Dynamic compensation switching method based on grounding transformer arc suppression coil

Abstract

The invention relates to the technical field of power protection, and discloses a dynamic compensation switching method based on a grounding transformer arc suppression coil. The method comprises the following steps: collecting zero-sequence voltage, zero-sequence current and line impedance of a photovoltaic system in real time through a multi-branch collector to obtain a power grid operation state parameter set; according to the phase relationship between the zero-sequence current and the zero-sequence voltage, identifying a photovoltaic internal fault through a fault direction discrimination algorithm; inputting the fault confirmation signal into an arc suppression coil inductance adjusting system, and dynamically calculating a compensation inductance parameter according to the capacitive current of the fault point; the inductance of the arc suppression coil is adjusted through the combination of the graded reactors controlled by the thyristors, so that the inductive current and the capacitive current are counteracted with each other, and the grounding arc is extinguished. The technical problems that a traditional arc suppression coil cannot distinguish photovoltaic internal and external faults, compensation parameters are fixed and the response speed is low are solved, and the accuracy of ground fault processing and the dynamic response capability of compensation switching are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power protection, and in particular to a grounding transformer-based arc suppression coil dynamic compensation switching method. BACKGROUND

[0002] Traditional photovoltaic power generation system grounding protection mainly relies on fixed parameter arc suppression coils for neutral grounding fault handling. The arc suppression coil compensates for the capacitive ground current in the system by generating inductive current. When a single-phase ground fault occurs, the inductive current and capacitive current are used to realize the natural extinction of the arc according to the principle of mutual offset. The existing arc suppression coil system adopts a preset inductance parameter configuration. The basic parameters of the arc suppression coil are determined according to the rated capacity and operating conditions of the power grid. The arc suppression coil is connected to the neutral point of the power grid through the grounding transformer. In the normal operating state, the arc suppression coil is in standby state. When a ground fault signal is detected, it is put into operation for fault current compensation.

[0003] The arc suppression coil system in the prior art has significant technical deficiencies. The main problem is that the lack of automatic measurement system makes it impossible to measure the grid capacitance current and displacement voltage in real time. When the grid operating mode or grid parameters change, the capacitive current can only be estimated manually, which has a large estimation error and cannot effectively control the residual current and suppress the arc overvoltage in a timely manner. The traditional arc suppression coil adopts a fixed adjustment stage according to the voltage level, which has a small stage number and a large stage current difference, and the compensation accuracy is seriously insufficient. At the same time, the tuning process requires power outage and the arc suppression coil is removed, which loses the continuity of arc compensation and has a slow response speed. In the case of system abnormalities or accidents, the adjustment cannot be made in time, which may easily lead to loss of control. Many operating arc suppression coils have insufficient capacity configuration and can only operate in an under-compensation state for a long time. Due to the lack of damping resistance, a series resonance circuit is formed with the grid capacitance, and when the under-compensation state encounters a grid line breakage fault, voltage resonance may occur, which is more harmful to the insulation of the power system than arc grounding overvoltage.

[0004] The fundamental problem of the existing arc suppression coil technology is the lack of fault direction recognition capability and dynamic parameter adjustment mechanism, which cannot distinguish between internal faults and external faults in the photovoltaic system, resulting in blindness and inefficiency of the compensation strategy. The traditional fixed parameter configuration cannot adapt to the dynamic changes of the operating conditions of the photovoltaic power generation system. In particular, when the grid parameters change due to photovoltaic output fluctuations, the preset arc suppression coil parameters cannot provide optimal compensation effect, thereby affecting the timeliness and accuracy of fault handling. More importantly, the existing technology lacks a fault recognition algorithm for the characteristics of the photovoltaic system, which cannot determine the specific location and nature of the fault according to the phase relationship of the zero sequence electric quantity. This technical limitation directly restricts the optimization of the arc suppression coil compensation effect and the improvement of the safety of the power grid operation. SUMMARY

[0005] The application provides a grounding transformer-based arc suppression coil dynamic compensation switching method, which is used to solve the technical problems that the traditional arc suppression coil cannot distinguish between internal and external faults of a photovoltaic system, the compensation parameters are fixed, and the response speed is slow, and improves the accuracy of ground fault processing and the dynamic response capability of compensation switching.

[0006] In a first aspect, the application provides a grounding transformer-based arc suppression coil dynamic compensation switching method, which comprises: A plurality of branch collectors connected by a total control unit collect and process photovoltaic system zero sequence voltage, zero sequence current, and line impedance in real time to obtain a set of power grid operating state parameters; According to the phase relationship between the zero sequence current and the zero sequence voltage in the set of power grid operating state parameters, a fault direction discrimination algorithm is used to identify internal and external single-phase ground faults to obtain a photovoltaic internal fault confirmation signal; The photovoltaic internal fault confirmation signal is input to an arc suppression coil inductance adjustment system, and the inductance value of the arc suppression coil is dynamically calculated and processed according to the capacitive current at the fault point to obtain compensation inductance parameters; Based on the compensation inductance parameters, a thyristor-controlled hierarchical reactor combination is used to adjust the inductance of the arc suppression coil, so that the inductive current generated by the arc suppression coil and the capacitive current at the fault point are offset, and a grounded arc extinguishing state is obtained.

[0007] Optionally, the real-time collection and processing of photovoltaic system zero sequence voltage, zero sequence current, and line impedance by the plurality of branch collectors connected by the total control unit comprises: The branch 1 collector, the branch 2 collector, and the branch 3 collector are respectively connected to the corresponding photovoltaic branch; Each branch collector is configured with an independent A / D conversion module, and the sampling frequency is set to 10 kHz; The total control unit establishes a communication connection with each branch collector through a grounding transformer controller to form a multi-branch parallel monitoring network, and records the amplitude and phase information of the zero sequence component in real time; a power grid-to-ground capacitance current database is established to store the capacitance current variation law under different operating conditions to obtain a set of power grid operating state parameters.

[0008] Optionally, the process of the fault direction discrimination algorithm for identifying internal and external single-phase ground faults is as follows: The sliding window technique is used to pre-process the zero sequence voltage and zero sequence current in the set of power grid operating state parameters, and each window contains 20 sampling points; The fundamental component is extracted by fast Fourier transform to calculate the phase difference between the zero sequence current and the zero sequence voltage; when the angle by which the zero sequence current leads the zero sequence voltage satisfies the range of 0 degrees to 150 degrees, a positive internal fault identification result of the photovoltaic system is generated; When the angle of the zero sequence current lags behind the zero sequence voltage meets the range of negative 150 degrees to 0 degrees, a reverse external fault recognition result is generated; and the photovoltaic internal fault confirmation signal is output based on the photovoltaic system forward internal fault recognition result.

