A voltage regulation system for reactive power compensation equipment

Abstract

This invention relates to the field of power system technology and discloses a voltage regulation system for reactive power compensation equipment. The system first eliminates harmonic interference through a voltage and reactive power sensing unit to obtain accurate grid parameters. Then, a load trend prediction unit analyzes historical load curves and real-time trends to generate future prediction signals. A dynamic compensation calculation unit combines the current grid status with this prediction to calculate the optimal compensation capacity and generate switching commands. Upon receiving the command, the hybrid switching actuator first triggers the thyristor to achieve inrush-free conduction. After the current stabilizes, it switches the contactor to maintain low power consumption. Simultaneously, the compensation status monitoring unit adjusts parameter deviations in a closed loop, a multi-dimensional protection unit prevents overvoltage, overcurrent, and overheating, dual power supplies ensure continuous operation when the main power supply is abnormal, and an intelligent communication unit enables data uploading and remote collaboration. This system achieves rapid response switching of the reactive power compensation equipment and ensures the safety of a large-scale power grid, effectively suppressing inrush current impacts during the switching process.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a voltage regulation system for reactive power compensation equipment. Background Technology

[0002] In modern power systems, with the continuous growth of industrial loads and the widespread integration of distributed power sources, the reactive power balance problem of the power grid is becoming increasingly prominent. Frequent fluctuations in reactive power can cause the grid voltage to deviate from its rated value, leading to a series of problems such as voltage drops, voltage rises, decreased power factor, and increased line losses, seriously affecting power quality and the economic efficiency and stability of grid operation. Currently, static var compensation (SVC) technology is more widely accepted. SVC refers to the use of different static switches to switch reactors or capacitors, enabling them to generate and absorb reactive current, thereby improving the power factor of the power system, stabilizing the system voltage, and suppressing system oscillations.

[0003] For example, the patent with announcement number CN105406482B discloses a voltage and reactive power control method and system based on a reactive power compensation device. The core steps include: determining the location of the compensation point based on the branch topology and load distribution; deploying sensors at the beginning, end and compensation points to collect voltage and current data in real time; analyzing reactive power deficit and voltage deviation through algorithms such as model predictive control; and dynamically adjusting the capacity of each compensation point to form a hierarchical and partitioned compensation strategy.

[0004] Regarding the above and existing related technologies, the inventors believe that the following defects often exist: Traditional reactive power compensation equipment, such as parallel capacitor banks, usually adopts contactor switching, which has a slow response speed. During the switching process, inrush current and overvoltage are easily generated, which can cause impact on the equipment itself and the power grid, and it is difficult to adapt to rapidly changing power grid loads. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that the existing technology has the disadvantages of large inrush current when the voltage is unstable due to the dynamic fluctuation of reactive load in the power grid, and the response lag of traditional compensation methods. To this end, we propose a voltage regulation system for reactive power compensation equipment.

[0006] To achieve the above objectives, this application adopts the following technical solution: a voltage regulation system for reactive power compensation equipment, comprising:

[0007] The voltage and reactive power sensing unit is used to collect the three-phase voltage, current and reactive power of the power grid, and provides standardized data after filtering to eliminate harmonics.

[0008] The load trend prediction unit, based on historical data and real-time change rate analysis, predicts the reactive load fluctuation trend of the power grid in the future and outputs a prediction signal;

[0009] The dynamic compensation calculation unit is used to calculate the compensation capacity and generate switching commands based on real-time data, prediction signals and grid voltage ratings using the reactive power balance formula.

[0010] The hybrid switching actuator adopts a hybrid structure of "thyristor pre-switching + contactor stable operation" to achieve switching to low-energy operation of the contactor after inrush-free switching;

[0011] The compensation status monitoring unit is used to monitor the grid voltage, reactive power factor and operating status of compensation equipment in real time after switching. When a deviation occurs, it generates a feedback signal, triggers secondary regulation and stores the regulation data.

[0012] The multi-dimensional protection unit integrates overvoltage, overcurrent, overtemperature and module fault diagnosis functions. When the power grid and compensation equipment are abnormal, it cuts off the power supply, locates the fault and sends an alarm signal.

[0013] The dual power supply unit uses a combination of main power and backup lithium battery for power supply. When the power grid is abnormal, it automatically switches to backup power to maintain the continuous operation of the system.

[0014] The intelligent communication and interaction unit supports 5G+LoRa dual-mode communication, uploads system operation data to the remote dispatch center and receives manual control commands issued by the dispatch center, realizing two-way interaction between local automatic adjustment and remote monitoring.

[0015] Preferably, the load trend prediction unit specifically includes:

[0016] The historical load database is used to classify and store the reactive load data of the power grid for the past month. It supports automatic cleaning of redundant data from three months ago, provides a data foundation for trend analysis, and has the function of backing up data during power outages.

[0017] The real-time rate of change calculation module is used to receive real-time current / voltage data from the voltage reactive power sensing unit and calculate the reactive load change rate per minute through a differential algorithm to identify the three states of load: "stable / rising / falling".

[0018] The prediction algorithm module is used to output the reactive load prediction curve for the next 5-15 minutes by using the built-in improved LSTM neural network algorithm, combined with historical data and real-time change rate.

[0019] The prediction result verification module is used to compare the prediction result with the actual load data of the last three times in the same time period. If the deviation exceeds 10%, the algorithm parameters will be automatically corrected to avoid long-term prediction drift.

[0020] The prediction duration adjustment module is used to adjust the prediction time range and supports setting three prediction durations through remote or local operation to meet the needs of power grid scenarios.

[0021] Preferably, the real-time change rate calculation module uses the following formula to calculate the reactive load change rate:

[0022]

[0023] In the formula: R represents the average absolute rate of change; This represents the change in reactive power; Indicates the change over time; i represents the sequence number or index number of the current sampling point; Q i Q represents the reactive power value at the current sampling point i; i-1 τ represents the reactive power value of the previous sampling point i-1; n represents the total number of sampling points within the calculation window; τ represents the sampling time interval.

[0024] Preferably, the dynamic compensation calculation unit specifically includes:

[0025] The rated parameter storage module is used to store the rated parameters of the power grid and compensation equipment, and to fix the basic parameters including the rated voltage of the power grid, the target power factor and the rated capacity of the compensation equipment. The parameters can be modified online and automatically backed up.

[0026] The reactive power balance calculation module is used to calculate the required compensation capacity and, based on the compensation capacity balance calculation formula, combined with the real-time data transmitted by the voltage reactive power sensing unit and the trend signal transmitted by the load trend prediction unit, calculates the capacitive or inductive compensation capacity required by the power grid.

[0027] The switching status monitoring module is used to obtain the switching status of the compensation equipment. It obtains the current "on" or "off" status of the compensation equipment in real time by detecting the auxiliary contact signal of the contactor, so as to avoid generating duplicate switching or missed switching commands.

[0028] The operating condition adaptation module is used to automatically adjust the calculation logic of the reactive power balance calculation module to adapt to different operating conditions of the power grid by identifying accident conditions, including "voltage drop" and "impact load".

[0029] The compensation instruction generation module is used to generate specific switching instructions. Specifically, it converts the compensation capacity calculated by the reactive power balance calculation module into switching instructions for specific compensation devices, and includes execution timing requirements to provide operation instructions for the hybrid switching actuator.

[0030] Preferably, the compensation capacity balance calculation formula for the reactive power balance calculation module is as follows:

[0031]

[0032] In the formula: C represents the comprehensive compensation capacity; N represents the length of the real-time measurement data window; M represents the length of the predictive data window; Q mk Q represents the k-th real-time measured reactive power value;pj Q represents the j-th predicted reactive power value; t The target reactive power value is represented by k; the real-time data sampling point number is represented by k; and the predicted data sampling point number is represented by j.

[0033] Preferably, the hybrid switching actuator specifically includes:

[0034] The thyristor drive module is used to drive the thyristor to operate. After receiving the switching command issued by the dynamic compensation calculation unit, it outputs a trigger signal to control the thyristor to conduct, so as to realize the inrush current switching of the compensation equipment. The inrush current multiple is ≤1.2 times the rated current during switching. It is also compatible with two drive voltages, 220V and 380V, to meet the drive requirements of different compensation equipment.

