Photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion
By dynamically calculating the adaptive disturbance amplitude through multi-source information fusion technology, the problem of detection blind spots and power quality deterioration in complex power grid environments by traditional island detection algorithms is solved, and the safe and stable operation of photovoltaic grid-connected systems is realized.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XI AN ZFE TECH CO LTD
- Filing Date
- 2026-05-25
- Publication Date
- 2026-06-23
Smart Images
Figure CN122260040A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of islanding detection technology, specifically to a photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion. Background Technology
[0002] With the large-scale grid connection of photovoltaic (PV) power generation, the safe and stable operation of PV grid-connected systems is crucial, and anti-islanding protection is a key link in ensuring this safety. The islanding effect refers to the situation where, when the grid is interrupted due to faults or maintenance, the PV grid-connected system fails to detect the power outage in time and continues to supply power to surrounding loads, forming a power island. This not only poses a serious threat to the personal safety of maintenance personnel but may also damage electrical equipment due to voltage and frequency fluctuations. Accurate and timely processing of anti-islanding data, quickly determining whether the islanding effect has occurred, and taking corresponding measures are the core means to avoid these hazards. However, the operating environment of PV grid-connected systems is complex, affected by multiple factors such as sunlight and temperature, making it difficult for a single data source to comprehensively and accurately reflect the system status. Therefore, using multi-source information fusion technology to comprehensively process PV grid-connected anti-islanding data, improving the accuracy and reliability of data processing, is of paramount practical significance for ensuring the safe and stable operation of PV grid-connected systems.
[0003] Traditional islanding detection algorithms often employ fixed-parameter active disturbance methods, such as the classic Sandia frequency offset method. The core of this method is to inject a disturbance signal of a specific frequency or amplitude into the power grid to disrupt the power balance in islanded states, thereby triggering protection devices. However, traditional algorithms have significant drawbacks: Firstly, fixed disturbance amplitudes (disturbance coefficients) are difficult to adapt to complex and variable power grid environments. In strong power grids, small disturbances may be "clamped" by the grid and fail to effectively detect islanding; in weak power grids or power matching scenarios, insufficient disturbances can lead to detection blind spots. Secondly, over-reliance on active disturbances can degrade power quality, especially in harmonic and flicker-sensitive scenarios, potentially causing harmonic exceedances, voltage fluctuations, and even interfering with the normal operation of user equipment. Summary of the Invention
[0004] To address the aforementioned technical problems, the present invention aims to provide a photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion. The specific technical solution adopted is as follows: One embodiment of the present invention provides a photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion, the method comprising: The basic disturbance amplitude at the current moment is obtained by using the preset minimum disturbance amplitude of the strong network, the preset maximum disturbance amplitude of the weak network, the reference short-circuit capacity, and the actual short-circuit capacity at the current moment; The average harmonic voltage effective value change rate at the current moment is obtained by comparing the effective values of each harmonic voltage at the current moment with those at the previous moment; the flicker value set at the current moment is formed by combining the flicker values at each moment within a preset time length before the current moment; the relaxation factor at the current moment is obtained based on the THD value, the flicker value set, the average harmonic voltage effective value change rate, and the fundamental voltage effective value at the current moment. The tightening factor for the current moment is obtained based on the current photovoltaic output active power, local load active power consumption, system frequency change rate, and the number of photovoltaic inverters operating in parallel. The adaptive perturbation amplitude at the current moment is calculated using the base perturbation amplitude, relaxation factor, and tightening factor; the islanding effect is detected by combining the adaptive perturbation amplitude at the current moment with the active perturbation method.
[0005] Preferably, the basic disturbance amplitude at the current moment is obtained by using a preset minimum disturbance amplitude for a strong network, a preset maximum disturbance amplitude for a weak network, a reference short-circuit capacity, and the actual short-circuit capacity at the current moment, including: A first mapping value is obtained by negatively mapping the ratio of the actual short-circuit capacity to the reference short-circuit capacity at the current moment using an exponential function with the natural constant as the base; the difference between the preset maximum disturbance amplitude of the weak network and the preset minimum disturbance amplitude of the strong network is obtained and multiplied by the first mapping value to obtain the multiplication result; the multiplication result is added to the preset minimum disturbance amplitude of the strong network to obtain the basic disturbance amplitude at the current moment.