[0009] Optionally, the process of dynamically calculating the arc suppression coil inductance value according to the fault point capacitive current by the arc suppression coil inductance adjustment system is as follows: The required compensation capacity is calculated according to the maximum active power of the photovoltaic system, the power factor before compensation and the target power factor; The target inductance value of the arc suppression coil is determined by the system angular frequency and the equivalent capacitance value; The constraint condition that the arc suppression coil capacity does not exceed 20% of the rated capacity of the transformer is verified; it is confirmed that the zero sequence voltage drop generated in the transformer by the zero sequence current flowing through the arc suppression coil does not exceed 10% of the rated phase voltage; and the compensation inductance parameter is generated based on the constraint condition.

[0010] Optionally, the process of inductance adjustment processing of the arc suppression coil by the thyristor-controlled grading reactor combination is as follows: The compensation inductance parameter is received by the coarse adjustment reactor group, and the basic inductance value adjustment is completed within 5 ms; The continuous fine adjustment of inductance is realized by the fine adjustment reactor group through PWM control, and the fine adjustment process is completed within 15 ms; The compensation effect verification module completes the compensation effect detection within 10 ms; and the total response time of the coarse adjustment reactor group, the fine adjustment reactor group and the compensation effect verification module is controlled to be within 30 ms; The inductive current generated by the arc suppression coil and the capacitive current at the fault point are offset by inductance adjustment within the total response time.

[0011] Optionally, the cooperative control process of the coarse adjustment reactor group and the fine adjustment reactor group is as follows: The coarse adjustment reactor group includes a plurality of thyristor switching reactors, and the fine adjustment reactor group adopts a controllable saturated reactor; The residual current after compensation is monitored in real time by a closed-loop control system; when the residual current exceeds a preset threshold, the fine adjustment reactor group automatically performs inductance fine adjustment; and the mutual offset effect of the inductive current and the capacitive current is maintained within a set range.

[0012] Optionally, an adjustable damping resistance is connected in series in the arc suppression coil loop for suppressing series resonance overvoltage; the adjustable damping resistance value is determined according to the ratio relationship between inductance and capacitance; and when it is detected that the system enters a resonance state, the adjustable damping resistance is automatically put into operation.

[0013] Optionally, the extinction state of the ground arc is confirmed by the following verification indexes: The fault phase voltage is restored to more than 95% of the rated voltage; The non-fault phase voltage is maintained within 105% of the rated voltage; The deviation of the zero sequence voltage and the zero sequence current product is controlled within 2% of the rated zero sequence voltage.

[0014] In the technical scheme provided in the application, the technical feature of the total control unit connecting the multi-branch collector for real-time acquisition and processing of the zero sequence voltage, the zero sequence current and the line impedance of the photovoltaic system overcomes the technical defect of the traditional arc extinction coil lacking an automatic measurement system, realizes continuous monitoring and data management of the power grid operation state parameters, eliminates the error problem of manual estimation of the capacitive current, and establishes a power grid-to-ground capacitive current database to provide a reliable data basis for dynamic compensation. The technical feature of the fault direction discrimination algorithm for internal and external identification and processing of single-phase ground faults solves the fundamental problem that the traditional technology cannot distinguish the fault location, realizes accurate identification of internal and external faults of the photovoltaic system through analysis of the phase relationship between the zero sequence current and the zero sequence voltage, avoids the blindness of the compensation strategy, and significantly improves the pertinence and effectiveness of fault processing. The technical feature of the arc extinction coil inductance adjustment system dynamically calculating and processing the inductance value according to the capacitive current at the fault point completely changes the limitations of the traditional arc extinction coil fixed parameter configuration, realizes accurate matching of the compensation parameters and the actual fault conditions, solves the technical problems of few stages, large stage difference current and low compensation accuracy, and ensures the optimal correspondence between the compensation capacity and the fault current through real-time calculation. The technical feature of the thyristor-controlled hierarchical reactor combination for inductance adjustment realizes fast response within 30ms, solves the technical problems of power outage and slow response speed of the traditional arc extinction coil tuning, guarantees the accuracy and speed of inductance adjustment through collaborative control of coarse adjustment and fine adjustment, makes the inductive current generated by the arc extinction coil accurately offset the capacitive current at the fault point, and ensures reliable extinguishing of the ground arc.

[0015] The fault direction discrimination algorithm plays a key role in the application of grounding protection in photovoltaic power generation system. The algorithm realizes the accurate analysis of the phase relationship of zero sequence electric quantity by the combination of sliding window technology and fast Fourier transform application. Its core contribution is to establish a fault type recognition mechanism based on phase difference range determination. When the zero sequence current leads the zero sequence voltage by 0 to 150 degrees, it is determined as internal fault of photovoltaic. When the zero sequence current lags the zero sequence voltage by negative 150 to 0 degrees, it is determined as external fault. This algorithm design is specially optimized for the electrical characteristics of photovoltaic system. Compared with the traditional amplitude comparison method, it has higher anti-interference ability and discrimination accuracy. The dynamic inductance calculation algorithm has significant technical advantages in arc suppression coil parameter optimization. The required compensation capacity is determined by real-time calculation of the relationship between the maximum active power of the photovoltaic system and the power factor before and after compensation. The target inductance value is accurately calculated by combining the system angular frequency and equivalent capacitance value. The algorithm considers the engineering constraint conditions such as arc suppression coil capacity constraint and zero sequence voltage drop limit, ensuring the engineering realizability of the calculation result. The contribution of this algorithm in the dynamic compensation application of photovoltaic power generation system is to realize the real-time matching of compensation parameters and system operating state, significantly improving the stability and reliability of compensation effect. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application. Those skilled in the art can obtain other drawings based on these drawings without creative labor.

[0017] Figure 1 An embodiment of the grounding transformer arc suppression coil dynamic compensation switching method in the embodiments of the present application is shown in the figure. Figure 2 The system architecture of the grounding transformer arc suppression coil dynamic compensation switching method in the embodiments of the present application is shown in the figure. DETAILED DESCRIPTION

[0018] The embodiment of the present application provides a grounding transformer arc suppression coil dynamic compensation switching method. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the term "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0019] For ease of understanding, the specific flow of the embodiment of the present application is described below, please refer to Figure 1 One embodiment of the grounding transformer arc suppression coil dynamic compensation switching method in the embodiment of the present application comprises: The multi-branch collector connected through the total control unit collects and processes the zero sequence voltage, zero sequence current and line impedance of the photovoltaic system in real time to obtain a power grid operating state parameter set; According to the phase relationship between the zero sequence current and the zero sequence voltage in the power grid operating state parameter set, the internal and external identification processing of the single-phase ground fault is performed through the fault direction discrimination algorithm to obtain a photovoltaic internal fault confirmation signal; The photovoltaic internal fault confirmation signal is input to the arc suppression coil inductance adjustment system, the arc suppression coil inductance value is dynamically calculated and processed according to the fault point capacitive current to obtain a compensation inductance parameter; Based on the compensation inductance parameter, the inductance adjustment processing of the arc suppression coil is performed through the thyristor-controlled grading reactor combination, so that the inductive current generated by the arc suppression coil and the capacitive current at the fault point are offset to each other to obtain a grounded arc extinguishing state.

