Photovoltaic inverter anti-inrush control method and system based on power signal coupling

Abstract

The application relates to the technical field of photovoltaic grid-connected control, and discloses a photovoltaic inverter anti-backflow control method and system based on power signal coupling, which comprises the following steps: collecting photovoltaic smooth power and load signals, combining a supervisory control and data acquisition system to analyze a deviation range and generate a dynamic signal; screening target adaptive rules to determine adjustment parameters, fusing data to generate a control instruction; updating system adjustment parameters and performing overload checking, monitoring an operation state to obtain a stable power output result. The method effectively improves the anti-backflow control precision of the photovoltaic inverter and the system stability through power signal coupling regulation and overload protection, and provides reliable technical support for photovoltaic grid-connected safety.

Description

Technical Field

[0001] This invention relates to the field of anti-reverse current control technology, and in particular to an anti-reverse current control method and system for photovoltaic inverters based on power signal coupling. Background Technology

[0002] Currently, the adoption rate of grid-connected photovoltaic power generation systems is increasing. Inverters, as core equipment, need to implement anti-reverse current control to prevent excess power from flowing back to the grid, causing voltage fluctuations and equipment losses. However, photovoltaic power is greatly affected by sunlight and temperature fluctuations, and the grid load also changes dynamically, placing stringent requirements on the real-time performance and accuracy of anti-reverse current control.

[0003] In one existing technology, photovoltaic inverters often use fixed power threshold judgment or simple PID regulation strategy to prevent reverse current. The equipment used includes a monitoring control and data acquisition system. After collecting the photovoltaic output power and load power in real time, the inverter output power is directly reduced when the difference exceeds the threshold. Some solutions even rely solely on power data at a single moment to perform control.

[0004] However, existing technologies struggle to adapt to the rapid dynamic fluctuations in photovoltaic power and load. Fixed thresholds can easily lead to control lag or over-adjustment, potentially leaving residual reverse current risks and wasting photovoltaic energy. Furthermore, they fail to incorporate historical fluctuation patterns and adapt to dynamic scenarios, making it impossible to adjust control strategies based on real-time changes in sunlight intensity and load, resulting in insufficient control precision. Therefore, existing technologies suffer from slow response to reverse current prevention in photovoltaic inverters, difficulty in adapting to dynamic changes in sunlight and load, and a lack of flexibility in control strategies. Summary of the Invention

[0005] This invention provides a photovoltaic inverter anti-reverse current control method and system based on power signal coupling, to solve the problem that the existing photovoltaic inverter anti-reverse current control relies on fixed power thresholds or single-moment data, resulting in response lag, poor dynamic adaptability, and thus affecting the reliability of anti-reverse current and the utilization rate of photovoltaic energy.

[0006] In a first aspect, the present invention provides a photovoltaic inverter anti-reverse current control method based on power signal coupling, comprising:

[0007] Real-time power data and grid load data of the photovoltaic system are collected and noise-reduced to obtain smooth power signals and load signals. The load signals are subtracted from the smooth power signals to obtain the difference signals.

[0008] Analyze the signal timing changes and fluctuation frequency of the difference signal to determine the amplitude range of the fluctuation amplitude and peak amplitude;

[0009] If the amplitude range exceeds the preset amplitude range, historical power data sequences are extracted from the preset historical database and trend analysis is performed to obtain the fluctuation trend.

[0010] The fluctuation trend is matched with a preset rule base to generate a dynamic signal;

[0011] By comparing the dynamic signal with the smoothed power signal, the deviation range is analyzed, and the adjustment parameters are determined.

[0012] If the deviation range does not exceed the preset deviation range, the adjustment parameters and the load signal are combined to generate a control command;

[0013] The photovoltaic system's adjustment settings are updated according to the control commands, and the operating status is monitored in conjunction with the preset overload protection mechanism and preset stable state indicators to obtain a stable power output result.

[0014] In one optional implementation, real-time power data and grid load data of the photovoltaic system are acquired and noise-reduced to obtain a smoothed power signal and a load signal. The load signal is then subtracted from the smoothed power signal to obtain a difference signal, including:

[0015] Real-time power data and grid load data of photovoltaic systems are collected and integrated to obtain raw collected data.