[0035] The contactor switching module is used to switch to the contactor operation mode. Within 300ms after the thyristor drive module completes the switching, it drives the contactor to engage to replace the thyristor for long-term operation, reducing the power consumption of the hybrid switching actuator to 1 / 5 of that of the thyristor operation mode.

[0036] The inrush current suppression module is used to suppress the inrush current during switching using a built-in current-limiting resistor. The current-limiting resistor is connected to the circuit at the moment the compensation equipment is switched on, and the current-limiting resistor is short-circuited after the current stabilizes, so as to prevent the inrush current at the moment of switching from damaging the compensation equipment or affecting the stability of the grid voltage.

[0037] The execution status feedback module is used to detect the current signal of the circuit after switching through the current sensor to determine whether the compensation equipment is normally put into operation or disconnected, and to send the status signal of "successful switching" or "failed switching" back to the dynamic compensation calculation unit and the compensation status monitoring unit to form a closed loop control.

[0038] The execution timing control module is used to control the timing of switching actions. According to the timing requirements in the instructions of the dynamic compensation calculation unit, it controls the action sequence of the thyristor drive module and the contactor switching module.

[0039] Preferably, the inrush current suppression module adopts a two-stage suppression strategy. The first stage limits di / dt through a magnetically saturated reactor, and the second stage suppresses overvoltage through a parallel RC absorption circuit, wherein the resistance value is 10Ω±5% and the capacitance value is 0.1μF±5%.

[0040] Preferably, the compensation state monitoring unit specifically includes:

[0041] The real-time parameter monitoring module is used to synchronously collect grid voltage, current and power factor data after the compensation equipment is switched on to obtain grid parameters after switching. The sampling frequency is consistent with that of the voltage and reactive power sensing unit, providing real-time data for judging the effect of the compensation status monitoring unit.

[0042] The target value comparison module is used to compare the voltage and power factor collected by the real-time parameter monitoring unit with the rated target values, calculate the voltage deviation value and power factor deviation value, and provide deviation data for the compensation state monitoring unit to judge the adjustment effect.

[0043] The deviation signal generation module is used to generate a deviation feedback signal. If the deviation value calculated by the target value comparison module exceeds the set threshold, a deviation signal is generated immediately and transmitted to the dynamic compensation calculation unit to trigger secondary adjustment.

[0044] The adjustment data storage module is used to store adjustment process data. It stores complete data for each compensation adjustment in the format of "time-compensation capacity-adjustment effect", retains 6 months of historical records and supports exporting data in Excel format, providing data basis for power grid operation and maintenance personnel to analyze the adjustment effect;

[0045] The monitoring frequency synchronization module is used to synchronize the monitoring frequency and the acquisition frequency. It works in conjunction with the acquisition frequency control module of the voltage and reactive power sensing unit to ensure that the sampling timestamp of the real-time parameter monitoring module is consistent with the acquisition timestamp of the voltage and reactive power sensing unit.

[0046] Preferably, the multi-dimensional protection unit specifically includes:

[0047] The overvoltage and overcurrent detection module is used to monitor the grid voltage and compensation circuit current in real time. When the grid voltage exceeds the rated value by ±10% or the compensation circuit current exceeds the rated value by 1.5 times, it immediately outputs an overvoltage or overcurrent protection trigger signal, providing trigger conditions for the multi-dimensional protection unit to start protection.

[0048] The equipment over-temperature monitoring module is used to monitor and compensate for the temperature of the equipment. It uses an NTC thermistor to collect the temperature of the thyristor and contactor in the hybrid switching actuator. When the temperature exceeds 60°C, it outputs an over-temperature protection trigger signal.

[0049] The compensation module diagnostic module is used to detect the impedance value of capacitors and reactors once per hour using the impedance detection method. If the impedance value deviates from the rated value by ±20%, the corresponding compensation module is determined to be faulty, the fault number is locked, and a fault trigger signal is output.

[0050] The protection command execution module is used to receive the protection trigger signals output by the overvoltage and overcurrent detection module, the equipment overtemperature monitoring module, and the compensation module diagnostic module, and then cut off the power circuit of the hybrid switching actuator within 100ms, while locking the compensation command receiving function of the dynamic compensation calculation unit.

[0051] The local alarm driver module is used to drive the local audible and visual alarm to activate when the protection command execution module starts protection, and at the same time display the fault type on the device panel;

[0052] The protection threshold setting module is used to adjust the protection threshold. It supports adjusting the overvoltage, overcurrent, and overtemperature protection thresholds through remote or local operation, enabling the multi-dimensional protection unit to adapt to power grid scenarios with different stability requirements.

[0053] Preferably, the dual power supply unit specifically includes:

[0054] The power switching control module is used to realize the switching between main and backup power. It monitors the grid power voltage through a voltage detection chip with a detection accuracy of ±0.5V. When the grid voltage is lower than 320V or higher than 450V, it automatically switches to the backup lithium battery pack within 50ms. The switching process is seamless to ensure uninterrupted power supply to the system.

[0055] The battery power monitoring module is used to collect voltage and current data of the backup lithium battery pack in real time to calculate and monitor the remaining power of the lithium battery. When the remaining power is less than 20%, it sends a "low battery" signal to the intelligent communication device to remind maintenance personnel to charge the lithium battery in time.

[0056] The power supply voltage regulator module is used to regulate the AC power input from the grid power access unit and the DC power output from the backup lithium battery pack, and outputs a stable DC power of 24V±0.5V to provide stable power supply for each module of the system.

[0057] The technical effects and advantages of this invention are as follows:

[0058] In this invention, the voltage and reactive power sensing unit filters grid parameters to eliminate harmonic interference, providing accurate data support for subsequent modules. The load trend prediction unit then analyzes historical load curves and real-time trends to generate prediction signals for future time periods. The dynamic compensation calculation unit combines the current grid status with the load prediction results to calculate the optimal compensation capacity and generate switching commands. Upon receiving the command, the hybrid switching actuator first triggers the thyristor to achieve inrush-free conduction. After the current stabilizes, it switches to the contactor to maintain low-power operation. This achieves rapid response switching of reactive power compensation, effectively suppresses inrush current impacts during the switching process, and reduces system operating energy consumption. The compensation status monitoring unit continuously collects grid parameters after switching. When the voltage or power factor deviation exceeds the threshold, dynamic compensation is immediately triggered. The calculation unit performs secondary adjustments and ensures that the compensation effect continues to meet the standards through a closed-loop monitoring mechanism, significantly reducing the frequency of manual intervention. The multi-dimensional protection unit monitors the status of the power grid and equipment in real time. Once overvoltage, overcurrent, or overheating is detected, the power circuit is immediately cut off, significantly reducing the risk of equipment failure and extending the service life of key components. The dual power supply unit automatically switches to the backup battery when the main power supply is abnormal, ensuring continuous system operation and further improving power supply reliability. The intelligent communication interaction unit uploads local operating data to the remote center and receives remote control commands to achieve coordinated adjustment. The load forecasting function allows the system to adjust the compensation capacity in advance, efficiently adapting to the rapid fluctuations of industrial loads. Overall, the timeliness and accuracy of voltage regulation are greatly improved, ultimately effectively ensuring the safety of large-scale power grids. Attached Figure Description

[0059] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts:

[0060] Figure 1 This is an overall architecture diagram of a voltage regulation system for a reactive power compensation device according to the present invention;

[0061] Figure 2 This is a schematic diagram illustrating the working principle of a voltage regulation system for a reactive power compensation device according to the present invention, which controls voltage by controlling reactive power.

[0062] Figure 3 This is a load trend prediction unit architecture diagram of a voltage regulation system for a reactive power compensation device according to the present invention;

[0063] Figure 4 This is a diagram of the dynamic compensation calculation unit architecture of the voltage regulation system of a reactive power compensation device according to the present invention.

[0064] Figure 5This is a diagram of a hybrid switching actuator architecture for a voltage regulation system of a reactive power compensation device according to the present invention.