[0020] It can be understood that the execution subject of the present application can be a grounding transformer arc suppression coil dynamic compensation switching system, and can also be a terminal or a server, and the specific place is not limited. The embodiment of the present application takes the server as the execution subject for example.

[0021] Specifically, the implementation process of the grounding transformer arc suppression coil dynamic compensation switching method first establishes communication connection with the branch 1 collector, the branch 2 collector and the branch 3 collector through the total control unit, and each collector is configured with an independent A / D conversion module to realize synchronous collection of zero sequence voltage, zero sequence current and line impedance of the photovoltaic system. The total control unit serves as a data aggregation center, receives original electrical parameter data from the three branch collectors, and sets the sampling frequency to 10 kHz to ensure fast response to transient changes in the power grid. Each branch collector obtains the amplitude and phase information of the zero sequence component through a high-precision sensor, forms a data structure containing a time stamp, amplitude and phase angle, and establishes a set of power grid operating state parameters. This parameter set not only records real-time data, but also builds a power grid capacitance current database, stores the capacitance current variation law under different operating conditions, and forms a historical data reference benchmark.

[0022] After receiving the set of power grid operating state parameters, the fault direction discrimination algorithm first uses the sliding window technique to preprocess the zero sequence voltage and zero sequence current, and each sliding window contains a data sequence of 20 consecutive sampling points. The sliding window technique smoothes the data in the time window to eliminate transient noise interference and extract stable electrical parameter features. The fast Fourier transform algorithm converts the time domain data in the window to the frequency domain, separates the fundamental component from the composite signal, and calculates the phase difference between the zero sequence current and the zero sequence voltage. When the phase difference calculation result shows that the angle of the zero sequence current leading the zero sequence voltage is within the range of 0 degrees to 150 degrees, the discrimination algorithm determines that it is a forward internal fault of the photovoltaic system, and generates a forward internal fault identification result of the photovoltaic system. On the contrary, when the angle of the zero sequence current lagging behind the zero sequence voltage is within the range of negative 150 degrees to 0 degrees, it is identified as a reverse external fault. Based on the forward internal fault identification result of the photovoltaic system, the algorithm outputs a photovoltaic internal fault confirmation signal to trigger the subsequent dynamic compensation switching process.

[0023] After receiving the photovoltaic internal fault confirmation signal, the arc suppression coil inductance regulation system immediately starts the inductance value dynamic calculation program. The calculation program first reads three key parameters of the photovoltaic system, namely the maximum active power, the power factor before compensation and the target power factor, and calculates the required reactive power compensation capacity through the power factor compensation formula. The inductance regulation system determines the target inductance value of the arc suppression coil according to the system angular frequency and the equivalent capacitance value using the inductance calculation formula. During the calculation process, the system verifies two important constraints: the capacity of the arc suppression coil must not exceed 20% of the rated capacity of the transformer, and the zero sequence voltage drop generated by the zero sequence current flowing through the arc suppression coil in the transformer must not exceed 10% of the rated phase voltage. After passing the constraint condition verification, the dynamic calculation program outputs the compensation inductance parameter, which includes the target inductance value, the regulation range and the regulation accuracy requirement.

[0024] After receiving the compensation inductance parameters, the thyristor-controlled grading reactor combination starts a three-stage inductance adjustment process. The coarse adjustment reactor group, as the first stage, contains multiple thyristor switching reactors, which adjust the inductance in a large range according to the target value in the compensation inductance parameters, and complete the fast switching of the basic inductance value within a 5 ms time window. The fine adjustment reactor group, as the second stage, uses controllable saturated reactor technology to achieve continuous fine adjustment of the inductance value through PWM pulse width modulation control, and completes the fine adjustment process within 15 ms. The compensation effect verification module, as the third stage, real-time monitors the residual current and voltage recovery after adjustment, and completes the quantitative evaluation of the compensation effect within 10 ms. The coordinated control of the three stages ensures that the total response time is controlled within 30 ms, so that the inductive current generated by the arc suppression coil and the capacitive current at the fault point can offset each other, and finally form an extinguishing state of the ground arc.

[0025] In a specific embodiment, the multi-branch collector connected by the total control unit collects the zero sequence voltage, zero sequence current and line impedance of the photovoltaic system in real time, including: The branch 1 collector, the branch 2 collector and the branch 3 collector are respectively connected to the corresponding photovoltaic branch; Each branch collector is configured with an independent A / D conversion module, and the sampling frequency is set to 10 kHz; The total control unit establishes a communication connection with each branch collector through the grounding transformer controller, forms a multi-branch parallel monitoring network, and records the amplitude and phase information of the zero sequence component in real time; establishes a power grid to ground capacitance current database, stores the capacitance current variation law under different operating conditions, and obtains a set of power grid operating state parameters.

[0026] Specifically, the branch 1 collector is specially responsible for the zero sequence voltage, zero sequence current and line impedance data collection of the first photovoltaic branch, and converts the high-voltage side electrical signal into a low-voltage signal suitable for A / D conversion module processing through the voltage transformer and current transformer. The branch 2 collector and the branch 3 collector use the same signal conditioning circuit, corresponding to the second and third photovoltaic branches respectively, forming a three-way parallel data acquisition architecture. The A / D conversion module built-in each branch collector uses a 16-bit resolution analog-to-digital converter, and the sampling frequency is set to 10 kHz, which means that each electrical parameter is digitally sampled 10000 times per second, so as to capture the fast changing signals in the power grid operation process.

[0027] After the A / D conversion module converts the analog signal into a digital signal, the original sampling data is preliminarily processed by the built-in digital signal processor, including digital filtering, signal amplification and zero drift compensation. The low-pass filter is used in the digital filtering process to eliminate high-frequency noise interference. The signal amplification section adjusts the signal amplitude according to the preset gain coefficient. The zero drift compensation function eliminates the direct current offset error introduced by the sensor and the conversion circuit. The processed digital signal is transmitted to the total control unit through the CAN bus protocol. The CAN bus uses differential signal transmission method, which has strong anti-interference ability and data transmission reliability.

[0028] The total control unit as the data aggregation and processing center, through the grounding transformer controller to establish communication connection with three branch collectors, the grounding transformer controller built-in multi-channel communication interface, support at the same time with multiple collectors for data exchange. The communication protocol uses master-slave mode, the total control unit as the master station periodically sends data request instructions to each branch collector, the collector as a slave station responds to the request and uploads the collected electrical parameter data. Each data packet contains a timestamp, a collector identifier, a parameter type and a value information during data transmission. The timestamp ensures the time synchronization of different branch data, the collector identifier distinguishes the data source, and the parameter type indicates whether it is zero sequence voltage, zero sequence current or line impedance data.