[0006] Preferably, the average harmonic voltage effective value change rate at the current moment is obtained based on the change of the effective value of each harmonic voltage at the current moment compared with the effective value of each harmonic voltage at the previous moment, including: The rate of change of the effective value of the first harmonic voltage at the current moment is obtained by dividing the absolute value of the difference between the effective value of the first harmonic voltage at the current moment and the effective value of the same harmonic voltage at the previous moment by the time interval between the current moment and the previous moment. The average rate of change of the effective value of each harmonic voltage at the current moment is obtained by calculating the average rate of change of the effective value of the current harmonic voltage.
[0007] Preferably, the relaxation factor for the current moment is obtained based on the current THD value, flicker value set, average harmonic voltage RMS change rate, and fundamental voltage RMS value, including: The THD margin at the current moment is obtained by dividing the difference between the national standard limit for THD and the current THD value by the national standard limit for THD. The flicker margin at the current moment is obtained by dividing the difference between the national standard limit for flicker and the current flicker value by the national standard limit for flicker. The characteristic value of the change in the effective value of harmonic voltage at the current moment is obtained by subtracting the ratio of the current moment's average effective value change rate to the maximum value among all times' average effective value change rates of harmonic voltage. The coefficient of variation of the flicker value set at the current moment is negatively correlated using an exponential function with the natural constant as the base to obtain the flicker stationarity characteristic value. The harmonic characteristic value is obtained by normalizing the ratio of the sum of the square roots of the squares of the effective values of all harmonic voltages at the current moment to the effective value of the fundamental voltage at the current moment by the first preset value. The relaxation factor at the current moment is obtained by multiplying the sum of the THD margin and the flicker margin, the sum of the characteristic value of the change in the effective value of harmonic voltage and the flicker stationarity characteristic value, and the harmonic characteristic value.
[0008] Preferably, the tightening factor for the current moment is obtained based on the current photovoltaic output active power, the local load active power consumption, the system frequency change rate, and the number of photovoltaic inverters operating in parallel, including: The power characteristic value is obtained by dividing the difference between the current photovoltaic output active power and the local load active power by the rated power and taking the absolute value; the system frequency characteristic value is obtained by normalizing the sum of the current system frequency change rate and the hyperparameters; the parallel number characteristic value is obtained by subtracting the ratio of the current number of parallel photovoltaic inverters to the parallel number threshold from the first preset value; the tightening factor at the current moment is obtained by obtaining the sum of the power characteristic value and the system frequency characteristic value and multiplying it by the parallel number characteristic value.
[0009] Preferably, the method for obtaining the system frequency change rate is as follows: The rate of change of the system frequency at the current moment is obtained by dividing the difference between the system frequency at the current moment and the system frequency at the previous moment by the time interval between the two moments and taking the absolute value.
[0010] Preferably, the adaptive perturbation amplitude at the current moment is calculated using the base perturbation amplitude, the relaxation factor, and the tightening factor at the current moment, including: The relaxation adjustment value is obtained by multiplying the current basic disturbance amplitude by the sum of the first preset value and the normalized value of the relaxation factor at the current time; the tightening adjustment value is obtained by multiplying the current basic disturbance amplitude by the difference between the first preset value and the normalized value of the tightening factor at the current time; the adaptive disturbance amplitude at the current time is obtained by calculating the mean of the relaxation adjustment value and the tightening adjustment value.
[0011] The embodiments of the present invention have at least the following beneficial effects: This application utilizes a preset minimum disturbance amplitude for a strong grid, a preset maximum disturbance amplitude for a weak grid, a reference short-circuit capacity, and the actual short-circuit capacity at the current moment to obtain the basic disturbance amplitude at the current moment. The basic disturbance amplitude is determined based on the short-circuit capacity, enabling the initial disturbance amplitude to be reasonably set according to the actual strength of the grid, thereby improving the rationality of disturbance amplitude adjustment. Then, based on the current THD value, flicker set, average harmonic voltage effective value change rate, and fundamental voltage effective value, the relaxation factor at the current moment is obtained. Based on the current photovoltaic output active power and local load consumption... The tightening factor is obtained at the current moment by measuring power output, system frequency change rate, and the number of photovoltaic inverters operating in parallel. Then, based on the current moment's base disturbance amplitude, relaxation factor, and tightening factor, the adaptive disturbance amplitude is calculated. This deeply integrates power quality parameters with system operating status to obtain the "relaxation factor" and "tightening factor," enabling bidirectional real-time adjustment of the disturbance coefficient. This effectively balances the sensitivity of islanding detection with the compliance of power quality, while eliminating the detection blind spot of traditional active disturbance methods in complex power grid scenarios, reducing the risk of false tripping, and providing a better technical path for the safe and stable operation of photovoltaic grid-connected systems. Attached Figure Description
[0012] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 The present invention provides a flowchart of a photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion. Detailed Implementation
[0014] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0016] The following description, in conjunction with the accompanying drawings, details a specific scheme for an adaptive anti-islanding data processing method for photovoltaic grid connection based on multi-source information fusion provided by the present invention.