[0016] The original acquired data is subjected to noise reduction processing, outliers are screened and corrected, and the processed signal is obtained;

[0017] The time alignment deviation of the processed signal is calibrated to obtain a smoothed power signal and load signal;

[0018] The difference between the smoothed power signal and the load signal is calculated to obtain the difference signal.

[0019] In one optional implementation, the step of analyzing the signal timing variations and fluctuation frequencies of the difference signal to determine the amplitude range of the fluctuation amplitude and peak amplitude includes:

[0020] By analyzing the signal timing changes and fluctuation frequencies of the difference signal, the timing change characteristics and fluctuation frequency characteristics are obtained.

[0021] Based on the time-series variation characteristics and the fluctuation frequency characteristics, fluctuation data and peak data are extracted to obtain the amplitude range. In an optional implementation, if the amplitude range exceeds a preset amplitude range, historical power data sequences are extracted from a preset historical database and trend analysis is performed to obtain the fluctuation trend, including:

[0022] By comparing the amplitude range of the fluctuation amplitude and the peak amplitude with the preset amplitude range, a result exceeding the preset range is obtained;

[0023] Based on the results that exceed the preset range, historical power data sequences are extracted from a preset historical database to obtain matching historical data;

[0024] By analyzing the fluctuation patterns of the historical matching data, the trend of fluctuations can be obtained.

[0025] In one optional implementation, matching the fluctuation trend with a preset rule base to generate a dynamic signal includes:

[0026] The fluctuation trend is matched with a preset rule base to obtain the trend matching result;

[0027] Based on the trend matching results, the matching rules in the preset rule base are filtered to obtain the target matching rules;

[0028] Based on the target adaptation rules, a dynamic signal for power fluctuation status is generated.

[0029] In one optional implementation, comparing the dynamic signal with the smoothed power signal, analyzing the deviation range, and determining the adjustment parameters includes:

[0030] By comparing the dynamic signal with the smoothed power signal, a signal comparison result is obtained;

[0031] Based on the signal comparison results, the deviation range is analyzed to obtain deviation range data;

[0032] The adjustment parameters are determined by combining the basic deviation range and the percentage of deviations exceeding the acceptable threshold in the deviation range data.

[0033] In one optional implementation, if the deviation range does not exceed a preset deviation range, the adjustment parameters and the load signal are fused to generate a control command, including:

[0034] If the deviation range does not exceed the preset deviation range, then the range compliance result is obtained;

[0035] Based on the range compliance results, the adjustment parameters and the load signal are fused to obtain fused data;

[0036] Based on the fused data, control commands are generated to determine the final power allocation ratio and adjust the response speed parameters.

[0037] In one optional implementation, the step of updating the adjustment settings of the photovoltaic system according to the control command, and monitoring the operating status in conjunction with a preset overload protection mechanism and preset stability indicators to obtain a stable power output result includes:

[0038] The system adjustment settings of the photovoltaic system are updated according to the control command to obtain the system adjustment parameters;

[0039] Based on the system adjustment parameters, and combined with the preset overload protection mechanism, an overload risk verification is performed to obtain the verification result.

[0040] Based on the verification results and combined with preset stability indicators, the operating status of the photovoltaic system is monitored and analyzed to obtain stable power output results.

[0041] Secondly, the present invention provides a photovoltaic inverter anti-reverse current control system based on power signal coupling, comprising:

[0042] The data acquisition and processing module is used to acquire real-time power data and grid load data of the photovoltaic system and perform noise reduction processing to obtain smooth power signal and load signal. The load signal is subtracted from the smooth power signal to obtain the difference signal.

[0043] The fluctuation amplitude analysis module is used to analyze the signal timing changes and fluctuation frequency of the difference signal, and determine the amplitude range of the fluctuation amplitude and peak amplitude.

[0044] The historical trend extraction module extracts historical power data sequences from a preset historical database and performs trend analysis if the amplitude range exceeds a preset amplitude range, thereby obtaining the fluctuation trend.

[0045] The rule base matching module matches the fluctuation trend with a preset rule base to generate a dynamic signal;

[0046] The deviation analysis and adjustment module is used to compare the dynamic signal with the smoothed power signal, analyze the deviation range, and determine the adjustment parameters.

[0047] A control command generation module is used to generate a control command by integrating the adjustment parameters and the load signal if the deviation range does not exceed a preset deviation range.