[0029] The establishment of the multi-branch parallel monitoring network is based on real-time data exchange and state synchronization mechanism. The total control unit sends a synchronization signal to all branch collectors every 100 milliseconds to ensure that each collector samples data at the same time, avoiding data analysis errors caused by sampling time differences. The amplitude and phase information of the zero sequence component is extracted through a digital signal processing algorithm. The amplitude calculation uses the root mean square algorithm for statistical analysis of the sampling data, and the phase information is determined by zero-crossing detection and correlation analysis algorithm. The amplitude data is stored in the form of effective value, and the phase data is recorded in the form of angle value. Both of them together constitute the description of the zero sequence component.

[0030] The database establishment process of the grid-to-ground capacitance current includes three links of data classification, storage and indexing. The data classification divides the collected capacitance current data into four states of normal operation, light load operation, heavy load operation and fault operation according to the operating conditions, and the data in each state is stored independently to form a sub-database. The storage structure adopts time series database format, which records the change rule of capacitance current in time sequence and associates the corresponding operating condition parameters, including load level, environmental temperature, humidity and other influencing factors. The indexing mechanism is based on a multi-dimensional indexing system of time, working condition type and parameter range, which supports fast retrieval of historical data under specific conditions.

[0031] The formation process of the power grid operation status parameter set involves correlating real-time acquired data with historical databases. Real-time data directly reflects the current state, while historical data serves as the benchmark for trend analysis and anomaly detection. The parameter set is stored using a multi-dimensional array structure: the first dimension represents the time series, the second represents the branch number, the third represents the parameter type, and the fourth represents the numerical value and status identifier. The status identifier includes data validity, data quality level, and anomaly flags. Data validity indicates whether the acquired data is within the normal range, the data quality level reflects the reliability of the data, and the anomaly flag indicates whether any anomalies have been detected.

[0032] Figure 2 This is a schematic diagram of the system architecture of the dynamic compensation switching method based on the grounding transformer arc suppression coil in the embodiments of this application. Figure 2 As shown, the system architecture includes a grounding transformer, an arc suppression coil, a compensation system, and a multi-branch acquisition network. DL0 represents the main control unit, and DL1, DL2, and DL3 represent branch 1, branch 2, and branch 3 acquisition units, respectively. Each branch acquisition unit establishes a communication connection with the main control unit through the grounding transformer controller, forming a multi-branch parallel monitoring network. The grounding transformer is connected to the neutral point of the power grid through the arc suppression coil. A series resistor R in the arc suppression coil circuit is used for damping control, and capacitor C represents the equivalent capacitance of the power grid to ground. Each branch acquisition unit monitors the zero-sequence voltage, zero-sequence current, and line impedance parameters of the corresponding photovoltaic branch in real time, and transmits the collected data to the main control unit for fault direction determination and compensation parameter calculation, realizing rapid identification and dynamic compensation switching control of single-phase grounding faults in the photovoltaic system.

[0033] In one specific embodiment, the process of the fault direction discrimination algorithm performing internal and external identification processing for single-phase ground faults is as follows: The sliding window technique is used to preprocess the zero-sequence voltage and zero-sequence current in the set of power grid operating state parameters. Each window contains 20 sampling points. The fundamental component is extracted by fast Fourier transform, and the phase difference between zero-sequence current and zero-sequence voltage is calculated. When the angle between the zero-sequence current and the zero-sequence voltage is within the range of 0 to 150 degrees, the positive internal fault identification result of the photovoltaic system is generated. When the angle of the zero-sequence current lagging behind the zero-sequence voltage is within the range of -150 degrees to 0 degrees, a reverse external fault identification result is generated; based on the positive internal fault identification result of the photovoltaic system, a photovoltaic internal fault confirmation signal is output.

[0034] Specifically, the zero-sequence voltage and the zero-sequence current in the power grid operating state parameter set are preprocessed by a sliding window technique. The sliding window technique is a time-domain signal processing method, which continuously moves a fixed-length data window on a time series, and each window contains continuous 20 sampling point data. Since the sampling frequency is 10 kHz, each window corresponds to a time span of 2 milliseconds, and the data in the window represents the instantaneous change characteristics of the zero-sequence voltage and the zero-sequence current in that time period. The moving step of the sliding window is set to 1 sampling point, which means that after each window moves, the oldest 1 data point is discarded, and the latest 1 data point is added, forming a continuous updating data processing process. The data preprocessing process includes two steps of outlier detection and smoothing filtering. Outlier detection identifies abnormal data by comparing the deviation of the current data point from other data points in the window, and smoothing filtering uses a moving average method to eliminate random noise interference for subsequent analysis.

[0035] The fast Fourier transform algorithm receives the preprocessed window data, converts the time-domain signal to the frequency-domain signal for spectral analysis. The core principle of the fast Fourier transform algorithm is based on the fast calculation method of discrete Fourier transform, which divides N-point discrete Fourier transform into multiple smaller-scale transform operations through divide-and-conquer strategy, significantly reducing the computational complexity. The algorithm performs butterfly operation on 20 sampling point zero-sequence voltage data. Butterfly operation is the basic calculation unit of fast Fourier transform, which realizes frequency domain transformation through complex multiplication and addition operations. The transformation result contains the frequency spectrum information of the zero-sequence voltage signal, among which the fundamental component corresponds to the frequency component of power frequency 50 Hz. The algorithm extracts the amplitude and phase information of the fundamental component through the spectral peak detection method. The zero-sequence current data uses the same fast Fourier transform processing flow to obtain the amplitude and phase parameters of the zero-sequence current fundamental component. In the process of extracting the fundamental component, the algorithm performs peak search on the spectral data, identifies the frequency point with the largest amplitude as the fundamental frequency, and the corresponding complex value contains the amplitude and phase information of the fundamental component.

[0036] The phase difference calculation process is based on mathematical operations on complex representations of the fundamental components of zero-sequence current and zero-sequence voltage. The phase angle of the complex number is calculated by the inverse tangent function, and the phase angle of the zero-sequence current is subtracted from the phase angle of the zero-sequence voltage to obtain the phase difference value. The sign and value of the phase difference calculation result directly reflect the leading or lagging relationship of the zero-sequence current relative to the zero-sequence voltage. When the calculated phase difference is positive, it indicates that the zero-sequence current leads the zero-sequence voltage, and when the phase difference is negative, it indicates that the zero-sequence current lags behind the zero-sequence voltage. The algorithm performs angle range judgment on the phase difference value. When the phase difference is within the range of 0 degrees to 150 degrees, the discrimination logic generates a positive internal fault recognition result of the photovoltaic system, which is based on the electrical characteristics that the zero-sequence current leads the zero-sequence voltage when a positive internal fault occurs. When the phase difference is within the range of negative 150 degrees to 0 degrees, the discrimination logic generates a negative external fault recognition result, which corresponds to the electrical phenomenon that the zero-sequence current lags behind the zero-sequence voltage when an external fault occurs.