[0065] Figure 6 This is an architecture diagram of the compensation status monitoring unit of the voltage regulation system of a reactive power compensation device according to the present invention;

[0066] Figure 7 This is a multi-dimensional protection unit architecture diagram of a voltage regulation system for a reactive power compensation device according to the present invention;

[0067] Figure 8 This is a diagram of the dual-power supply unit architecture of a voltage regulation system for a reactive power compensation device according to the present invention. Detailed Implementation

[0068] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0069] Reference Figures 1-2 As shown, the present invention provides a technical solution: a voltage regulation system for a reactive power compensation device, comprising:

[0070] Voltage and reactive power sensing unit: Real-time acquisition of three-phase voltage, current signals and reactive power values ​​of the power grid, elimination of harmonic interference through built-in filtering algorithm, and synchronous transmission of standardized data to the calculation module and monitoring module, providing accurate raw data support for subsequent regulation.

[0071] Load trend prediction unit: Creatively introduces historical load database and real-time load change rate analysis algorithm. Based on the data transmitted by the sensing module, it predicts the reactive load fluctuation trend of the power grid in the next 5-15 minutes (such as the load increase trend before the peak of industrial electricity consumption) and outputs the prediction signal to the compensation calculation module in advance to avoid the problem of traditional "lagging adjustment".

[0072] Dynamic compensation calculation unit: Receives real-time data from the sensing module and trend signals from the prediction module, combines the grid voltage rating (e.g., 380V / 10kV), calculates the required compensation capacity (capacitive / inductive) using the reactive power balance formula, and simultaneously determines the switching status of the current compensation equipment (e.g., capacitor banks, reactors), generates the optimal switching command, and transmits it to the execution module.

[0073] Hybrid switching actuator: Breaking through the traditional single contactor or thyristor switching mode, it adopts a hybrid structure of "thyristor pre-switching + contactor stable operation". After receiving the instruction from the calculation module, it first achieves inrush-free switching through the thyristor. After the current stabilizes, it switches to contactor operation to reduce energy consumption, completes the precise input or output of compensation capacity, and ensures no voltage surge during the regulation process.

[0074] Compensation Status Monitoring Unit: Tracks the voltage value, reactive power factor and operating status of the power grid after switching in real time, compares them with the target value set by the calculation module (such as power factor ≥ 0.95), and if a deviation occurs (such as voltage still being too low), generates a feedback signal to the dynamic compensation calculation unit to trigger secondary adjustment; at the same time, it stores the adjustment data to provide a basis for subsequent operation and maintenance.

[0075] Multi-dimensional protection unit: integrates overvoltage, overcurrent, overtemperature and module fault diagnosis functions. When the mains voltage exceeds the rated value by ±10%, the compensation equipment current is overloaded or the actuator temperature exceeds 60℃, the power supply to the switching actuator is immediately cut off. If a compensation module (such as a capacitor) is damaged, the faulty module is locked by the built-in fault location algorithm and the fault signal is sent to the communication module and local alarm light simultaneously to avoid equipment damage or power grid accidents.

[0076] Dual power supply unit: It adopts a dual power supply structure of "main power (grid power) + backup lithium battery". When the main power is normal, it supplies power to the entire system. When the grid fails or the voltage is abnormal, it automatically switches to the backup power to ensure uninterrupted operation of the sensing, protection and communication modules and avoid regulation interruption caused by power failure.

[0077] Intelligent communication interaction unit: Supports 5G+LoRa dual-mode communication. On the one hand, it uploads system operation data (voltage, power factor, fault information) to the remote dispatch center. On the other hand, it receives manual adjustment instructions issued by the dispatch center (such as customized compensation parameters under special working conditions), realizing two-way interaction between local autonomous adjustment and remote monitoring, and improving system flexibility.

[0078] In addition, the voltage reactive power sensing unit specifically includes:

[0079] The three-phase voltage acquisition module, as the voltage signal acquisition component of the voltage reactive power sensing unit, is used to acquire the three-phase voltage signals of the power grid (A, B, and C phases). Specifically, it achieves signal acquisition through a high-precision voltage transformer with an error ≤0.2%, adapting to multiple voltage levels of 10kV / 380V. At the same time, it converts the acquired analog voltage signals into 0-5V standard analog outputs, providing accurate raw three-phase voltage data for the voltage reactive power sensing unit.

[0080] The three-phase current acquisition module, as the current signal acquisition component of the voltage reactive power sensing unit, is used to acquire the three-phase circuit current signal of the power grid in real time. Specifically, it avoids the interference of large current on the acquisition accuracy by using Hall current sensors, synchronously outputs an analog signal that is proportional to the actual current, and ensures that the output signal is consistent with the acquisition timing of the three-phase voltage acquisition unit, so as to provide the voltage reactive power sensing unit with synchronous three-phase current raw data.

[0081] The harmonic filtering module, as a signal purification component of the voltage and reactive power sensing unit, is used to eliminate the interference of power grid harmonics on the collected data. Specifically, it adopts a combination of a second-order RC low-pass filter circuit and a Kalman digital filter algorithm to filter out the 3rd, 5th, and 7th harmonic components in the power grid, avoid voltage / current data distortion caused by harmonics, and ensure the accuracy of the output data of the voltage and reactive power sensing unit.

[0082] The acquisition frequency control module, as the acquisition rhythm control component of the voltage and reactive power sensing unit, is used to adjust the data acquisition frequency. Specifically, it supports setting the sampling frequency from 5 to 20 Hz through remote or local operation. The default sampling interval is 10 Hz to balance data real-time performance and system computing power consumption. At the same time, the acquired voltage and current data are temporarily stored in the buffer area and transmitted to other cooperating components of the voltage and reactive power sensing unit after standardization processing.

[0083] Specifically, the voltage and reactive power sensing unit eliminates harmonic interference through filtering, providing accurate grid parameters for subsequent modules. The load trend prediction unit analyzes historical load curves and real-time trends to generate prediction signals for future time periods. The dynamic compensation calculation unit integrates the current grid status and future load forecasts to calculate the optimal compensation capacity and generate switching commands. Upon receiving the command, the hybrid switching actuator first triggers the thyristor to achieve inrush-free conduction, and then switches to the contactor to maintain low-power operation after the current stabilizes. The compensation status monitoring unit continuously collects grid parameters after switching, and triggers the dynamic compensation calculation unit for secondary adjustment when voltage or power factor deviations exceed thresholds. The multi-dimensional protection unit monitors the grid and equipment status in real time, and immediately cuts off the power circuit when overvoltage, overcurrent, or overheating is detected. The dual power supply unit automatically switches to the backup battery when the main power supply is abnormal, ensuring continuous system operation. The intelligent communication interaction unit uploads local operating data to the remote center and simultaneously receives remote control commands to achieve coordinated adjustment.

[0084] The above solutions enable rapid response switching of reactive power compensation equipment, effectively suppressing inrush current surges during the switching process and improving the timeliness and accuracy of voltage regulation. Load forecasting allows the system to adjust compensation capacity in advance to adapt to rapid fluctuations in industrial loads. A closed-loop monitoring mechanism ensures continuous achievement of compensation standards, reducing the frequency of manual intervention. Multiple protection functions reduce the risk of equipment failure and extend the service life of key components. The dual-power supply design guarantees the system's continuous operation during grid anomalies, improving power supply reliability and effectively ensuring the safety of large-scale power grids.

[0085] Reference Figure 1 and Figure 3 As shown, this application further proposes a load trend prediction unit, including a historical load database, a real-time change rate calculation module, a prediction algorithm operation module, a prediction result verification module, and a prediction duration adjustment module.

[0086] The historical load database, which serves as the data storage component of the load trend prediction unit, stores historical reactive load data of the power grid. Specifically, it stores reactive load data of the power grid for the past month in categories of day, hour, and minute. It has the function of automatically cleaning up redundant data from three months ago and also supports data backup after power outages, providing basic data support for trend analysis of the load trend prediction unit.

[0087] The real-time rate of change calculation module, which serves as the load status identification component of the load trend prediction unit, is used to calculate the reactive load change rate. Specifically, it receives real-time current and voltage data transmitted by the voltage and reactive power sensing unit, calculates the reactive load change rate (ΔQ / Δt) per minute through a differential algorithm, and identifies the load as being in one of three states: stable, rising, or falling, based on the change rate value, thus providing real-time status data for the prediction algorithm of the load trend prediction unit.