[0048] The system regulation and monitoring module is used to update the regulation settings of the photovoltaic system according to the control command, and monitor the operating status in combination with the preset overload protection mechanism and preset stability index to obtain stable power output results.

[0049] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the photovoltaic inverter anti-reverse current control method based on power signal coupling as described in any one of the above claims.

[0050] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the photovoltaic inverter anti-reverse current control method based on power signal coupling described above.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] (1) This invention directly analyzes the real-time power data and grid load data of the photovoltaic system, calculates the time-series changes and fluctuation frequency of the difference signal between the two, captures the essential characteristics of supply and demand imbalance, realizes rapid perception and response to reverse flow risk, and compares with the preset amplitude range to identify abnormal power fluctuation trends before the reverse flow risk actually occurs, so that the system can start the control preparation earlier, changing the traditional anti-reverse flow scheme only passively responds to the reverse flow that has already occurred, and solving the problem of the traditional method's reliance on complex prediction of a single power signal;

[0053] (2) This invention utilizes a preset historical database for trend analysis and rule matching, finds fluctuation patterns in similar scenarios from historical data, generates dynamic adjustment signals by integrating historical fluctuation trends, and achieves intelligent decision-making by combining a preset rule library. It can adapt to different weather patterns and load types, improve system operation stability, and solve the problem that traditional fixed threshold control cannot effectively cope with power mutations under complex working conditions.

[0054] (3) This invention optimizes control commands through deviation analysis and multiple verification mechanisms, taking into account both response speed and adjustment accuracy, ensuring a dynamic balance between power output quality and grid safety, and solving the problems of response lag and large overshoot in traditional control. At the same time, this invention integrates overload protection mechanism and multiple stability state index monitoring, continuously monitoring key indicators such as power fluctuation amplitude and response delay, ensuring that the system operates within the electrical and control stability range while eliminating the risk of reverse flow, thus enhancing the level of protection for the overall system operation safety. Attached Figure Description

[0055] Figure 1 This is a schematic flowchart of a photovoltaic inverter anti-reverse current control method based on power signal coupling provided in the first embodiment of the present invention;

[0056] Figure 2 This is a schematic diagram of a photovoltaic inverter anti-reverse current control system based on power signal coupling, provided in the second embodiment of the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Reference Figure 1 The first embodiment of the present invention provides a photovoltaic inverter anti-reverse current control method based on power signal coupling, comprising the following steps:

[0059] S1. Collect real-time power data and grid load data of the photovoltaic system and perform noise reduction processing to obtain a smoothed power signal and load signal. Subtract the load signal from the smoothed power signal to obtain a difference signal.

[0060] S2, Analyze the signal timing changes and fluctuation frequency of the difference signal to determine the amplitude range of the fluctuation amplitude and peak amplitude;

[0061] S3, if the amplitude range exceeds the preset amplitude range, extract the historical power data sequence from the preset historical database and perform trend analysis to obtain the fluctuation trend.

[0062] S4, match the fluctuation trend with a preset rule base to generate a dynamic signal;

[0063] S5. Compare the dynamic signal with the smoothed power signal, analyze the deviation range, and determine the adjustment parameters;

[0064] S6, if the deviation range does not exceed the preset deviation range, the adjustment parameters and the load signal are combined to generate a control command;

[0065] S7. Update the adjustment settings of the photovoltaic system according to the control command, and monitor the operating status in combination with the preset overload protection mechanism and preset stable state index to obtain a stable power output result.

[0066] In step S1, real-time power data and grid load data of the photovoltaic system are collected and noise-reduced to obtain a smoothed power signal and a load signal. The load signal is subtracted from the smoothed power signal to obtain a difference signal, including:

[0067] S11 collects real-time power data and grid load data of the photovoltaic system and integrates them to obtain the raw collected data;

[0068] S12, the original acquired data is subjected to noise reduction processing, outliers are screened and corrected, and a processed signal is obtained;

[0069] S13, calibrate the time alignment deviation of the processed signal to obtain a smoothed power signal and load signal;

[0070] S14, calculate the difference between the smoothed power signal and the load signal to obtain the difference signal.

[0071] In step S11, real-time power data and grid load data of the photovoltaic system are collected and integrated to obtain the raw collected data.