[0017] In this embodiment, the main application scenario of the present invention is as follows: Based on various data of photovoltaic grid connection, this application performs corresponding analysis to adaptively obtain the disturbance coefficient in the traditional active disturbance method, realizes bidirectional real-time adjustment of the disturbance coefficient, and then determines whether islanding has occurred.
[0018] Please see Figure 1 The diagram illustrates a flowchart of a photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion, provided by an embodiment of the present invention. The method includes the following steps: Step S1: Obtain the basic disturbance amplitude at the current moment using the preset minimum disturbance amplitude of the strong network, the preset maximum disturbance amplitude of the weak network, the reference short-circuit capacity, and the actual short-circuit capacity at the current moment.
[0019] Before adaptively acquiring the perturbation amplitude in the traditional active perturbation method, it is necessary to first obtain the relevant data required for the calculation, and then obtain the relevant data required for subsequent calculations.
[0020] Preset minimum disturbance amplitude for strong power grid: The lower limit of disturbance to avoid harmonic pollution under strong power grid (e.g., 0.5%), which is set according to the power grid harmonic standard (e.g., IEEE 519) or the equipment tolerance capability, and is usually provided by the power grid company or equipment manufacturer; Preset maximum disturbance amplitude in weak network: The upper limit of disturbance that ensures effective detection under weak network conditions (e.g., 3%), determined through experiments or simulations, requires a balance between detection sensitivity and system stability, and should refer to the experience values of similar projects. Actual short-circuit capacity at each moment: Based on system parameters (power supply capacity, line impedance, transformer parameters), an equivalent circuit is established, and the short-circuit current is calculated using the per-unit method, then... Therefore, the actual short-circuit capacity is equal to × Rated voltage × Actual current, collected in seconds; Reference short-circuit capacity: Obtain typical values directly from power grid planning data, or estimate based on system parameters, assuming an infinite power source (with zero impedance), and take the rated capacity of the power source as a reference; The fundamental voltage RMS value and the RMS values of each harmonic voltage are collected in seconds by a power quality monitoring device. Flicker value: also known as short-term flicker value, is collected by a flicker meter at a statistical cycle of 10 minutes (directly read from inside the device), and is collected at the second level; Rated voltage: Static acquisition via system parameter configuration table (e.g., 220V / 380V); Photovoltaic output active power and local load consumption active power: collected in seconds via power transmitter; System frequency: acquired in seconds via a frequency meter; Number of photovoltaic inverters operating in parallel: Data is collected in seconds via the communication management unit (obtained through protocol).
[0021] In power grid operation, active disturbance injection is crucial for realizing functions such as islanding detection, and the key is to set the disturbance amplitude appropriately. Different short-circuit capacities of power grids result in different degrees of "rigidity", which determines the initial setting of the basic disturbance amplitude and provides a benchmark reference for the disturbance amplitude.
[0022] The short-circuit capacity of a power grid is a key indicator of its "rigidity" and directly impacts the effectiveness of active disturbance injection. In a strong power grid environment, a larger short-circuit capacity means a smaller equivalent impedance and stronger disturbance suppression capability. Even with a large active disturbance, it's difficult to generate significant changes in electrical quantities within the grid, hindering functions like islanding detection and potentially increasing harmonic pollution, thus affecting power quality. Conversely, in a weak power grid, a smaller short-circuit capacity and a larger equivalent impedance make it more sensitive to disturbances. Smaller disturbances can produce significant effects, improving detection sensitivity. Therefore, determining the base disturbance amplitude based on the short-circuit capacity allows for the reasonable setting of the initial disturbance amplitude according to the actual strength of the power grid. This provides a benchmark reference consistent with grid characteristics for subsequent adaptive adjustments, ensuring the entire adaptive process starts from a point aligned with the grid's basic conditions.
[0023] Therefore, the basic disturbance amplitude at the current moment is obtained by using the preset minimum disturbance amplitude of the strong network, the preset maximum disturbance amplitude of the weak network, the reference short-circuit capacity, and the actual short-circuit capacity at the current moment.