[0088] The prediction algorithm operation module, as the core calculation component of the load trend prediction unit, is used to generate reactive load prediction results. Specifically, it incorporates an improved LSTM neural network algorithm that integrates grid operating conditions and combines historical data from the historical load database with real-time change rate data from the real-time change rate calculation unit to output a reactive load prediction curve for the next 5-15 minutes. The prediction error is controlled within ±8%, providing the core prediction signal for the load trend prediction unit.

[0089] The prediction result verification module serves as the prediction accuracy correction component for the load trend prediction unit. It is used to calibrate the prediction result deviation. Specifically, it compares the prediction result output by the prediction algorithm running unit with the actual load data of the same time period in the last three times. If the deviation exceeds 10%, it automatically corrects the parameters of the LSTM neural network algorithm (such as adjusting the neural network weights) to avoid long-term prediction drift problems in the load trend prediction unit.

[0090] The prediction duration adjustment module, as a scenario adaptation component of the load trend prediction unit, is used to adjust the prediction time range. Specifically, it supports setting three prediction durations of 5 minutes, 10 minutes, and 15 minutes through remote or local operation. It can be selected according to the needs of the power grid scenario (such as selecting 15 minutes for peak industrial electricity consumption scenarios and 5 minutes for residential electricity consumption scenarios), so that the load trend prediction unit can adapt to the prediction needs of different power grid operation scenarios.

[0091] Specifically, the historical load database stores recent reactive power load data from the power grid, providing a training and reference basis for the prediction algorithm. Its automatic cleanup mechanism avoids the impact of historical data overload on analysis efficiency. The real-time rate of change calculation module continuously receives real-time parameters from the power grid and calculates the reactive power load change rate every minute through differential operations, classifying the load status into three types: stable, rising, or falling, providing a basis for the prediction algorithm to judge the current operating condition. The prediction algorithm operation module combines historical data patterns and real-time trends, utilizing the time series processing capabilities of an improved LSTM neural network to output a reactive power load prediction curve for a specific future time window. The prediction result verification module compares the prediction results with the actual data from the three most recent time periods. When the deviation exceeds a set threshold, it automatically triggers algorithm parameter correction to prevent the prediction model from accumulating errors due to environmental changes. The prediction duration adjustment module provides three adjustable prediction time ranges, allowing the prediction time window to be adjusted via a local interface or remote commands, enabling the prediction cycle to adapt to the speed of power grid load changes.

[0092] The above solution effectively addresses the response lag problem caused by insufficient prediction accuracy in traditional reactive power compensation equipment. By integrating historical patterns with real-time change characteristics, it achieves accurate load prediction, enabling the compensation device to prepare the required compensation capacity in advance. A dynamic verification mechanism ensures the prediction model continuously adapts to changes in grid operating conditions, avoiding the accumulation of prediction deviations due to long-term use. The adjustable prediction duration function allows the system to match the needs of different scenarios, such as rapid fluctuations in industrial loads and stable changes in commercial loads, significantly improving the scenario adaptability of the reactive power compensation system.

[0093] This application further proposes the following formula for calculating the reactive load change rate in the real-time change rate calculation module:

[0094]

[0095] In the formula: R represents the average absolute rate of change; This represents the change in reactive power; Indicates the change over time; i represents the sequence number or index number of the current sampling point; Q i Q represents the reactive power value at the current sampling point i; i-1τ represents the reactive power value of the previous sampling point i-1; n represents the total number of sampling points within the calculation window; τ represents the sampling time interval.

[0096] The average absolute rate of change, R, is the ratio of the sum of the absolute differences in reactive power values ​​among all adjacent sampling points within the calculation window to the time interval and the total number of sampling points. It can be implemented using a sliding window algorithm to eliminate noise interference from single-point sampling and reflect the overall trend of reactive load changes. The reactive power change, ΔQ, is the reactive power difference between adjacent sampling points, calculated using a differential algorithm to capture the instantaneous changes in reactive load and avoid misjudgments caused by relying solely on amplitude changes. The time change, Δt, is the time interval between adjacent sampling points, implemented using a high-precision clock module to combine reactive power change with the time dimension, quantifying the dynamic rate of load change. The total number of sampling points, n, refers to the number of sampling points included in the calculation window. This can be set using configurable parameters to control the sample range covered by the calculation, balancing real-time performance and data integrity requirements. The sampling time interval, τ, is the data acquisition period, adjusted by changing the data acquisition frequency to adapt to the accuracy requirements of rate of change calculations in different power grid scenarios.

[0097] Specifically, the average absolute rate of change R is obtained by summing the absolute differences of reactive power values ​​at all adjacent sampling points within the window and normalizing this sum with the time interval and the total number of sampling points. By introducing a time-dimensional parameter, the instantaneous change in reactive power is transformed into a dynamic rate indicator, thus avoiding misjudgments caused by data fluctuations at a single point in time. During the calculation, the sampling point sequence number i ensures that the data is processed continuously in chronological order, the total number of sampling points n controls the sample range covered by the calculation, and the sampling time interval τ matches the actual operating scenario of the power grid. For example, in scenarios with frequent load fluctuations, the calculation frequency can be increased by shortening the sampling time interval τ, while in scenarios with stable loads, the calculation stability can be improved by increasing the total number of sampling points n. Thus, this formula can dynamically adjust the calculation parameters to achieve accurate quantification of the reactive load change rate.

[0098] Reference Figure 1 and Figure 4 As shown, this application further proposes a voltage regulation system for reactive power compensation equipment, whose dynamic compensation calculation unit includes a rated parameter storage module, a reactive power balance calculation module, a switching status monitoring module, an operating condition adaptation module, and a compensation command generation module.

[0099] The rated parameter storage module serves as the basic parameter storage component for the dynamic compensation calculation unit. It stores the rated parameters of the power grid and compensation equipment, specifically fixing basic parameters such as the rated voltage of the power grid (e.g., 10kV±5%), the target power factor (e.g., ≥0.95), and the rated capacity of the compensation equipment (e.g., 20kvar for a single capacitor). It supports online modification of parameters and automatic backup, providing parameter basis for the compensation capacity calculation of the dynamic compensation calculation unit.

[0100] The reactive power balance calculation module, as the core calculation component of the dynamic compensation calculation unit, is used to calculate the required compensation capacity. Specifically, it is based on the compensation capacity balance calculation formula, combined with the real-time data transmitted by the voltage reactive power sensing unit and the trend signal transmitted by the load trend prediction unit, to calculate the capacitive or inductive compensation capacity required by the power grid. The calculation accuracy is accurate to 1 kvar, providing capacity data for the dynamic compensation calculation unit to generate switching commands.

[0101] The switching status monitoring module, as the equipment status sensing component of the dynamic compensation calculation unit, is used to obtain the switching status of the compensation equipment. Specifically, it obtains the "on" or "off" status of the current compensation equipment (capacitor bank / reactor) in real time by detecting the auxiliary contact signal of the contactor, so as to avoid the dynamic compensation calculation unit from generating duplicate switching or missed switching commands and ensure the accuracy of the operation of the compensation equipment.

[0102] The operating condition adaptation module, as the operating condition response component of the dynamic compensation calculation unit, is used to adapt to special operating conditions of the power grid. Specifically, by identifying special operating conditions such as "sudden voltage drop" and "impact load" of the power grid, it automatically adjusts the calculation logic of the reactive power balance calculation module (such as increasing the compensation capacity redundancy by 10% under the impact load condition) to avoid the dynamic compensation calculation unit from experiencing adjustment lag problems under special operating conditions.

[0103] The compensation instruction generation module, as the instruction output component of the dynamic compensation calculation unit, is used to generate specific switching instructions. Specifically, it converts the compensation capacity calculated by the reactive power balance calculation module into switching instructions for specific compensation equipment (such as "connecting capacitor #2" and "disconnecting reactor #1"), while also including execution timing requirements (such as a 2-second interval between switching of multiple devices), providing operation instructions for the hybrid switching actuator.