[0037] The positive internal fault recognition result of the photovoltaic system serves as a trigger condition for subsequent processing. When the algorithm detects the recognition result, it immediately outputs a photovoltaic internal fault confirmation signal. The photovoltaic internal fault confirmation signal is in the form of a digital signal, which includes fault type identification, fault time, and fault branch information. The fault type identification distinguishes between positive internal faults and other fault types, the fault time records the accurate timestamp of detecting the fault, and the fault branch information indicates the specific photovoltaic branch number where the fault occurs. The generation process of the confirmation signal also includes continuity verification. The algorithm requires that the analysis results of three consecutive sliding windows all point to the same fault type before outputting the confirmation signal, in order to avoid false positives caused by transient interference.

[0038] When a positive internal fault occurs in the photovoltaic system, the zero-sequence voltage and the zero-sequence current satisfy the relationship ; wherein represents the zero-sequence voltage, represents the zero-sequence current, represents the zero-sequence impedance of the photovoltaic branch. When a negative external fault occurs in the photovoltaic system, the zero-sequence voltage and the zero-sequence current satisfy the relationship ; wherein represents the zero-sequence impedance of the power supply, represents the zero-sequence impedance of the line. Based on these two mathematical relationships, the phase difference calculation uses complex number operation method to calculate the phase angle difference between the zero-sequence current and the zero-sequence voltage through the function. The phase difference satisfies when a positive internal fault occurs, and satisfies when a negative external fault occurs. The value range of the phase difference directly corresponds to the discrimination result of the fault type.

[0039] In a specific embodiment, the arc suppression coil inductance adjustment system dynamically calculates the inductance value of the arc suppression coil according to the fault point capacitive current, and the process is as follows: According to the maximum active power of the photovoltaic system, the power factor before compensation, and the target power factor, the required compensation capacity is calculated. The target inductance value of the arc suppression coil is determined by the system angular frequency and the equivalent capacitance value. Verify that the arc suppression coil capacity does not exceed 20% of the rated capacity of the transformer; confirm that the zero sequence voltage drop generated by the zero sequence current flowing through the arc suppression coil in the transformer does not exceed 10% of the rated phase voltage; generate compensation inductance parameters based on the constraint condition.

[0040] Specifically, the required compensation capacity is calculated by three key parameters: the maximum active power of the photovoltaic system, the power factor before compensation, and the target power factor. The maximum active power of the photovoltaic system is obtained from the set of grid operating state parameters, which reflects the power output capability of the photovoltaic system under full load operating state, and is collected in real time by the power measurement device and transmitted to the inductance adjustment system. The power factor before compensation is calculated by measuring the ratio of active power and reactive power output by the photovoltaic system, and the value of the power factor directly reflects the balance state of capacitive and inductive loads in the grid. The target power factor is pre-set according to the grid operating requirements, usually set to a value close to 1 to achieve the purpose of power factor correction. The compensation capacity calculation uses the power triangle principle, and the required compensation capacity is determined by calculating the difference between the reactive power before and after compensation, which involves inverse cosine function operation and trigonometric function transformation. After converting the power factor to phase angle, vector operation is performed.

[0041] The determination process of the system angular frequency and the equivalent capacitance value is based on the comprehensive analysis of the basic parameters of the grid and the characteristic parameters of the photovoltaic system. The system angular frequency is equal to 2 times the circumference of the circle multiplied by the power frequency of 50 Hz, resulting in a fixed value of 314 radians per second, which is used as the basic frequency parameter for calculating the inductance of the arc suppression coil. The calculation process of the equivalent capacitance value is more complex, as it needs to consider the parallel effect of the ground capacitance of each branch of the photovoltaic system. The total equivalent capacitance value is obtained by parallel calculation of the ground capacitance values of branch 1, branch 2, and branch 3. The ground capacitance value is measured by a capacitance measurement device under normal operating conditions of the photovoltaic system. The measurement process uses the AC impedance method, which measures the capacitive impedance by applying an AC signal of known frequency and amplitude, and then calculates the capacitance value based on the inverse relationship between impedance and capacitance. The calculation of the target inductance value of the arc suppression coil is based on the series resonance principle. When the inductive reactance of the arc suppression coil is equal to the equivalent capacitive reactance of the system, complete compensation is achieved. The target inductance value is equal to 1 divided by the product of the square of the system angular frequency and the equivalent capacitance value.

[0042] The constraint verification process includes capacity constraint and voltage drop constraint. The capacity constraint verification is achieved by comparing the ratio of arc suppression coil capacity to transformer rated capacity. The arc suppression coil capacity is equal to the product of arc suppression coil inductance, system angular frequency and rated voltage square. The transformer rated capacity is obtained from the transformer nameplate parameters. The ratio result must be less than 20% to meet the constraint condition. The voltage drop constraint verification process is more complex. It needs to calculate the voltage drop generated by the zero sequence current flowing through the arc suppression coil on the transformer zero sequence impedance. The size of the zero sequence current is equal to the fault point capacitive current. The transformer zero sequence impedance is obtained by transformer parameter calculation or measurement. The zero sequence voltage drop is calculated by Ohm's law. The zero sequence current multiplied by the transformer zero sequence impedance obtains the voltage drop value. The ratio of the value to the rated phase voltage must be less than 10% to meet the operation requirements.

[0043] The generation process of compensation inductance parameters comprehensively processes the calculation results and constraint verification results to form a parameter set containing target inductance value, adjustment range and safety margin. The target inductance value is the theoretical calculation result. The adjustment range considers the fluctuation range of grid parameters in actual operation. The safety margin ensures that the arc suppression coil can still work normally under extreme working conditions. The parameter generation process also includes a dynamic adjustment mechanism. When it is detected that the grid parameters change, the inductance adjustment system automatically recalculates the compensation inductance parameters to ensure the continuous effectiveness of the compensation effect.

[0044] It should be noted that the calculation of the target inductance value of the arc suppression coil is based on the series resonance compensation principle. The target inductance value is determined by the formula , wherein represents the target inductance value, represents the system angular frequency, represents the equivalent capacitance value of the grid. The system angular frequency , wherein is 50 Hz, and the equivalent capacitance value is obtained by parallel calculation of the capacitance to ground of each photovoltaic branch. The compensation capacity calculation uses the power factor correction formula , wherein represents the required compensation capacity, represents the maximum active power of the photovoltaic system, represents the power factor angle before compensation, represents the target power factor angle. The calculation of the damping resistance value is based on the critical damping condition. The resistance value is determined by the formula , wherein represents the damping resistance value, L represents the inductance value of the arc suppression coil, and C represents the capacitance value to ground of the grid.

