is calculated based on the equivalent impedance, and capacitor banks of the corresponding capacity are put into operation in batches; when there is a slight undervoltage, capacitors are put into operation one group at a time according to the single-group cyclic switching strategy, and the voltage is detected again after a fixed interval after each group is put into operation; the power factor judgment logic is shielded throughout the undervoltage state.

[0033] In step S5, the smooth mode switching adopts a dynamic hysteresis mechanism. The specific process is as follows: a dynamically adjustable hysteresis width is set based on the qualified voltage threshold to form a dual switching threshold for rising and falling edges; when the voltage rises from the undervoltage range to the hysteresis range, the addition of new capacitors is stopped, and the current compensation capacity is maintained; when the voltage continues to be higher than the qualified threshold and exceeds the set stable time, redundant capacitors are gradually removed, and the power factor detection logic is started at the same time to smoothly transition to the power factor priority control mode.

[0034] The method also includes capacitor remaining life calculation and online capacity monitoring steps: the remaining life coefficient of each capacitor is calculated based on the number of switching, operating temperature and capacity decay rate, and the capacitor group with higher remaining life coefficient is selected first when switching; the current of each capacitor branch is detected in real time and the actual output reactive power is calculated. When the capacity decay exceeds the set threshold, an alarm is generated and the faulty capacitor group is shielded.

[0035] Step S2, the identification of the line equivalent impedance, employs a dual-mechanism approach: During initial operation or periodic calibration, a switching disturbance calibration is used: a capacitor of known capacity is switched in, and the line equivalent reactance is calculated based on the voltage and reactive power changes before and after switching. During continuous daily operation, a recursive least squares method with a forgetting factor is used to iteratively correct the impedance parameters based on a discrete model of line voltage drop, and parameter updates are only performed during system steady-state operation. Specifically, the switching disturbance calibration involves calculating the equivalent reactance using the reactive power disturbance generated by switching a capacitor of known capacity: recording the effective value of the steady-state bus phase voltage U0 and the load reactive power Q0 before switching; switching in a capacitor with a rated capacity of ∆Q... C The common capacitor is used to wait for the voltage to stabilize after 5 power frequency cycles; the effective value of the steady-state bus phase voltage U1 and the total reactive power Q1 of the system are recorded after switching; the voltage change ΔU=U1-U0 and the reactive power change ΔQ=Q1-Q0 are calculated; the reactance of the line at the end of the distribution network is much larger than the resistance, and the active voltage drop can be ignored, which can be calculated by ΔU=ΔQ·X / U N Derive the equivalent reactance of the line; X = ΔU·U N / ΔQ; where UN is the system's rated phase voltage. Identify the equivalent resistance R by switching two sets of capacitors with different capacities, and establish a system of two linear equations to solve for R and X. Use the recursive least squares method to establish a discrete model of the line voltage drop: U(k) = U N -(P(k)·R+Q(k)·X) / U N ); k represents the k-th discrete sampling time (the sampling period in this scheme is 1s), U(k) represents the measured effective value of the phase voltage at the end of the distribution network at the k-th time, U N The system's rated phase voltage is given by P(k), the active power transmitted by the line at time k, and the reactive power transmitted by the line at time k. The equivalent resistance R and equivalent reactance X of the power supply line are iteratively updated using a recursive least squares algorithm with a forgetting factor. Parameter updates are only performed when the mode smoothing switching module determines that the system is in a steady state. Iteration is paused during transient processes to avoid interference from fluctuating data with identification accuracy.

[0036] The hysteresis width is dynamically adjusted: the active power fluctuation rate of the load is calculated in real time, and the hysteresis width and mode stability determination time are adjusted according to the fluctuation rate. The larger the load fluctuation rate, the larger the hysteresis width and stability determination time, thus suppressing frequent switching in the case of severe fluctuations.