[0024] Specifically, a first mapping value is obtained by negatively mapping the ratio of the actual short-circuit capacity to the reference short-circuit capacity at the current moment using an exponential function with the natural constant as the base; the difference between the preset maximum disturbance amplitude of the weak network and the preset minimum disturbance amplitude of the strong network is obtained and multiplied by the first mapping value to obtain a multiplication result; the multiplication result is added to the preset minimum disturbance amplitude of the strong network to obtain the basic disturbance amplitude at the current moment.
[0025] The specific calculation model for the basic disturbance amplitude is as follows: , Where K0i represents the basic disturbance amplitude at time i (the current time); Kmin and Kmax are the preset minimum disturbance amplitude (0.5%) for a strong network and the preset maximum disturbance amplitude (2%) for a weak network, respectively; Sci represents the actual short-circuit capacity at the current time; S0 represents the reference short-circuit capacity; and e represents the natural constant. The first mapping value indicates that when the Sci of a strong power grid is large, the exponential term approaches 0, the basic disturbance amplitude approaches its minimum, and ineffective disturbances are suppressed. When the Sci of a weak power grid is small, the exponential term approaches 1, the basic disturbance amplitude approaches its maximum, and detection sensitivity is improved. It should be noted that "active disturbance method" is a general term representing various active disturbance methods (Active Frequency Shift (AFD), Sliding Mode Frequency Shift (SMS), Sandia Frequency Shift method, etc.). Different methods result in different basic disturbance amplitudes, and the settings of Kmin and Kmax also differ.
[0026] Step S2: Obtain the average harmonic voltage effective value change rate at the current moment based on the change of each harmonic voltage effective value at the current moment compared to the previous moment; assemble the flicker value set at the current moment by including the flicker values at each moment within a preset time length before the current moment; obtain the relaxation factor at the current moment based on the THD value, flicker value set, average harmonic voltage effective value change rate, and fundamental voltage effective value.
[0027] The above obtains the basic disturbance amplitude at the current moment. However, relying solely on the basic disturbance amplitude is far from sufficient. Power quality and system operating status have complex and contradictory requirements for the disturbance amplitude. Power quality has a certain tolerance margin, allowing for appropriate relaxation of the disturbance amplitude. System operating status, on the other hand, is related to the effectiveness and risk of islanding detection, and the disturbance amplitude needs to be tightened when necessary. Therefore, dynamically correcting the basic disturbance amplitude by comprehensively considering the basic disturbance amplitude and the relaxation and tightening factors obtained from power quality and system operating status to obtain the final adaptive disturbance amplitude has become an inevitable choice to ensure the safe, stable, and efficient operation of the power grid.
[0028] There is a contradictory yet unified relationship between power quality and system operating status's demands for active disturbances. Obtaining a relaxation factor from a power quality perspective is because, under the premise of ensuring power quality compliance, the system has a certain "tolerance margin" that allows for appropriately amplified disturbance amplitudes. When the total harmonic distortion (THD) margin and flicker margin are large, it indicates that the system has a strong capacity to withstand additional harmonics and frequency fluctuations. In this case, the disturbance amplitude can be appropriately relaxed to improve the efficiency of functions such as islanding detection. Conversely, obtaining a tightening factor from a system operating status perspective is because the system operating status directly affects the effectiveness of islanding detection and system risk. When the power imbalance is small, the frequency change rate is large, and the islanding risk level is high, the difficulty of islanding detection increases, and the system faces greater risks. In this case, the disturbance amplitude must be tightened to ensure timely and reliable islanding detection and guarantee the safe and stable operation of the system. Obtaining factors from both perspectives allows for a comprehensive and integrated consideration of various factors' demands on the disturbance amplitude, making the adjustment of the disturbance amplitude more scientific and reasonable.
[0029] The logic for obtaining the relaxation factor focuses on power quality margin. First, it uses the real-time margins of total harmonic distortion (THD) and flicker as the core to directly quantify the system's current remaining harmonic and fluctuation tolerance capacity; a larger margin allows for greater permissible disturbance amplification. Second, it introduces two dynamic features: the rate of change of the effective value of harmonic voltage and the flicker trend, to capture the deterioration trend of power quality. If the rate of change is small or the fluctuation is stable, the disturbance amplitude can be relaxed to improve adaptability to transient processes. Finally, it considers the masking effect of high-frequency background noise; if the noise level is low, the restrictions are relaxed. These features are normalized and weighted to form a relaxation factor that reflects the power quality safety boundary, ensuring that the disturbance amplitude is dynamically adjusted within compliance requirements.