[0045] In a specific embodiment, the process of the thyristor-controlled grading reactor combination for inductance adjustment of the arc suppression coil is as follows: The coarse-adjustment reactor group receives the compensation inductance parameters and completes the adjustment of the basic inductance value within 5ms; The fine-tuning reactor group achieves continuous fine-tuning of the inductance through PWM control, completing the fine-tuning process within 15ms; The compensation effect verification module completes the compensation effect detection within 10ms; the total response time of the coarse adjustment reactor group, the fine adjustment reactor group, and the compensation effect verification module is controlled within 30ms. By adjusting the inductance during the total response time, the inductive current generated by the arc suppression coil and the capacitive current at the fault point are mutually canceled out.

[0046] Specifically, the coarse-adjustment reactor group receives compensation inductance parameters and adjusts the base inductance value. The coarse-adjustment reactor group comprises multiple stages of thyristor-switched reactors, each with a different inductance value. Coarse adjustment control of the inductance value is achieved through binary encoding. The compensation inductance parameters include three key data points: the target inductance value, the current inductance value, and the adjustment step size. The coarse-adjustment controller first calculates the difference between the target and current inductance values, and then determines the reactor combinations to be switched on or off based on the inductance values ​​of each reactor stage. The switching decision employs an optimization algorithm, finding the combination with the smallest difference from the target inductance value by traversing all reactor combination schemes. The switching command is transmitted to each stage of the thyristor switch via digital signals. Upon receiving the switching command, the thyristor switch executes the switching action at the AC voltage zero-crossing point to avoid arcing and inrush current during the switching process. The data flow of the entire coarse-adjustment process includes five stages: parameter reception, difference calculation, combination optimization, command generation, and execution confirmation. Data buffering and state mechanisms ensure the continuity and reliability of data processing between these stages. The 5-millisecond settling time constraint requires the controller to use a high-speed digital signal processor with a clock frequency set above 100 MHz to ensure that complex combinatorial optimization algorithms can complete calculations and output results within the time limit.

[0047] The fine-tuning reactor group receives the coarse-tuning completion signal and starts the continuous fine-tuning process controlled by PWM. PWM is the application of pulse width modulation technology, which realizes the continuous control of the controllable saturated reactor by adjusting the duty cycle of the pulse signal. The inductance value of the controllable saturated reactor has a nonlinear relationship with the control current, and a small change in the control current leads to a continuous change in the inductance value. The PWM controller calculates the required control current value based on the residual error between the target inductance value and the inductance value after coarse tuning. The control current calculation uses a lookup table method and a linear interpolation algorithm. The lookup table method is based on a pre-established table of inductance values and corresponding control currents. The linear interpolation algorithm handles the numerical calculation between table intervals to ensure that the control accuracy reaches the level of one thousandth of the inductance value. The PWM signal generation process includes three steps: carrier signal generation, modulation signal calculation, and pulse width determination. The carrier signal is in the form of a triangular wave with a frequency set to 20 kHz. The modulation signal is calculated in real time based on the control current requirement. The pulse width is determined by comparing the amplitude relationship between the modulation signal and the carrier signal. The data feedback mechanism of the fine-tuning process monitors the adjustment effect in real time through an inductance value measurement device. The measurement device uses an impedance analysis method to apply a test signal with a known frequency and amplitude to the controllable saturated reactor. The inductance value is calculated by measuring the phase and amplitude of the response signal. After comparing the measurement result with the target value, an error signal is generated, which is input to the PWM controller to form a closed-loop control system. The 15 ms fine-tuning time includes multiple iteration adjustment processes, with each iteration period being 3 ms, including measurement, calculation, adjustment, and verification. Five iterations ensure that the inductance value adjustment accuracy meets the compensation requirements.

[0048] The compensation effect verification module starts the rapid detection program after the inductance adjustment is completed. The detection process includes residual current measurement, voltage recovery detection, and phase relationship verification. The residual current measurement obtains the adjusted zero-sequence current value through the zero-sequence current transformer and compares it with the capacitive current before the fault. In the ideal compensation state, the residual current should be close to zero. In actual engineering, the residual current is controlled within five percent of the capacitive current to be considered successful compensation. Voltage recovery detection monitors the changes in the fault phase voltage and non-fault phase voltage. The recovery degree of the fault phase voltage is a direct indicator of the compensation effect, and the stability of the non-fault phase voltage reflects the impact of the compensation process on other parts of the grid. Phase relationship verification analyzes the phase difference change between the adjusted zero-sequence voltage and zero-sequence current. A phase difference close to zero indicates that the inductive current and capacitive current are basically canceled out, and the degree of deviation of the phase difference from zero quantifies the compensation accuracy. The verification algorithm uses a multi-parameter comprehensive evaluation method to calculate the comprehensive evaluation score by weighting the residual current, voltage recovery, and phase relationship according to the weight coefficients. If the score exceeds the pre-set threshold, the compensation is considered successful, and if the score is below the threshold, the secondary adjustment program is triggered. The 10 ms detection time is achieved through parallel processing technology, with the three detection items simultaneously collecting and analyzing data to finally form the comprehensive evaluation result.

[0049] The control mechanism of the total response time uniformly manages the working timing of the three processing modules through a timing coordinator, which receives a photovoltaic internal fault confirmation signal as a starting trigger, and simultaneously sends starting instructions and time synchronization signals to the coarse-tuning reactor group, the fine-tuning reactor group and the compensation effect verification module. Real-time clocks and countdown timers are arranged inside each module to ensure that the respective data processing tasks are completed within the specified time. The timing coordinator continuously monitors the execution status of each module, and when the total execution time approaches the upper limit of 30 milliseconds, it forcibly ends the current adjustment process and outputs the current optimal result. The 30-millisecond time constraint solves the problem of slow response speed of traditional arc suppression coils, and the fast response capability ensures timely compensation after a fault occurs, avoiding greater damage to the power grid caused by the continuous burning of ground arcs.

[0050] In a specific embodiment, the cooperative control process of the coarse-tuning reactor group and the fine-tuning reactor group is as follows: The coarse-tuning reactor group contains multiple thyristor switching reactors, and the fine-tuning reactor group uses controllable saturated reactors. The residual current after compensation is monitored in real time through a closed-loop control system. When the residual current exceeds the preset threshold, the fine-tuning reactor group automatically adjusts the inductance. This ensures that the mutual cancellation effect of inductive current and capacitive current is maintained within the set range.

[0051] Specifically, the cooperative control process of the coarse-tuning reactor group and the fine-tuning reactor group establishes a double-layer inductance adjustment architecture in the grounding transformer arc suppression coil dynamic compensation switching method. The coarse-tuning reactor group contains multiple thyristor switching reactors as the main means of large-range inductance adjustment. Each thyristor switching reactor is connected to a reactor component with different inductance values, and the on and off states of the thyristor are controlled through digital control signals to realize the switching in and out of the reactor. The multiple thyristor switching reactors use binary weight configuration, with the first level corresponding to the basic inductance unit, the second level corresponding to twice the basic inductance unit, the third level corresponding to four times the basic inductance unit, and so on, forming a binary inductance adjustment system. The control logic calculates the required adjustment amount based on the difference between the target inductance value and the current inductance value, and then converts the adjustment amount into binary code. Each binary bit corresponds to the state of a thyristor switching reactor, with high level indicating switching in and low level indicating switching out. The thyristor switching controller generates the corresponding trigger pulse sequence after receiving the binary control code, and the trigger pulse is sent at the zero-crossing point of the alternating voltage, ensuring that the thyristor turns on or off when the current is zero, avoiding arc and impact phenomena during switching.