[0037] The voltage sag classification and judgment logic in step S3 satisfies the following: real-time calculation of the effective value U of the three-phase voltage. a U b U c Take the minimum value of the three phases U min =min(U a U b U c ); Set two threshold levels and three status levels: Severe undervoltage: U min th1 Mild undervoltage: U th1 ≤U min th2 Voltage qualified: U min ≥U th2。 The method for calculating and selecting the graded compensation capacity in step S4 is as follows: When there is a severe undervoltage, the lowest phase voltage is raised to U. th2 To achieve the target, calculate the required total reactive power compensation capacity: Q need =(U th2 -U min )U N / X; where X is the real-time identified equivalent line reactance. In the available capacitor bank, all common-complementary capacitor combinations are traversed, and the combination with the total capacitance closest to Q is selected. need ​​The combination of capacitors is used; among multiple combinations with similar capacities, the combination with the highest average remaining life factor is selected first, and they are put into operation in batches at once, with the shortest allowable switching interval of 10 seconds. In this mode, the power factor determination logic and overcompensation determination logic are disabled, and voltage recovery is the sole control objective. When there is a slight undervoltage, one set of common compensation capacitors with the smallest rated capacity is put into operation at a time, and after each set is put into operation, a switching interval T is waited. in (Default 30s) Check the voltage again until U min ≥U th2 Or there are no available capacitors.

[0038] The dynamic hysteresis algorithm for smooth mode switching in step S5 is as follows: using the qualified voltage threshold U th2 Based on this, a dynamically adjustable hysteresis width ΔU is set. h This forms a bidirectional switching threshold: rising edge switching point (undervoltage → qualified): U on =U th2 Falling edge switching point (qualified → undervoltage): U off =U th2 -ΔU h The voltage is at [U] off U on When within the specified range, maintain the current operating mode to avoid frequent boundary switching.

[0039] The dynamic adjustment method for hysteresis width is as follows: First, calculate the active power fluctuation rate of the load in real time to satisfy: ∆P r =|P(k)-P(k-1)| / P(k-1)∆T×100%; where ∆T is the calculation period (default 1s). Adjust the hysteresis width and stability determination time according to the fluctuation rate classification: smooth fluctuations satisfy: (∆P r <5% / s): ΔU h =6V, stable judgment time 30s; moderate fluctuations satisfy: (5% / s≤∆P) r <15% / s): ΔU h =8V, stability check time 60s; violent fluctuations: (∆P) r ≥15% / s): ΔU h =12V, stability determination time 90s. In the formula, P(k) is the active power collected in the current second, P(k-1) is the active power collected in the previous second, |P(k)-P(k-1)| is the absolute value of the change in active power between the two seconds (only considering the magnitude of the change, not distinguishing between increase / decrease), ∆T is the power collection period, fixed at 1 second; ∆P r This represents the percentage fluctuation of active power per second, expressed as % / s.

[0040] Secondly, the transition from voltage priority mode to power factor mode is performed in three steps to avoid impact: Locking phase: After the voltage rises and enters the hysteresis range, all switching operations are immediately locked, with no additional switching or disconnection, maintaining the current compensation capacity; Stability determination phase: The voltage remains higher than U on After reaching the set stable duration, the state is determined to be stable; smooth transition phase: redundant capacitors are removed group by group, with an interval of T between each group. in During the cut-off process, the power factor is monitored in real time. When the power factor drops to the input threshold PFon, the cut-off is stopped immediately, and the system officially enters the power factor priority control mode.

[0041] The formula for calculating the remaining life factor of a capacitor is: L = w1·L s +w2·L t +w3·L c Where L is the remaining lifetime coefficient, w1, w2, and w3 are weighting coefficients, w1=0.4, w2=0.3, w3=0.3, and L s For the switching life component, L s =N a / N r， Where N a To accumulate the number of throws, N r L is the rated switching number of the capacitor. t For temperature aging components, L c For the capacity decay component, L c =1-δ / δ li δ is the current capacity decay rate, δ li The failure threshold is set to 20% by default. The temperature aging component is calculated based on the Arrhenius aging model and the equivalent aging time is converted from the cumulative operating temperature.