[0072] It should be noted that the sensors include photovoltaic power sensors and grid load sensors, deployed at the output end of the photovoltaic array and the grid connection end, respectively. Real-time power data acquisition refers to capturing the real-time output power of the components using the photovoltaic power sensor, while grid load data acquisition refers to obtaining the real-time power consumption of the grid using the grid load sensor. The system adopts a multi-rate acquisition architecture: to achieve millisecond-level fast control, the acquisition channel frequency of the differential signal should be no less than 1Hz. Key sensors need to support event-triggered reporting, immediately executing instantaneous power adjustment when the power change rate exceeds a threshold, especially within milliseconds to seconds of detecting reverse flow, rapidly reducing the active power output of the photovoltaic inverter, and directly reducing the net power flowing into the grid to zero through closed-loop control. Simultaneously, to perform trend analysis and identify abnormal power fluctuation trends before the actual occurrence of reverse flow risk, a data channel with 10s / times can be maintained in parallel for data acquisition under normal conditions. This avoids missing key fluctuations in photovoltaic power and grid load due to excessively long acquisition intervals, prevents data redundancy caused by high-frequency acquisition, and adapts to the dual requirements of data timeliness and conciseness for subsequent differential signal trend analysis. The integrated raw data refers to the association of two types of data according to the collection timestamp. Each data entry includes the collection time, real-time photovoltaic power value, and real-time grid load value. The preset data standard range is 0 to 1000kW for real-time photovoltaic power value, which is suitable for the common output range of distributed photovoltaic systems; and 0 to 1200kW for real-time grid load value, which covers the range of grid load fluctuations in the covered area. Data outside the range is marked as abnormal and pending processing.

[0073] For example, some records in the original collected data were 09:01:10, 850kW, 900kW; 09:01:20, 1050kW, 880kW; 09:01:30, 720kW, 1300kW. Because the photovoltaic power of 1050kW > 1000kW and the grid load of 1300kW > 1200kW, they were marked as abnormal items to be processed.

[0074] In step S12, the original acquired data is subjected to noise reduction processing, outliers are screened and corrected, and a processed signal is obtained.

[0075] It should be noted that the 1-minute sliding window method for calculating the mean of the data sequence, i.e., taking data within a continuous 1-minute interval each time, reduces high-frequency noise from the sensor caused by environmental fluctuations, is suitable for a sampling frequency of 10 seconds per sampling, and avoids single fluctuations interfering with data validity. Outlier screening refers to the preset range in step S11, i.e., photovoltaic power 0 to 1000kW and grid load 0 to 1200kW; values ​​exceeding this range are marked as outliers. Outlier correction uses linear interpolation, estimating a reasonable value based on the mean of three normal data points before and after the outlier, ensuring data continuity. This method is suitable for dynamic photovoltaic system scenarios due to its simple calculation and close alignment with the gradual changes in power and load data. The processed signal needs to be correlated with the sampling time to ensure temporal consistency and provide a reliable data foundation for subsequent difference calculations.

[0076] For example, the original data records 09:01:20, 1050kW, 880kW (PV power out of range), 09:01:30, 720kW, 1300kW (load out of range). After marking the anomalies, the power at 09:01:20 is corrected using 09:01:10, 850kW, 09:01:40, 820kW, with an average of 835kW; the load at 09:01:30 is corrected using 09:01:20, 880kW, 09:01:40, 890kW, with an average of 885kW. The processed signals are 09:01:20, 835kW, 880kW; 09:01:30, 720kW, 885kW.

[0077] In step S13, the time alignment deviation of the processed signal is calibrated to obtain a smoothed power signal and a load signal.

[0078] It should be noted that the time alignment deviation stems from the different deployment locations of the photovoltaic power sensor and the grid load sensor. Their hardware clocks may have a ±10s offset, causing a mismatch between power and load data at the same timestamp. Calibration uses the grid load sensor time as a reference because its grid frequency is fixed, with a clock drift of less than 0.1s per day, resulting in superior stability. A time axis correction tool is used to shift the photovoltaic power data timestamp to match the load data; isolated data points with a deviation exceeding 10s are discarded. After calibration, a 2-minute sliding window is used to calculate the mean and smooth the data, eliminating minor fluctuations. The final smoothed power signal includes the calibration timestamp and the smoothed photovoltaic power value, while the load signal includes the same timestamp and the smoothed load value, ensuring a perfect time sequence correspondence between the two.