[0104] Specifically, the rated parameter storage module stores modifiable grid rated parameters, providing a dynamic adjustment basis for compensation calculations. The reactive power balance calculation module synchronously receives real-time measurement data and load trend prediction signals, and calculates the current and future required compensation capacity using the compensation capacity balance formula. The switching status monitoring module continuously collects contactor auxiliary contact signals, automatically blocking duplicate switching commands when it detects that the equipment is in the switched-on state. The operating condition adaptation module analyzes the waveform distortion characteristics during voltage drops and triggers an emergency compensation mode to adjust the calculation logic. The compensation command generation module converts the final calculation results into a command sequence containing execution timing, controlling the hybrid switching actuator to complete the switching action according to the preset timing.

[0105] This application achieves proactive compensation through a calculation method that integrates real-time data and predictive signals. Traditional systems lack equipment status monitoring capabilities, leading to frequent malfunctions. This application effectively blocks erroneous commands through hardware-level monitoring of contactor auxiliary contact signals. Conventional compensation algorithms still employ standard calculation logic under fault conditions. This application automatically switches to a calculation mode adapted to different scenarios through operating condition identification. This solution solves the problem of insufficient adaptability caused by fixed parameters in traditional systems, avoids compensation lag caused by relying solely on real-time data, eliminates the risk of malfunctions due to the lack of equipment status monitoring, overcomes the technical deficiency of a single calculation mode in handling complex operating conditions, and achieves accurate dynamic compensation under different operating states of the power grid.

[0106] This application further proposes the following formula for calculating the compensation capacity balance of the reactive power balance calculation module:

[0107]

[0108] In the formula: C represents the comprehensive compensation capacity; N represents the length of the real-time measurement data window; M represents the length of the predictive data window; Q mk Q represents the k-th real-time measured reactive power value; pj Q represents the j-th predicted reactive power value; t The target reactive power value is represented by k; the real-time data sampling point number is represented by k; and the predicted data sampling point number is represented by j.

[0109] The real-time measurement data window length N refers to the number of real-time reactive power data sampling points used to calculate the compensation capacity. This can be implemented using a sliding window algorithm, adjusting the window length to balance calculation accuracy and response speed. The predictive data window length M refers to the number of predictive reactive power data sampling points used to calculate the compensation capacity. This can be implemented using a dynamic matching algorithm, dynamically adjusting the weight of predictive data in the calculation based on load change trends. The target reactive power value Q... tThis refers to the expected balance value of reactive power in the power grid, which can be set through the grid's rated parameters and operating strategies, providing a dynamic benchmark for compensation capacity calculation. Real-time measurement of reactive power value Q. mk This refers to the current reactive power data of the power grid collected by the voltage reactive power sensing unit. Specifically, synchronous sampling technology can be used to obtain this data, ensuring consistency with the actual operating conditions of the power grid. The predicted reactive power value Q... pj This refers to the reactive power prediction data for future periods output by the load trend prediction unit. Specifically, it can be generated through an improved LSTM neural network algorithm to achieve dynamic tracking of load trends.

[0110] Specifically, in the reactive power balance calculation process, the real-time measurement data window length N and the predicted data window length M are simultaneously incorporated into the calculation model. This is achieved by incorporating the real-time reactive power Q... mk With the predicted reactive power Q pj Respectively compared with the target reactive power Q t The sum of squared deviations is calculated to form a composite compensation capacity assessment index. When the real-time data window length N is set to a shorter value, such as a window with 5 sampling points, it can quickly respond to sudden changes in the reactive power of the power grid; when the prediction data window length M is set to a longer value, such as a window with 10 sampling points, it can effectively smooth prediction errors. Target reactive power Q t As a dynamic benchmark, it is automatically updated when the power grid operating conditions change. For example, when a voltage drop is detected, the target value can be temporarily adjusted to the required value to maintain voltage stability. The weighted comprehensive calculation of real-time measurement and prediction data ensures that the compensation capacity includes both immediate feedback on the current power grid status and prediction information of future load trends, thereby avoiding the problems of compensation lag or lead caused by a single data source.

[0111] Compared to existing technologies, traditional reactive power compensation equipment calculates capacity based solely on real-time measurement data or fixed historical trends, failing to address scenarios with rapid load fluctuations. For example, existing methods often result in inaccurate compensation capacity calculations during impact loads due to excessively short real-time data windows, or redundant compensation due to missing predictive data during stable load periods. This solution, however, establishes a dynamic coupling model between real-time and predictive data. During load surges, it prioritizes short-window real-time data to improve response speed, while combining long-window predictive data to enhance calculation accuracy during stable load periods, achieving adaptive compensation under different operating conditions.

[0112] The above solution effectively solves the problem of calculation deviation in compensation capacity caused by the single data source in traditional methods. By integrating real-time measurement and predictive data into a composite calculation model, compensation requirements are accurately matched when the grid load changes rapidly, avoiding voltage instability caused by insufficient compensation or frequent equipment switching due to overcompensation. Simultaneously, the dynamic adjustment capability of the window length allows the compensation strategy to adapt to different grid operating states, optimizing response speed while ensuring calculation accuracy and improving the overall regulation efficiency of the reactive power compensation system.

[0113] Reference Figure 1 and Figure 5 As shown, this application further proposes a hybrid switching actuator, including a thyristor drive module, a contactor switching module, an inrush current suppression module, an execution status feedback module, and an execution timing control module.

[0114] The thyristor drive module, as the inrush-free switching component of the hybrid switching actuator, is used to drive the thyristor to operate. Specifically, after receiving the switching command issued by the dynamic compensation calculation unit, it outputs a trigger signal to control the thyristor to conduct, realizing inrush-free switching of the compensation equipment. During switching, the inrush current multiple is ≤1.2 times the rated current. It is also compatible with two driving voltages, 220V and 380V, to meet the driving requirements of different compensation equipment.

[0115] The contactor switching module serves as a low-power operation switching component for the hybrid switching actuator. It is used to switch to the contactor operation mode. Specifically, within 300ms after the thyristor drive module completes the switching (after the circuit current stabilizes), the contactor is driven to engage, replacing the thyristor for long-term operation. This reduces the operating power consumption of the hybrid switching actuator to 1 / 5 of that in the thyristor operation mode, achieving low-energy operation.

[0116] The inrush current suppression module serves as a secondary protection component for the hybrid switching actuator, further suppressing the inrush current during switching. Specifically, it has a built-in current-limiting resistor. When the compensation equipment is switched on, the current-limiting resistor is connected to the circuit. After the current stabilizes, the current-limiting resistor is short-circuited to prevent the inrush current during switching from damaging the compensation equipment or affecting the stability of the power grid voltage.

[0117] The execution status feedback module serves as the status feedback component of the hybrid switching actuator. It is used to provide feedback on the switching execution results. Specifically, it detects the current signal of the circuit after switching through a current sensor to determine whether the compensation equipment is normally engaged or disengaged. It then sends the status signal of "successful switching" or "failed switching" back to the dynamic compensation calculation unit and the compensation status monitoring unit to form a closed-loop control.

[0118] The execution timing control module, as the action rhythm control component of the hybrid switching actuator, is used to control the timing of switching actions. Specifically, it strictly follows the timing requirements in the instructions of the dynamic compensation calculation unit (such as the switching interval of multiple compensation devices) to control the action sequence of the thyristor drive module and the contactor switching module, so as to avoid grid voltage fluctuations caused by the simultaneous switching of multiple devices.

[0119] Specifically, after the dynamic compensation calculation unit issues the switching command, the thyristor drive module first triggers conduction to achieve contactless rapid switching, utilizing the millisecond-level response characteristics of semiconductor devices to complete the initial connection. After the thyristor conduction reaches a stable state, the contactor switching module immediately drives the mechanical contacts to close, switching the main circuit to low-impedance operation. The inrush current suppression module automatically connects the current-limiting resistor at the moment of switching, and after the current stabilizes, it short-circuits the resistor through a bypass relay, forming a two-stage current buffer mechanism. The execution status feedback module continuously monitors the circuit current, and when it detects a stable current within a preset threshold range, it determines that the switching is successful and feeds back the status signal to the control system to form a closed-loop verification. The execution timing control module precisely controls the time interval between the thyristor trigger pulse and the contactor drive signal through preset timing logic to avoid device action conflicts.