[0030] Therefore, it is necessary to obtain the average rate of change of the effective value of the harmonic voltage at the current moment based on the change of the effective value of each harmonic voltage at the current moment compared with the effective value of each harmonic voltage at the previous moment.
[0031] Specifically, the rate of change of the effective value of the first harmonic voltage at the current moment is obtained by dividing the absolute value of the difference between the effective value of the first harmonic voltage at the current moment and the effective value of the same harmonic voltage at the previous moment by the time interval between the current moment and the previous moment. The average rate of change of the effective value of the effective value of each harmonic voltage at the current moment is then calculated to obtain the average rate of change of the effective value of the effective value of the harmonic voltage at the current moment. For example, the rate of change of the effective value of the nth harmonic voltage at the current moment can be obtained by dividing the absolute value of the difference between the effective value of the nth harmonic voltage at the current moment and the effective value of the nth harmonic voltage at the previous moment by the time interval between the current moment and the previous moment.
[0032] Furthermore, the trend change of the flicker value needs to be combined with the flicker values of other times in a local area at a given moment. Thus, the flicker values of each moment within a preset time length before the current moment, including the current moment, are combined to form the flicker value set of the current moment. Preferably, in this application, the preset time length is 1 minute, that is, the flicker values of each moment within the most recent 1 minute including the current moment are combined to form the flicker value set of the current moment.
[0033] Furthermore, the relaxation factor for the current moment is obtained based on the current THD value, flicker value set, average harmonic voltage effective value change rate, and fundamental voltage effective value.
[0034] Specifically, the THD margin at the current moment is obtained by dividing the difference between the national standard limit for THD and the current THD value by the national standard limit for THD; the flicker margin at the current moment is obtained by dividing the difference between the national standard limit for flicker and the flicker value at the current moment by the national standard limit for flicker; the characteristic value of the change in the effective value of harmonic voltage at the current moment is obtained by subtracting the ratio of the current moment's average effective value change rate to the maximum value among all moments' average effective value change rates from the first preset value; the coefficient of variation of the flicker value set at the current moment is negatively correlated using an exponential function with the natural constant as the base to obtain the flicker stationarity characteristic value; the harmonic characteristic value is obtained by normalizing the ratio of the sum of the square roots of the squares of the effective values of all harmonic voltages at the current moment to the effective value of the fundamental voltage at the current moment from the first preset value; and the relaxation factor at the current moment is obtained by multiplying the sum of the THD margin and the flicker margin, the sum of the characteristic value of the change in the effective value of harmonic voltage and the flicker stationarity characteristic value, and the harmonic characteristic value.
[0035] The specific calculation model for the relaxation factor is as follows: , Where Ai represents the relaxation factor at time i (the current time); Tl represents the national standard limit for THD (usually 5%), exceeding which results in substandard power quality; the acquisition of THD values is based on existing technology and will not be elaborated further here; Ti represents the THD value at the current time. Pl represents the THD margin, i.e., the remaining available harmonic space; Pl represents the national standard limit for flicker (1.0), and Pi represents the flicker value at the current moment. This represents the flicker margin, i.e., the remaining available flicker space; Dti is the rate of change of the average effective value of harmonic voltage at the current moment, and Dtm is the maximum value among the rates of change of the average effective value of harmonic voltage at all moments (used for normalization, eliminating dimensions). The first preset value is 1. The value represents the characteristic value of the effective value change of harmonic voltage, indicating that the smaller the rate of change of the average effective value of harmonic voltage at the current moment, the more relaxed the disturbance amplitude should be; e is the natural constant. Tpi is the coefficient of variation of the flicker value set at the current moment, and Tpi is the standard deviation of the flicker values in the flicker value set at the current moment (reflecting the degree of fluctuation). This represents the mean of the flicker values in the set of flicker values at the current moment. The flicker stability characteristic value is represented by the coefficient of variation of the flicker value set at the current moment. The smaller the coefficient of variation, the more stable the flicker, and the wider the disturbance amplitude should be. Uhi represents the effective value of the h-th harmonic voltage at the current moment, where h ranges from the starting number of the high-frequency harmonic (e.g., 20th) to the ending number of the high-frequency harmonic (e.g., 50th). U represents the effective value of the fundamental voltage at the current moment, and norm is the normalization function. This is a harmonic characteristic value, indicating that the lower the background noise level, the more the disturbance should be relaxed. Thus, the relaxation factor for the current moment can be obtained.