[0052] The fine-tuning reactor group uses controllable saturated reactor to realize continuous fine-tuning of inductance value. The controllable saturated reactor is composed of a core reactor and a direct current control winding. The size of the direct current passing through the direct current control winding determines the magnetic saturation degree of the core, thereby changing the equivalent inductance value of the reactor. There is a nonlinear relationship between the control current and the inductance value. When the control current is small, the core is in a linear working area, and the inductance value remains a large value. When the control current increases, the core gradually enters the saturation area, and the inductance value decreases accordingly. The calculation of the control current is based on the pre-established inductance-current characteristic curve. The characteristic curve is obtained by experimental measurement, and the corresponding inductance value under different control currents is recorded to form a lookup table data structure. The fine-tuning controller looks up the corresponding control current value according to the residual inductance error after coarse tuning. When the error value is between the table data points, a linear interpolation algorithm is used to calculate the accurate control current value.

[0053] The data processing process of the closed-loop control system monitors the compensation effect in real time through a residual current detection device. The residual current detection device includes a zero-sequence current transformer and a signal conditioning circuit. The zero-sequence current transformer is installed in the arc suppression coil loop to detect the actual current value flowing through the arc suppression coil. The signal conditioning circuit amplifies, filters, and analog-to-digital converts the weak signal output by the transformer to obtain a digitized current value. Residual current calculation is achieved through vector operation, which subtracts the detected inductive current from the known capacitive current to obtain the amplitude and phase information of the residual current. The amplitude of the residual current directly reflects the compensation effect, and the smaller the amplitude, the better the compensation effect. The phase information reflects the nature of the residual current. A leading phase indicates that the capacitive component is dominant, and a lagging phase indicates that the inductive component is dominant.

[0054] The setting of the preset threshold is based on the safe operation requirements of the power grid and the arc suppression effect standard, and is usually set to five percent of the capacitive current amplitude before the fault. When the residual current exceeds the threshold, the automatic fine-tuning program of the fine-tuning reactor group is triggered. The fine-tuning program first analyzes the nature of the residual current to determine whether it is capacitive over-compensation or inductive under-compensation, and then calculates the required inductance adjustment amount. The adjustment amount calculation uses a proportional-integral control algorithm. The proportional term determines the adjustment amplitude according to the current error size, and the integral term determines the adjustment direction according to the historical error accumulation. The two terms form the control output. The control output is converted into the control current increment of the controllable saturated reactor, and the inductance value is fine-tuned by modifying the direct current control current.

[0055] The maintenance process of the mutual offset effect between the inductive current and the capacitive current is achieved by continuous monitoring and dynamic adjustment. The monitoring period is set to one-tenth of the fundamental period of the power grid, i.e., once every 2 milliseconds. Each time the monitoring obtains the current residual current value and compares it with the set range. The set range includes an upper threshold value corresponding to the maximum allowed residual current value and a lower threshold value corresponding to the minimum requirement for compensation accuracy. When the residual current is within the set range, the current adjustment state is maintained. When the residual current exceeds the set range, the corresponding adjustment program is started. The execution priority of the adjustment program is determined according to the degree of deviation. For slight deviation, fine-tune the reactor group for fine-tuning. For severe deviation, start the coarse-tune reactor group and the fine-tune reactor group for joint adjustment.

[0056] In a specific embodiment, a adjustable damping resistance is connected in series in the arc-extinguishing coil loop for suppressing series resonance overvoltage; the adjustable damping resistance value is determined according to the ratio of inductance to capacitance; when it is detected that the system enters the resonance state, the adjustable damping resistance is automatically put into operation.

[0057] The arc extinguishing state is confirmed by the following verification indicators: the fault phase voltage is restored to more than 95% of the rated voltage; the non-fault phase voltage is maintained within 105% of the rated voltage; the deviation of the product of the zero sequence voltage and the zero sequence current is controlled within 2% of the rated zero sequence voltage.

[0058] Specifically, the damping resistance input process is precisely controlled by a resonance detection algorithm and a resistance value calculation program. The resonance detection algorithm continuously monitors the voltage and current waveforms in the arc-extinguishing coil loop, identifies the occurrence of series resonance state through frequency domain analysis, and determines the resonance state when the voltage amplitude abnormally rises and the current phase is close to the voltage phase. The calculation of the adjustable damping resistance value is based on the ratio of inductance to capacitance, and the resistance value is calculated under the critical damping condition, which is equal to the square root of 2 times the ratio of inductance to capacitance. This calculation process needs to obtain the current inductance value and the equivalent capacitance value of the power grid in real time. The inductance value is obtained by an inductance measuring device, which applies a high-frequency test signal to the arc-extinguishing coil and analyzes the response characteristics to obtain the inductance value. The capacitance value is obtained by parallel calculation of the capacitance of each photovoltaic branch to ground. After the resistance value calculation is completed, the adjustable damping resistance controller adjusts the input combination of the resistance array according to the calculation result. The resistance array includes multiple resistance units with different resistance values, and the required resistance value is achieved by controlling the series-parallel combination of each unit through relay switches. When the resonance detection algorithm confirms that the resonance state is detected, the control signal immediately triggers the automatic input program of the damping resistance, and the input process is executed at the current zero-crossing point to avoid the generation of impulse current.

[0059] The verification process of the grounded arc extinguishing state is realized through data acquisition and analysis of three key indicators. The fault phase voltage recovery detection uses a high-precision voltage transformer to monitor the real-time value of the fault phase voltage and compares it with the rated voltage. The voltage recovery rate is calculated by dividing the current voltage amplitude by the rated voltage amplitude. When the percentage reaches or exceeds 95, the fault phase voltage recovery indicator is confirmed to be qualified. The non-fault phase voltage maintenance detection simultaneously monitors the voltage amplitude changes of the remaining two phases to prevent overvoltage effects on the non-fault phase during compensation. The ratio of the non-fault phase voltage to the rated voltage needs to be controlled within the range of 0.95 to 1.05 to be considered qualified. The calculation process of the zero sequence voltage and zero sequence current product deviation is more complex. It requires simultaneous acquisition of the instantaneous values of the zero sequence voltage and zero sequence current and multiplication operation. The product represents the size of the zero sequence power, which should be close to zero under ideal compensation conditions. The deviation is calculated by dividing the difference between the actual zero sequence power and the theoretical zero value by the rated zero sequence voltage to obtain the percentage deviation. When the deviation is within 2%, the zero sequence power indicator is confirmed to be qualified.