[0120] The above scheme achieves a balance between rapid response and low-loss operation during the switching process of reactive power compensation equipment, effectively suppressing current surges generated during switching operations and ensuring the operational safety of the compensation equipment and the power grid. A closed-loop state feedback mechanism ensures the effectiveness of each switching action, avoiding potential malfunctions that may occur with traditional open-loop control. The timing control strategy resolves the timing conflicts arising from the coordinated operation of multiple types of switching devices, providing a reliable control foundation for the engineering application of hybrid switching structures.

[0121] This application further proposes a two-stage suppression strategy for the inrush current suppression module. The first stage limits the rate of current change through a magnetically saturated reactor, and the second stage suppresses overvoltage through a parallel resistor-capacitor absorption circuit. The resistance value is 10Ω ± 5%, and the capacitance value is 0.1μF ± 5%. The magnetically saturated reactor refers to an inductor with nonlinear magnetization characteristics, specifically implemented using an iron-based amorphous alloy magnetic core winding structure. Its permeability automatically adjusts with changes in current intensity, suppressing the current rise rate by increasing the inductance during sudden current changes. The parallel resistor-capacitor absorption circuit is an energy dissipation circuit composed of a resistor and a capacitor connected in series. Specifically, it can be implemented using a combination of a metal film resistor and a polypropylene film capacitor. The resistor converts transient electrical energy into heat energy, while the capacitor absorbs high-frequency voltage spikes.

[0122] Specifically, at the moment the compensation equipment switches on and off, the magnetically saturated reactor first responds to the sudden current change, using its magnetic saturation characteristics to generate a back electromotive force that limits the rate of current change, thus controlling the inrush current amplitude within a safe range. Subsequently, the parallel resistor-capacitor absorption circuit begins operation, absorbing the transient overvoltage energy generated when the switch contacts open through the capacitor, while the resistor limits the circuit current and dissipates the absorbed energy. This two-stage suppression strategy works in tandem; the magnetically saturated reactor focuses on suppressing the initial stage of the current surge, while the resistor-capacitor circuit focuses on eliminating the subsequent stage of the voltage spike. Together, they achieve impact suppression throughout the entire process.

[0123] The above solutions effectively reduce the risk of impact damage to power electronic devices and grid equipment caused by the switching operation of reactive power compensation equipment, suppress contact contact arcing and capacitor insulation aging, improve the operational reliability of the compensation system under frequent switching conditions, and control the equipment size and operating energy consumption while ensuring the suppression effect by optimizing the matching of resistor and capacitor parameters.

[0124] Reference Figure 1 and Figure 6 As shown, this application further proposes a compensation state monitoring unit, including a real-time parameter monitoring module, a target value comparison module, a deviation signal generation module, an adjustment data storage module, and a monitoring frequency synchronization module.

[0125] The real-time parameter monitoring module, as the data acquisition component of the compensation status monitoring unit, is used to acquire grid parameters after switching. Specifically, it synchronously collects grid voltage, current, and power factor data after the compensation equipment is switched, and the sampling frequency is consistent with that of the voltage and reactive power sensing unit (default 10Hz) to ensure the timeliness of the collected data and provide real-time data for judging the effect of the compensation status monitoring unit.

[0126] The target value comparison module, as the deviation identification component of the compensation state monitoring unit, is used to compare the monitoring data with the target value. Specifically, it compares the voltage (e.g., 10kV) and power factor (e.g., 0.92) collected by the real-time parameter monitoring module with the rated target values ​​(10kV±5%, ≥0.95), calculates the voltage deviation value (e.g., -3%) and power factor deviation value (e.g., -0.03), and provides deviation data for the compensation state monitoring unit to judge the adjustment effect.

[0127] The deviation signal generation module serves as the feedback signal output component of the compensation status monitoring unit. It generates deviation feedback signals. Specifically, if the deviation value calculated by the target value comparison unit exceeds the set threshold (such as voltage deviation ±3% or power factor deviation ±0.02), a deviation signal (such as "voltage is too low and capacity needs to be increased") is immediately generated and transmitted to the dynamic compensation calculation unit to trigger secondary adjustment.

[0128] The adjustment data storage module, as the data recording component of the compensation status monitoring unit, is used to store adjustment process data. Specifically, it stores complete data for each compensation adjustment in the format of "time-compensation capacity-adjustment effect", retains 6 months of historical records, and supports exporting data in Excel format, providing data basis for power grid operation and maintenance personnel to analyze the adjustment effect.

[0129] The monitoring frequency synchronization module, as a timing coordination component of the compensation state monitoring unit, is used to synchronize the monitoring frequency and the acquisition frequency. Specifically, it is linked with the acquisition frequency control unit of the voltage and reactive power sensing unit to ensure that the sampling timestamp of the real-time parameter monitoring module is consistent with the acquisition timestamp of the voltage and reactive power sensing unit, thus avoiding errors in the comparison between the monitoring data and the original acquisition data caused by timing deviations.

[0130] Specifically, the real-time parameter monitoring module collects grid parameters after switching at a fixed frequency, and its sampling frequency is synchronized with the front-end sensing unit to avoid data mismatch caused by sampling rate differences. The target value comparison module calculates the collected voltage, power factor and rated values ​​in real time to generate deviation values ​​as a basis for evaluating the adjustment effect. When the deviation exceeds a preset threshold, the deviation signal generation module triggers a feedback signal to the dynamic compensation calculation unit to start the secondary adjustment process. The adjustment data storage module associates and stores the time of each adjustment, compensation capacity and deviation results to form a traceable historical database. The monitoring frequency synchronization module ensures that the sampling time of the monitoring module and the sensing unit are strictly aligned through a clock signal linkage mechanism, eliminating data analysis errors caused by timestamp misalignment.

[0131] The above scheme achieves dynamic tracking and closed-loop control of the compensation and regulation effect, ensuring that grid parameter deviations are identified in a timely manner and trigger secondary regulation, thus avoiding the continuous deterioration of voltage fluctuations caused by monitoring delays in traditional equipment. Structured data storage provides a complete historical record for operation and maintenance analysis, and the synchronous sampling mechanism ensures the temporal consistency between monitoring data and raw data acquisition, effectively improving the accuracy of the regulation strategy.

[0132] Reference Figure 1 and Figure 7 As shown, this application further proposes a multi-dimensional protection unit, including an overvoltage and overcurrent detection module, an equipment overtemperature monitoring module, a compensation module diagnostic module, a protection command execution module, a local alarm drive module, and a protection threshold setting module.

[0133] The overvoltage and overcurrent detection module, as a power grid parameter protection component of the multi-dimensional protection unit, is used to detect the overvoltage and overcurrent status of the power grid. Specifically, it monitors the power grid voltage and compensation circuit current in real time. When the power grid voltage exceeds the rated value by ±10% or the compensation circuit current exceeds the rated value by 1.5 times, it immediately outputs an overvoltage or overcurrent protection trigger signal, providing trigger conditions for the multi-dimensional protection unit to start protection.

[0134] The equipment over-temperature monitoring module, as a temperature protection component of the multi-dimensional protection unit, is used to monitor the temperature of the compensation equipment. Specifically, it collects the temperature of the thyristor and contactor in the hybrid switching actuator through an NTC thermistor. When the temperature exceeds 60°C, it outputs an over-temperature protection trigger signal to prevent the compensation equipment from burning out due to overheating.

[0135] The compensation module diagnostic module, as a fault diagnosis component of the multi-dimensional protection unit, is used to detect faults in the compensation module. Specifically, it adopts the impedance detection method, which detects the impedance value of the capacitor and reactor once per hour. If the impedance value deviates from the rated value by ±20%, the corresponding compensation module is determined to be faulty and the fault number is locked (such as "3# capacitor fault"), and a fault trigger signal is output.