[0036] Step S3: Obtain the tightening factor at the current moment based on the photovoltaic output active power, local load active power consumption, system frequency change rate, and the number of photovoltaic inverters operating in parallel.
[0037] The logic for obtaining the tightening factor revolves around the reliability of island detection and system risk. First, power imbalance and frequency change rate are key factors. Lower system power matching simplifies island detection, thus the disturbance amplitude should be tightened to avoid excessive disturbance. Simultaneously, a smaller absolute value of the system frequency change rate indicates greater system stability, allowing for smaller-scale tightening of the disturbance amplitude for protection. Second, the superposition effect of multiple parallel units must be considered; a larger number of parallel units results in smaller disturbance amplitude per unit, preventing the total disturbance from exceeding the limit. These features are normalized and weighted to form a tightening factor reflecting the urgency of detection, ensuring dynamic tightening of the disturbance amplitude within a controllable risk range. The two factors are dynamically adjusted from opposing perspectives of "relaxation" and "tightening," ultimately achieving an adaptive balance of the disturbance coefficient through fusion.
[0038] Therefore, the tightening factor at the current moment is obtained based on the photovoltaic output active power, the local load active power consumption, the system frequency change rate, and the number of photovoltaic inverters operating in parallel.
[0039] Specifically, the power characteristic value is obtained by dividing the difference between the current photovoltaic output active power and the local load active power consumption by the rated power and taking the absolute value; the system frequency characteristic value is obtained by normalizing the sum of the current system frequency change rate and the hyperparameters; the parallel number characteristic value is obtained by subtracting the ratio of the current number of parallel photovoltaic inverters to the parallel number threshold from the first preset value; and the tightening factor at the current moment is obtained by summing the power characteristic value and the system frequency characteristic value and multiplying it by the parallel number characteristic value.
[0040] The specific calculation model for the tightening factor is as follows: , Where Bi is the tightening factor corresponding to 0 at the i-th time (current time); and These represent the current photovoltaic output active power and the local load consumed active power, respectively. Pbase represents the rated power (W) used for normalization. Here, represents the power characteristic value corresponding to the current moment, indicating that the lower the system power matching degree, the simpler the islanding detection. In this case, the perturbation coefficient should be tightened to avoid excessive perturbation; norm represents the normalization function; Fi is the system frequency change rate at the current moment, which is obtained by dividing the difference between the system frequency at the current moment and the system frequency at the previous moment by the time interval between the two moments and taking the absolute value; c is a hyperparameter used to prevent the denominator from being 0, and is generally taken as a small positive number. The system frequency characteristic value indicates that the smaller the absolute value of the system frequency change rate, the more stable the system is, and the smaller the disturbance coefficient protection can be tightened; Ni represents the number of photovoltaic inverters operating in parallel at the current moment, and Nz represents the threshold number of parallel units (e.g., 10 units). The characteristic value for the number of parallel units indicates that the more units connected in parallel, the smaller the disturbance amplitude of a single unit should be to avoid exceeding the total disturbance limit.
[0041] Step S4: Calculate the adaptive perturbation amplitude at the current moment using the base perturbation amplitude, relaxation factor, and tightening factor; use the adaptive perturbation amplitude at the current moment in combination with the active perturbation method to detect the islanding effect.
[0042] The base disturbance amplitude provides the initial disturbance amplitude based on the fundamental characteristics of the power grid for the entire adaptive process, serving as the foundation for adaptive adjustment. The relaxation and tightening factors dynamically adjust the base disturbance amplitude from the perspectives of power quality and system operating status, respectively. The relaxation factor for power quality allows for an appropriate increase in disturbance amplitude when the system's power quality is good and has sufficient margin, thereby improving the performance of related functions. The tightening factor for system operating status forces an increase in disturbance amplitude to ensure the reliability of detection when the system faces difficulties in islanding detection or high risks. Multiplying the base disturbance amplitude by these two factors and clamping it allows for a comprehensive consideration of the basic characteristics of the power grid, power quality, and system operating status. This ensures that the final adaptive disturbance amplitude conforms to the actual situation of the power grid, meets power quality requirements, and effectively achieves functions such as islanding detection under various operating conditions. This realizes adaptive adjustment under multi-source information fusion, ensuring the safe, stable, and efficient operation of the entire system.
[0043] Therefore, the adaptive perturbation amplitude at the current moment is calculated using the base perturbation amplitude, the relaxation factor, and the tightening factor at the current moment.