[0060] The data processing of the verification indicators uses a multi-channel synchronous acquisition and real-time calculation architecture. The data acquisition channels include the fault phase voltage channel, the non-fault phase voltage channel, the zero sequence voltage channel, and the zero sequence current channel. Each channel uses an independent A / D converter to ensure the synchronization and accuracy of data acquisition. The collected data is first subjected to digital filtering to eliminate high-frequency noise interference, then effective value calculation and phase analysis are performed to obtain the amplitude and phase information of each parameter. The calculation process of the fault phase voltage recovery rate includes three steps: amplitude extraction, reference comparison, and percentage conversion. The amplitude extraction uses the root mean square algorithm to statistically calculate the sampling data. The reference comparison divides the extracted amplitude by the pre-stored rated voltage value. The percentage conversion multiplies the division result by 100 to obtain the recovery rate value. The data processing of the non-fault phase voltage maintenance detection uses a range judgment algorithm. The algorithm first calculates the ratio of the non-fault phase voltage to the rated voltage, then determines whether the ratio falls within the pre-set qualified range. If it exceeds the range, an unqualified signal is output and the deviation value is recorded. The calculation of the zero sequence power deviation uses an instantaneous power integration algorithm. The algorithm integrates the product of the zero sequence voltage and zero sequence current over time to obtain the average power value. The difference between the average power and zero value is the power deviation. The ratio of the deviation value to the rated zero sequence voltage forms the final percentage deviation indicator.

[0061] The above examples are only used to illustrate the technical solutions of the present application, and not to limit them. Although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that they can modify the technical solutions described in the foregoing examples, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A grounding transformer-based arc suppression coil dynamic compensation switching method, characterized in that, The method comprises: The multi-branch collector connected by the total control unit collects and processes the zero sequence voltage, zero sequence current and line impedance of the photovoltaic system in real time to obtain a set of grid operating state parameters; According to the phase relationship between the zero sequence current and the zero sequence voltage in the set of grid operating state parameters, a fault direction discrimination algorithm is used to identify the internal and external single-phase grounding faults to obtain a photovoltaic internal fault confirmation signal; The photovoltaic internal fault confirmation signal is input to an arc suppression coil inductance adjustment system, and the inductance value of the arc suppression coil is dynamically calculated and processed according to the fault point capacitive current to obtain a compensation inductance parameter; Based on the compensation inductance parameter, a thyristor-controlled hierarchical reactor combination is used to adjust the inductance of the arc suppression coil, so that the inductive current generated by the arc suppression coil and the capacitive current at the fault point are offset to each other to extinguish the grounding arc.

2. The method of claim 1, wherein, The multi-branch collector connected by the total control unit collects and processes the zero sequence voltage, zero sequence current and line impedance of the photovoltaic system in real time, comprising: The branch 1 collector, the branch 2 collector and the branch 3 collector are connected to the corresponding photovoltaic branch respectively; Each branch collector is configured with an independent A / D conversion module, and the sampling frequency is set to 10 kHz; The total control unit establishes a communication connection with each branch collector through a grounding transformer controller to form a multi-branch parallel monitoring network, and records the amplitude and phase information of the zero sequence component in real time; a grid-to-ground capacitance current database is established to store the variation law of the capacitance current under different operating conditions to obtain a set of grid operating state parameters.

3. The method of claim 1, wherein, The process of the fault direction discrimination algorithm for identifying the internal and external single-phase grounding faults is as follows: The sliding window technology is used to pre-process the zero sequence voltage and zero sequence current in the set of grid operating state parameters, and each window contains 20 sampling points; The fundamental component is extracted by fast Fourier transform to calculate the phase difference between the zero sequence current and the zero sequence voltage; when the angle of the zero sequence current leading the zero sequence voltage satisfies the range of 0 degrees to 150 degrees, a positive internal fault identification result of the photovoltaic system is generated; When the angle of the zero sequence current lagging the zero sequence voltage satisfies the range of negative 150 degrees to 0 degrees, a reverse external fault identification result is generated; the photovoltaic internal fault confirmation signal is output based on the positive internal fault identification result of the photovoltaic system.

4. The method of claim 1, wherein, The process of the arc suppression coil inductance adjustment system dynamically calculating and processing the inductance value of the arc suppression coil according to the fault point capacitive current is as follows: The required compensation capacity is calculated according to the maximum active power of the photovoltaic system, the power factor before compensation and the target power factor; The target inductance value of the arc suppression coil is determined by the system angular frequency and the equivalent capacitance value; It is verified that the arc suppression coil capacity does not exceed 20% of the rated capacity of the transformer; It is confirmed that the zero sequence voltage drop generated by the zero sequence current flowing through the arc suppression coil in the transformer does not exceed 10% of the rated phase voltage; the compensation inductance parameter is generated based on the constraint condition.

5. The method of claim 1, wherein, The process of the thyristor-controlled hierarchical reactor combination adjusting the inductance of the arc suppression coil is as follows: The coarse adjustment reactor group receives the compensation inductance parameter and completes the basic inductance value adjustment within 5 ms; The fine-tuning reactor group realizes continuous fine-tuning of inductance through PWM control, and completes the fine-tuning process within 15 ms; The compensation effect verification module completes compensation effect detection within 10 ms; the total response time of the coarse-tuning reactor group, the fine-tuning reactor group and the compensation effect verification module is controlled to be within 30 ms; The inductive current generated by the arc-extinguishing coil and the capacitive current at the fault point are offset through inductance adjustment within the total response time.

6. The method of claim 5, wherein, The cooperative control process of the coarse-tuning reactor group and the fine-tuning reactor group is as follows: The coarse-tuning reactor group comprises a plurality of thyristor switching reactors, and the fine-tuning reactor group adopts a controllable saturated reactor; The residual current after compensation is monitored in real time through a closed-loop control system; when the residual current exceeds a preset threshold, the fine-tuning reactor group automatically performs inductance fine-tuning; and the mutual offset effect of the inductive current and the capacitive current is ensured to be maintained within a set range.

7. The method of claim 1, wherein, Further comprising: A tunable damping resistor is connected in series in the arc-extinguishing coil circuit, for suppressing series resonance overvoltage; the value of the tunable damping resistor is determined according to the ratio relationship between inductance and capacitance; when it is detected that the system enters a resonance state, the tunable damping resistor is automatically put into operation.

8. The method of claim 1, wherein, The extinguishing state of the ground arc is confirmed by the following verification indexes: The fault-phase voltage is restored to more than 95% of the rated voltage; The non-fault-phase voltage is maintained within 105% of the rated voltage; The deviation of the product of the zero-sequence voltage and the zero-sequence current is controlled to be within 2% of the rated zero-sequence voltage.

Images