[0136] The protection command execution module, as the protective action execution component of the multi-dimensional protection unit, is used to execute protection commands. Specifically, after receiving the protection trigger signals output by the overvoltage and overcurrent detection module, the equipment overtemperature monitoring module, and the compensation module diagnostic module, it cuts off the power circuit of the hybrid switching actuator within 100ms, and at the same time locks the compensation command receiving function of the dynamic compensation calculation unit to prevent the fault from escalating.

[0137] The local alarm drive module serves as the local alarm component of the multi-dimensional protection unit. It is used to trigger local alarms. Specifically, when the protection command execution unit starts protection, it drives the local audible and visual alarm (red alarm light + buzzer) to operate, and at the same time displays the fault type (such as "overvoltage protection" or "capacitor fault") on the equipment panel, which facilitates on-site maintenance personnel to quickly locate the fault.

[0138] The protection threshold setting module serves as the protection parameter adaptation component for the multi-dimensional protection unit. It is used to adjust the protection thresholds, specifically supporting remote or local operation to adjust the overvoltage, overcurrent, and overtemperature protection thresholds (such as adjusting the voltage protection threshold to ±8%), so that the multi-dimensional protection unit can adapt to power grid scenarios with different stability requirements.

[0139] Specifically, when the grid voltage fluctuation exceeds ±10% of the rated value, the overvoltage and overcurrent detection module outputs a trigger signal to the protection command execution module. The latter cuts off the power circuit of the hybrid switching actuator within 100ms and simultaneously locks the command receiving port of the dynamic compensation calculation unit. The equipment over-temperature monitoring module collects the thyristor heatsink temperature in real time via an NTC thermistor, triggering protection action when the temperature exceeds 60℃. The compensation module diagnostic module injects a 1kHz test signal into the capacitor every hour, determining device faults based on impedance values ​​deviating from the rated value by ±20%. After the protection command execution module activates, the local alarm drive module initiates an audible and visual alarm and displays the fault type code on the panel. The protection threshold setting module allows adjustment of the overvoltage threshold to ±15% or the overcurrent multiple to twice the rated value via the human-machine interface to adapt to different grid scenario requirements.

[0140] The above solution enables rapid isolation between power grid anomalies and equipment failures, effectively preventing equipment damage under overload conditions. The combination of multi-dimensional detection and adjustable threshold settings solves the problem of malfunctions caused by fixed parameters in traditional protection devices. The periodic self-diagnostic function of the impedance detection module avoids the risk of system paralysis caused by latent capacitor faults. The combined output of fault code display and audible / visual alarms significantly shortens the troubleshooting time for on-site maintenance personnel.

[0141] Reference Figure 1 and Figure 8 As shown, this application further proposes a dual power supply unit, including a power switching control module, a battery power monitoring module, and a power voltage regulator module.

[0142] The main power supply module serves as the main power supply access component for the dual power supply unit. It is used to connect to the main power grid, specifically through an air switch to connect to the 380V AC power grid. It has a built-in surge protector with a surge voltage withstand of 10kV to prevent damage to the dual power supply unit and subsequent system modules from lightning strikes or voltage surges, thus providing a stable main power input for the system.

[0143] The backup lithium battery pack, which serves as a backup power component for the dual power supply unit, is used to provide backup power. Specifically, it uses lithium iron phosphate batteries with a specification of 12V / 20Ah, which support deep discharge (discharge depth ≥80%). When fully charged, it can ensure the continuous operation of the voltage and reactive power sensing unit, multi-dimensional protection unit, and intelligent communication interaction unit for 4 hours, providing backup power for the system in case of power failure.

[0144] The power switching control module, which serves as the power switching component of the dual power supply unit, is used to realize the switching between the main and backup power supplies. Specifically, it monitors the mains power voltage through a voltage detection chip with a detection accuracy of ±0.5V. When the mains voltage is lower than 320V or higher than 450V, it automatically switches to the backup lithium battery pack within 50ms, and there is no power interruption gap during the switching process, ensuring uninterrupted power supply to the system.

[0145] The battery power monitoring module serves as a backup power status monitoring component for the dual-power supply unit. It monitors the lithium battery power, specifically by collecting real-time voltage and current data of the backup lithium battery pack and calculating the remaining power (SOC). When the SOC is below 20%, it sends a "low battery" signal to the intelligent communication interface to remind maintenance personnel to charge the lithium battery in a timely manner.

[0146] The power supply voltage regulator module serves as a power stabilization component in the dual-power supply unit. It is used to output a stable power supply, specifically to regulate the AC power input from the grid power access module and the DC power output from the backup lithium battery pack, and output a stable DC power supply of 24V±0.5V to provide stable power supply for each module in the system and avoid voltage fluctuations affecting the operating accuracy of the modules.

[0147] Specifically, when the grid power supply is normal, the main power supply provides power, and the power switching control module continuously monitors the grid voltage. When the voltage exceeds a preset threshold, the switching logic controls a relay or solid-state switch to switch the power supply circuit to the backup lithium battery pack within a very short time. During the switching process, the circuit voltage is maintained stable through energy storage capacitors to avoid power interruption. The battery power monitoring module periodically collects the terminal voltage and discharge current of the lithium battery pack, calculates the remaining power through an integral algorithm, and sends an alarm message to the communication module when the power level falls below a set threshold. The power voltage regulation module rectifies and filters the AC input of the main power supply, and simultaneously steps down the DC output of the lithium battery to ultimately output a stable DC voltage, ensuring that the system continuously obtains clean power during grid voltage fluctuations or power switching.

[0148] The above scheme enables uninterrupted switching of power supply circuits when the power grid is abnormal, avoiding the failure of control functions due to power outages; the real-time monitoring and early warning mechanism of the remaining power of the backup power supply effectively prevents secondary power outage accidents; the voltage stabilization module ensures the stable operation of the system under complex power supply conditions and ensures that the reactive power compensation equipment continues to perform voltage regulation functions during power grid voltage fluctuations.

[0149] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A voltage regulation system for a reactive power compensation device, characterized in that, include: The voltage and reactive power sensing unit is used to collect the three-phase voltage, current and reactive power of the power grid, and provides standardized data after filtering to eliminate harmonics. The load trend prediction unit, based on historical data and real-time change rate analysis, predicts the reactive load fluctuation trend of the power grid in the future and outputs a prediction signal; The dynamic compensation calculation unit is used to calculate the compensation capacity and generate switching commands based on real-time data, prediction signals and grid voltage ratings using the reactive power balance formula. The hybrid switching actuator adopts a hybrid structure of "thyristor pre-switching + contactor stable operation" to achieve switching to low-energy operation of the contactor after inrush-free switching; The compensation status monitoring unit is used to monitor the grid voltage, reactive power factor and operating status of compensation equipment in real time after switching. When a deviation occurs, it generates a feedback signal, triggers secondary regulation and stores the regulation data. The multi-dimensional protection unit integrates overvoltage, overcurrent, overtemperature and module fault diagnosis functions. When the power grid and compensation equipment are abnormal, it cuts off the power supply, locates the fault and sends an alarm signal. The dual power supply unit uses a combination of main power and backup lithium battery for power supply. When the power grid is abnormal, it automatically switches to backup power to maintain the continuous operation of the system. The intelligent communication and interaction unit supports 5G+LoRa dual-mode communication, uploads system operation data to the remote dispatch center and receives manual control commands issued by the dispatch center, realizing two-way interaction between local automatic adjustment and remote monitoring; The load trend prediction unit specifically includes: The historical load database is used to classify and store the reactive load data of the power grid for the past month. It supports automatic cleaning of redundant data from three months ago, provides a data foundation for trend analysis, and has the function of backing up data during power outages. The real-time rate of change calculation module is used to receive real-time current / voltage data from the voltage reactive power sensing unit and calculate the reactive load change rate per minute through a differential algorithm to identify the three states of load: "stable / rising / falling". The prediction algorithm module is used to output the reactive load prediction curve for the next 5 to 15 minutes by using the built-in improved LSTM neural network algorithm, combined with historical data and real-time change rate. The prediction result verification module is used to compare the prediction result with the actual load data of the last three times in the same time period. If the deviation exceeds 10%, the algorithm parameters will be automatically corrected to avoid long-term prediction drift. The prediction duration adjustment module is used to adjust the prediction time range and supports setting three prediction durations through remote or local operation to meet the needs of power grid scenarios. The real-time change rate calculation module uses the following formula to calculate the reactive load change rate: In the formula: R represents the average absolute rate of change; This represents the change in reactive power; Indicates the change over time; i represents the sequence number or index number of the current sampling point; Q i Q represents the reactive power value at the current sampling point i; i-1 This represents the reactive power value of the previous sampling point i-1; n represents the total number of sampling points within the calculation window; τ represents the sampling time interval. The dynamic compensation calculation unit specifically includes: The rated parameter storage module is used to store the rated parameters of the power grid and compensation equipment, and to fix the basic parameters including the rated voltage of the power grid, the target power factor and the rated capacity of the compensation equipment. The parameters can be modified online and automatically backed up. The reactive power balance calculation module is used to calculate the required compensation capacity and, based on the compensation capacity balance calculation formula, combined with the real-time data transmitted by the voltage reactive power sensing unit and the trend signal transmitted by the load trend prediction unit, calculates the capacitive or inductive compensation capacity required by the power grid. The switching status monitoring module is used to obtain the switching status of the compensation equipment. It obtains the current "on" or "off" status of the compensation equipment in real time by detecting the auxiliary contact signal of the contactor, so as to avoid generating duplicate switching or missed switching commands. The operating condition adaptation module is used to automatically adjust the calculation logic of the reactive power balance calculation module to adapt to different operating conditions of the power grid by identifying accident conditions including "voltage drop" and "impact load". The compensation instruction generation module is used to generate specific switching instructions. Specifically, it converts the compensation capacity calculated by the reactive power balance calculation module into switching instructions for specific compensation equipment, and also includes execution timing requirements to provide operation instructions for the hybrid switching actuator. The compensation capacity balance calculation formula of the reactive power balance calculation module is as follows: In the formula: C represents the comprehensive compensation capacity; N represents the length of the real-time measurement data window; M represents the length of the predictive data window; Q mk Q represents the k-th real-time measured reactive power value; pj Q represents the j-th predicted reactive power value; t The target reactive power value is represented by k; the real-time data sampling point number is represented by k; and the predicted data sampling point number is represented by j.