[0044] Specifically, the relaxation adjustment value is obtained by multiplying the current basic disturbance amplitude by the sum of the first preset value and the normalized value of the relaxation factor at the current time; the tightening adjustment value is obtained by multiplying the current basic disturbance amplitude by the difference between the first preset value and the normalized value of the tightening factor at the current time; and the adaptive disturbance amplitude at the current time is obtained by calculating the average of the relaxation adjustment value and the tightening adjustment value.
[0045] The specific calculation model for the adaptive perturbation amplitude is as follows: , Where Ki represents the adaptive perturbation magnitude at the i-th time (the current time); The relaxed adjustment value indicates that the adaptive base disturbance amplitude increases as the relaxation factor increases; The tightening adjustment value indicates that the adaptive base disturbance amplitude decreases as the tightening factor increases; the above norm is the normalization function; 1 / 2 represents the average of the relaxed and tightened adjustment values.
[0046] This allows us to obtain the adaptive perturbation amplitude at the current moment. Similarly, for each moment, we can obtain the corresponding adaptive perturbation amplitude.
[0047] After obtaining the adaptive disturbance amplitude, the calculated amplitude is first injected into the control loop of the photovoltaic grid-connected system in real time, and detected using the active disturbance method. Next, key system parameters are comprehensively monitored. While injecting the disturbance, changes in parameters such as voltage, frequency, and power at the grid connection point are closely observed. High-precision sensors and advanced data acquisition systems are used to obtain accurate values of these parameters in real time and record their trends over time, providing detailed data for subsequent analysis. Subsequently, the parameter changes are analyzed in depth, comparing the parameters after the disturbance injection with those under normal grid connection conditions to determine if islanding effects exist. If the voltage or frequency deviates from the normal range and... If the situation continues to worsen, a comprehensive assessment is made based on power changes to determine whether an islanding effect has occurred. Finally, appropriate measures are taken based on the analysis results. If an islanding effect is determined to have occurred, protective actions are immediately triggered to disconnect the grid-connected switch, ensuring the safety of equipment and personnel. Simultaneously, power quality monitoring data is used to ensure that harmonics, flicker, and other indicators do not exceed limits during disturbance injection, avoiding adverse effects on the power grid and user equipment. Furthermore, the disturbance strategy is dynamically adjusted according to the system's operating status, prioritizing detection reliability in emergencies and automatically tightening the disturbance amplitude when power quality is critical, achieving a balance between islanding detection sensitivity and power quality compliance, ensuring the safe and stable operation of the photovoltaic grid-connected system.
[0048] In summary, this application adaptively acquires the disturbance amplitude or disturbance coefficient in the traditional active disturbance method, deeply integrating power quality parameters with system operating status. By dynamically calculating "relaxation factors" and "tightening factors," it achieves bidirectional real-time adjustment of the disturbance amplitude. First, the basic disturbance amplitude is obtained based on the short-circuit capacity. Then, the relaxation and tightening factors are obtained from both the power quality and system operating status perspectives. Finally, the basic disturbance amplitude and the two factors are fused to obtain the final adaptive disturbance amplitude. The photovoltaic grid-connected anti-islanding data is comprehensively processed based on the adaptive disturbance amplitude. When the grid is healthy, the disturbance amplitude can be appropriately relaxed to accelerate islanding detection. When power quality is critical, the disturbance is automatically tightened to avoid exceeding limits. In emergency situations, priority is given to ensuring detection reliability, even at the expense of short-term power quality. This effectively balances the sensitivity of islanding detection with the compliance of power quality, eliminating the detection blind spots of traditional algorithms in complex grid scenarios, reducing the risk of false alarms, and providing a superior technical path for the safe and stable operation of photovoltaic grid-connected systems.
[0049] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.
[0050] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion, characterized in that, The method includes: The basic disturbance amplitude at the current moment is obtained by using the preset minimum disturbance amplitude of the strong network, the preset maximum disturbance amplitude of the weak network, the reference short-circuit capacity, and the actual short-circuit capacity at the current moment; The average harmonic voltage effective value change rate at the current moment is obtained by comparing the effective values of each harmonic voltage at the current moment with those at the previous moment; the flicker value set at the current moment is formed by combining the flicker values at each moment within a preset time length before the current moment; the relaxation factor at the current moment is obtained based on the THD value, the flicker value set, the average harmonic voltage effective value change rate, and the fundamental voltage effective value at the current moment. The tightening factor for the current moment is obtained based on the current photovoltaic output active power, local load active power consumption, system frequency change rate, and the number of photovoltaic inverters operating in parallel. The adaptive perturbation amplitude at the current moment is calculated using the base perturbation amplitude, relaxation factor, and tightening factor; the islanding effect is detected by combining the adaptive perturbation amplitude at the current moment with the active perturbation method.