[0042] The online monitoring of capacitor capacity specifically involves: after the capacitor is put into operation and stabilizes, calculating the actual output reactive power capacity using the effective value of the branch current, and then calculating the capacity decay rate; when the capacity decay rate exceeds a first threshold, reducing the switching priority of the corresponding capacitor; when the capacity decay rate exceeds a second threshold, generating a critical alarm and removing the capacitor from the switching queue. After the capacitor is put into operation and stabilizes, the effective value of the sampled branch current I is used to... C Calculate the actual output reactive power capacity Q a =3U p I C The capacity decay rate is δ=(Q r -Q a ) / Q r ×100%; where Q is... a The value represents the actual three-phase total reactive power output of the capacitor, expressed in var (var). A lower value indicates more severe capacitor aging and degradation. 3 represents the calculation of the total three-phase capacity of the corresponding three-phase common compensation capacitor, U.p I represents the effective value of the phase voltage of the power grid, i.e., the actual operating phase voltage across the capacitor; C This represents the effective current value of the capacitor branch, directly acquired by the controller's sampling module. δ represents the capacitance decay rate; a larger value indicates more severe aging. Q r This refers to the factory-rated reactive power of the capacitor bank; Q r -Q a This represents the absolute value of capacity decay, i.e., how much capacity has been lost compared to the rated value. When δ > 10%, an early warning is generated, and the switching priority of the capacitor is reduced. Within 10% is within the normal natural aging range of the capacitor and does not affect normal use. If it exceeds 10%, it indicates that aging has occurred. By reducing the switching priority, the number of times the capacitor is operated is reduced, further deterioration is delayed, and an early warning message is pushed to the cloud to remind maintenance personnel to pay attention.

[0043] When δ > 20%, a critical alarm is generated, the capacitor is automatically removed from the switching queue, and the switching action is blocked. The power industry typically uses a capacity degradation exceeding 20% ​​as the failure criterion for low-voltage power capacitors: at this point, the compensation effect drops significantly, and continued operation not only fails to meet the reactive power compensation / voltage rise requirements, but may also lead to safety risks such as breakdown and bulging due to internal dielectric deterioration.

[0044] In step S6, the cloud-edge collaborative parameter self-tuning is specifically as follows: the cloud platform, based on the historical operating data of the transformer area, takes maximizing the voltage qualification rate and minimizing the number of switching operations as dual optimization objectives, obtains the optimal control parameters and issues them; after receiving the parameters, the local controller first performs boundary legality verification, and after the verification passes, the updated parameters take effect, while retaining the original parameter backup to support abnormal rollback.

[0045] It should be noted that, depending on the implementation needs, the various components described in the embodiments of the present invention can be divided into more components, or two or more components or parts of components can be combined into new components to achieve the purpose of the embodiments of the present invention. The above embodiments only illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be pointed out that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A reactive power compensation controller for low voltage management at the end of a power grid, characterized in that, It includes a modular housing, a main control unit (3), a sampling unit (5), a switching control interface (7), a remote communication unit (8), and a power supply unit (4). The main control unit (3) has a built-in online line impedance identification module, a voltage grade priority control module, a power factor optimization control module and a mode smooth switching module; The line impedance online identification module is used to identify the equivalent impedance parameters of the power supply lines at the end of the distribution network based on real-time collected voltage and current data. The voltage grading priority control module is used to match the corresponding grading compensation strategy according to the drop level of the three-phase lowest voltage. Under the undervoltage state, the power factor is shielded and the capacitor is put into operation first to raise the terminal voltage. The power factor optimization control module is used to switch capacitors on and off according to reactive power deficit under qualified voltage conditions to maintain the power factor in compliance with the standard. The mode smoothing switching module is used to perform hysteresis-free switching between voltage priority control mode and power factor priority control mode, suppressing switching oscillations near the threshold.

2. A reactive power compensation controller for low voltage management at the end of a power grid according to claim 1, characterized in that, The line impedance online identification module incorporates a dual identification mechanism: one is a switching disturbance calibration unit, which utilizes the changes in bus voltage and reactive power before and after capacitor switching. The system rated voltage satisfies the formula: X=ΔU·U N / ΔQ; Calculate the equivalent reactance X of the line, where ΔU is the voltage change, ΔQ is the reactive power change, and U N The first is the rated phase voltage of the system; the second is the recursive least squares correction unit, which is used to iteratively update the equivalent resistance and equivalent reactance parameters of the line under steady-state conditions based on the discrete model of line voltage drop.

3. A reactive power compensation controller for low voltage management at the end of a power grid according to claim 1, characterized in that, The voltage grading priority control module has two built-in voltage thresholds, dividing the voltage state into three levels: severe undervoltage, mild undervoltage, and voltage compliance. Under severe undervoltage conditions, based on the identified equivalent reactance X, the voltage is controlled using formula Q. need =(U th2 -U min )·U N / X calculates the target compensation capacity Q need Batch deployment of common compensation capacitor banks of corresponding capacity, of which U th2 For the qualified voltage threshold, U min This is the lowest three-phase voltage; under slight undervoltage conditions, step compensation is performed according to the single-group sequential connection strategy until the voltage rises back to the qualified range.