2. The voltage regulation system of the reactive power compensation equipment according to claim 1, characterized in that: The hybrid switching actuator specifically includes: The thyristor drive module is used to drive the thyristor to operate. After receiving the switching command issued by the dynamic compensation calculation unit, it outputs a trigger signal to control the thyristor to conduct, so as to realize the inrush current switching of the compensation equipment. The inrush current multiple is ≤1.2 times the rated current during switching. It is also compatible with two driving voltages, 220V and 380V, to meet the driving requirements of different compensation equipment. The contactor switching module is used to switch to the contactor operation mode. Within 300ms after the thyristor drive module completes the switching, it drives the contactor to engage to replace the thyristor for long-term operation, reducing the power consumption of the hybrid switching actuator to 1 / 5 of that of the thyristor operation mode. The inrush current suppression module is used to suppress the inrush current during switching using a built-in current-limiting resistor. The current-limiting resistor is connected to the circuit at the moment the compensation equipment is switched on, and the current-limiting resistor is short-circuited after the current stabilizes, so as to prevent the inrush current at the moment of switching from damaging the compensation equipment or affecting the stability of the grid voltage. The execution status feedback module is used to detect the current signal of the circuit after switching through the current sensor to determine whether the compensation equipment is normally put into or cut off, and to send the status signal of "successful switching" or "failed switching" back to the dynamic compensation calculation unit and the compensation status monitoring unit to form a closed loop control. The execution timing control module is used to control the timing of switching actions. According to the timing requirements in the instructions of the dynamic compensation calculation unit, it controls the action sequence of the thyristor drive module and the contactor switching module.

3. The voltage regulation system of the reactive power compensation equipment according to claim 2, characterized in that: The inrush current suppression module adopts a two-stage suppression strategy. The first stage limits di / dt through a magnetically saturated reactor, and the second stage suppresses overvoltage through a parallel RC absorption circuit, wherein the resistance value is 10Ω±5% and the capacitance value is 0.1μF±5%.

4. The voltage regulation system of the reactive power compensation equipment according to claim 1, characterized in that: The compensation status monitoring unit specifically includes: The real-time parameter monitoring module is used to synchronously collect grid voltage, current and power factor data after the compensation equipment is switched on to obtain grid parameters after switching. The sampling frequency is consistent with that of the voltage and reactive power sensing unit, providing real-time data for judging the effect of the compensation status monitoring unit. The target value comparison module is used to compare the voltage and power factor collected by the real-time parameter monitoring unit with the rated target values, calculate the voltage deviation value and power factor deviation value, and provide deviation data for the compensation state monitoring unit to judge the adjustment effect. The deviation signal generation module is used to generate a deviation feedback signal. If the deviation value calculated by the target value comparison module exceeds the set threshold, a deviation signal is generated immediately and transmitted to the dynamic compensation calculation unit to trigger secondary adjustment. The adjustment data storage module is used to store adjustment process data. It stores complete data for each compensation adjustment in the format of "time-compensation capacity-adjustment effect", retains 6 months of historical records and supports exporting data in Excel format, providing data basis for power grid operation and maintenance personnel to analyze the adjustment effect; The monitoring frequency synchronization module is used to synchronize the monitoring frequency and the acquisition frequency. It works in conjunction with the acquisition frequency control module of the voltage and reactive power sensing unit to ensure that the sampling timestamp of the real-time parameter monitoring module is consistent with the acquisition timestamp of the voltage and reactive power sensing unit.

5. The voltage regulation system of the reactive power compensation equipment according to claim 1, characterized in that: The multi-dimensional protection unit specifically includes: The overvoltage and overcurrent detection module is used to monitor the grid voltage and compensation circuit current in real time. When the grid voltage exceeds the rated value by ±10% or the compensation circuit current exceeds the rated value by 1.5 times, it immediately outputs an overvoltage or overcurrent protection trigger signal, providing trigger conditions for the multi-dimensional protection unit to start protection. The equipment over-temperature monitoring module is used to monitor and compensate for the temperature of the equipment. It uses an NTC thermistor to collect the temperature of the thyristor and contactor in the hybrid switching actuator. When the temperature exceeds 60°C, it outputs an over-temperature protection trigger signal. The compensation module diagnostic module is used to detect the impedance value of capacitors and reactors once per hour using the impedance detection method. If the impedance value deviates from the rated value by ±20%, the corresponding compensation module is determined to be faulty, the fault number is locked, and a fault trigger signal is output. The protection command execution module is used to receive the protection trigger signals output by the overvoltage and overcurrent detection module, the equipment overtemperature monitoring module, and the compensation module diagnostic module, and then cut off the power circuit of the hybrid switching actuator within 100ms, while locking the compensation command receiving function of the dynamic compensation calculation unit. The local alarm driver module is used to drive the local audible and visual alarm to activate when the protection command execution module starts protection, and at the same time display the fault type on the device panel; The protection threshold setting module is used to adjust the protection threshold. It supports adjusting the overvoltage, overcurrent, and overtemperature protection thresholds through remote or local operation, enabling the multi-dimensional protection unit to adapt to power grid scenarios with different stability requirements.

6. The voltage regulation system of the reactive power compensation equipment according to claim 1, characterized in that: The dual-power supply unit specifically includes: The power switching control module is used to realize the switching between main and backup power. It monitors the grid power voltage through a voltage detection chip with a detection accuracy of ±0.5V. When the grid voltage is lower than 320V or higher than 450V, it automatically switches to the backup lithium battery pack within 50ms. The switching process is seamless to ensure uninterrupted power supply to the system. The battery power monitoring module is used to collect voltage and current data of the backup lithium battery pack in real time to calculate and monitor the remaining power of the lithium battery. When the remaining power is less than 20%, it sends a "low battery" signal to the intelligent communication device to remind maintenance personnel to charge the lithium battery in time. The power supply voltage regulator module is used to regulate the AC power input from the grid power access unit and the DC power output from the backup lithium battery pack, and outputs a stable DC power of 24V±0.5V to provide stable power supply for each module of the system.

Images