2. The photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion according to claim 1, characterized in that, The process of obtaining the basic disturbance amplitude at the current moment using the preset minimum disturbance amplitude of a strong network, the preset maximum disturbance amplitude of a weak network, the reference short-circuit capacity, and the actual short-circuit capacity at the current moment includes: A first mapping value is obtained by negatively mapping the ratio of the actual short-circuit capacity to the reference short-circuit capacity at the current moment using an exponential function with the natural constant as the base; the difference between the preset maximum disturbance amplitude of the weak network and the preset minimum disturbance amplitude of the strong network is obtained and multiplied by the first mapping value to obtain the multiplication result; the multiplication result is added to the preset minimum disturbance amplitude of the strong network to obtain the basic disturbance amplitude at the current moment.
3. The photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion according to claim 1, characterized in that, The method of obtaining the average harmonic voltage effective value change rate at the current moment based on the change of each harmonic voltage effective value at the current moment compared to the previous moment includes: The rate of change of the effective value of the first harmonic voltage at the current moment is obtained by dividing the absolute value of the difference between the effective value of the first harmonic voltage at the current moment and the effective value of the same harmonic voltage at the previous moment by the time interval between the current moment and the previous moment. The average rate of change of the effective value of each harmonic voltage at the current moment is obtained by calculating the average rate of change of the effective value of the current harmonic voltage.
4. The photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion according to claim 1, characterized in that, The process of obtaining the relaxation factor at the current moment based on the current THD value, flicker value set, average harmonic voltage RMS change rate, and fundamental voltage RMS value includes: The THD margin at the current moment is obtained by dividing the difference between the national standard limit for THD and the current THD value by the national standard limit for THD. The flicker margin at the current moment is obtained by dividing the difference between the national standard limit for flicker and the current flicker value by the national standard limit for flicker. The characteristic value of the change in the effective value of harmonic voltage at the current moment is obtained by subtracting the ratio of the current moment's average effective value change rate to the maximum value among all times' average effective value change rates of harmonic voltage. The coefficient of variation of the flicker value set at the current moment is negatively correlated using an exponential function with the natural constant as the base to obtain the flicker stationarity characteristic value. The harmonic characteristic value is obtained by normalizing the ratio of the sum of the square roots of the squares of the effective values of all harmonic voltages at the current moment to the effective value of the fundamental voltage at the current moment by the first preset value. The relaxation factor at the current moment is obtained by multiplying the sum of the THD margin and the flicker margin, the sum of the characteristic value of the change in the effective value of harmonic voltage and the flicker stationarity characteristic value, and the harmonic characteristic value.
5. The photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion according to claim 1, characterized in that, The tightening factor for the current moment, obtained based on the current photovoltaic output active power, local load active power consumption, system frequency change rate, and the number of photovoltaic inverters operating in parallel, includes: The power characteristic value is obtained by dividing the difference between the current photovoltaic output active power and the local load active power by the rated power and taking the absolute value; the system frequency characteristic value is obtained by normalizing the sum of the current system frequency change rate and the hyperparameters; the parallel number characteristic value is obtained by subtracting the ratio of the current number of parallel photovoltaic inverters to the parallel number threshold from the first preset value; the tightening factor at the current moment is obtained by obtaining the sum of the power characteristic value and the system frequency characteristic value and multiplying it by the parallel number characteristic value.
6. The photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion according to claim 5, characterized in that, The method for obtaining the system frequency change rate is as follows: The rate of change of the system frequency at the current moment is obtained by dividing the difference between the system frequency at the current moment and the system frequency at the previous moment by the time interval between the two moments and taking the absolute value.
7. The photovoltaic grid-connected adaptive anti-islanding data processing method based on multi-source information fusion according to claim 1, characterized in that, The calculation of the adaptive perturbation amplitude at the current moment using the base perturbation amplitude, relaxation factor, and tightening factor at the current moment includes: The relaxation adjustment value is obtained by multiplying the current basic disturbance amplitude by the sum of the first preset value and the normalized value of the relaxation factor at the current time; the tightening adjustment value is obtained by multiplying the current basic disturbance amplitude by the difference between the first preset value and the normalized value of the tightening factor at the current time; the adaptive disturbance amplitude at the current time is obtained by calculating the mean of the relaxation adjustment value and the tightening adjustment value.