4. A reactive power compensation controller for low voltage management at the end of a power grid according to claim 1, characterized in that, The main control unit (3) also has a built-in capacitor life prediction and equalization switching module, which is used to calculate the remaining life coefficient of each capacitor group by weighted summation based on three dimensions: the number of capacitor switching, the operating environment temperature, and the actual capacity decay rate; when performing switching operations, the combination with the highest average remaining life coefficient is selected first among the capacitor combinations that meet the capacity requirements.

5. A reactive power compensation controller for low voltage management at the end of a power grid according to claim 1, characterized in that, The modular housing includes an upper main control compartment (1) and a lower base compartment (2), which are detachably connected to the positioning column by a buckle; the lower base compartment (2) has a built-in pluggable functional slot, and the sampling unit (5), power supply unit (4) and remote communication unit (8) all adopt an independent plug-in structure and are electrically connected to the main control unit (3) through the slot; the lower base compartment (2) has a standard DIN rail slot (10) integrated on the back.

6. A reactive power compensation controller for low voltage management at the end of a power grid according to claim 1, characterized in that, The remote communication unit (8) has a built-in Dianhong IoT operating system adaptation layer, supports 4G / 5G / NB-IoT wireless communication, can automatically report the device model to the power distribution IoT cloud platform and receive dynamic control parameters, and realize the control threshold self-tuning of cloud-edge collaboration; the sampling unit (5) supports automatic identification of three-phase four-wire, three-phase three-wire and single-phase two-wire wiring methods, and uses a 16-bit or higher precision ADC to realize synchronous sampling of three-phase voltage and three-phase current.

7. A reactive power compensation control method for low voltage mitigation at the end of a power grid, applied to the controller described in any one of claims 1-6, characterized in that, Includes the following steps: S1. Read the default control parameters and collect three-phase voltage and three-phase current data in real time; S2. Identify the equivalent impedance parameters of the end lines of the distribution network online and calculate the reactive power compensation capacity required for voltage rise under the current operating conditions. S3. Extract the lowest three-phase voltage and determine the voltage drop level; S4. If the voltage is undervoltage, enter the voltage grade priority control mode, match the corresponding grade strategy to put the capacitor into the field, and prioritize the increase of the terminal voltage. S5. If the voltage recovers to the qualified range, the system will smoothly switch to the power factor priority control mode through the hysteresis transition mechanism, and optimize the switching of capacitors according to the reactive power deficit. S6. Periodically update the running data and upload it to the cloud platform, receive the setting parameters issued by the cloud platform, and dynamically update the control threshold.

8. A reactive power compensation control method for low voltage management at the end of a power grid according to claim 7, characterized in that, In step S3, the voltage drop levels are divided into three levels: severe undervoltage, mild undervoltage, and voltage qualified. In step S4, when there is a severe undervoltage, the target compensation capacity is calculated based on the equivalent impedance, and capacitor banks of the corresponding capacity are put into operation in batches; when there is a slight undervoltage, capacitors are put into operation one group at a time according to the single-group cyclic switching strategy, and the voltage is detected again after a fixed interval after each group is put into operation; the power factor judgment logic is shielded throughout the undervoltage state.

9. A reactive power compensation control method for low voltage management at the end of a power grid according to claim 7, characterized in that, In step S5, the smooth mode switching adopts a dynamic hysteresis mechanism. The specific process is as follows: a dynamically adjustable hysteresis width is set based on the qualified voltage threshold to form a dual switching threshold for rising and falling edges; when the voltage rises from the undervoltage range to the hysteresis range, the addition of new capacitors is stopped, and the current compensation capacity is maintained; when the voltage continues to be higher than the qualified threshold and exceeds the set stable time, redundant capacitors are gradually removed, and the power factor detection logic is started to smoothly transition to the power factor priority control mode.

10. A reactive power compensation control method for low voltage management at the end of a power grid according to claim 7, characterized in that, The method also includes capacitor remaining life calculation and capacity online monitoring steps: the remaining life coefficient of each capacitor is calculated based on the number of switching, operating temperature and capacity decay rate, and the capacitor bank with the higher remaining life coefficient is selected first when switching. The current of each capacitor branch is monitored in real time and the actual output reactive power is calculated. When the capacity decay exceeds the set threshold, an alarm is generated and the faulty capacitor bank is blocked.

Images