[0079] For example, in the processed signal, the photovoltaic data is 09:01:30, 820kW, 09:01:40, 815kW, and the load data is 09:01:20, 885kW, 09:01:30, 890kW. If the photovoltaic time is detected to be 10 seconds late, after calibration, the photovoltaic time is changed to 09:01:20 and 09:01:30. After smoothing, the smoothed power signals are 09:01:20, 817.5kW, 09:01:30, 817.5kW, and the load signals are 09:01:20, 885kW, 09:01:30, 890kW.

[0080] In step S14, the difference between the smoothed power signal and the load signal is calculated to obtain the difference signal.

[0081] It should be noted that the calculation must be based on timestamps that correspond perfectly with the smoothed photovoltaic power value and the grid load value at the same timestamp. Since the timing has been calibrated in step S13, there will be no time mismatch issues. The positive or negative value of the difference result represents the supply and demand relationship. A positive difference indicates excess photovoltaic power, a negative difference indicates that the load demand is greater than the photovoltaic output, and a zero value indicates that the supply and demand are balanced. This result can directly reflect whether there is a risk of reverse power flow. The difference signal must retain the same timestamp as the smoothed power signal and the load signal, and the corresponding difference value must be recorded to ensure timing continuity and provide basic data for subsequent analysis of the difference fluctuation characteristics.

[0082] For example, the smoothed power signal is 09:01:20, 817.5kW, 09:01:30, 820kW; the load signal is 09:01:20, 885kW, 09:01:30, 890kW; and the calculated difference signal is 09:01:20, -67.5kW, 09:01:30, -70kW.

[0083] In step S2, the signal timing changes and fluctuation frequency of the difference signal are analyzed to determine the amplitude range of the fluctuation amplitude and peak amplitude, including:

[0084] S21, Analyze the signal timing changes and fluctuation frequency of the difference signal to obtain timing change characteristics and fluctuation frequency characteristics;

[0085] S22, Based on the time-series change characteristics and the fluctuation frequency characteristics, extract fluctuation data and peak data to obtain the amplitude range;

[0086] S23, extract the peak data of the fluctuation range to obtain the peak amplitude range.

[0087] In step S21, the signal timing changes and fluctuation frequencies of the difference signal are analyzed to obtain timing change characteristics and fluctuation frequency characteristics.

[0088] It should be noted that when analyzing the time-series variation characteristics, time intervals are divided into 30-minute intervals. This length is based on a sampling frequency of 10 seconds per interval, with each interval containing 180 data points, sufficient for statistically analyzing stable trends. By comparing the starting and ending values ​​of the difference signal within each interval, the direction of change—whether it is rising, falling, or stable—is determined. Simultaneously, the overall amplitude of the difference within the interval is recorded to form the time-series variation characteristics. When analyzing the fluctuation frequency characteristics, the difference signal is periodically statistically analyzed according to the time series, calculating the number of fluctuation repetitions within different time intervals. The interval of repeated fluctuations of the difference signal within 10 minutes is counted, and the power spectral density is calculated for the difference signal sequence. A peak period within the 1-2 minute interval is defined as high-frequency fluctuation, and a peak period within the 30-minute interval is defined as low-frequency fluctuation. High-frequency fluctuations are mostly caused by short-term changes in photovoltaic power due to cloud cover, while low-frequency fluctuations are mostly caused by periodic adjustments in grid load. Different periods correspond to different interference factors, forming fluctuation frequency characteristics. Both are correlated with the timestamp of the difference signal to ensure time series consistency, providing a basis for subsequently determining the amplitude range of fluctuations and peak values.

[0089] For example, the difference signals are: 09:01:20, -67.5kW; 09:01:30, -70kW; 09:02:20, -68kW; 09:02:30, -66kW; 09:03:20, -65kW. Dividing the signal into two intervals, 09:01-09:02 and 09:03-09:04, the initial difference in the 09:01-09:02 interval is -67.5kW, and the final difference is -66kW. The trend is stable followed by a slow increase, exhibiting time-series variation characteristics. Periodic analysis of the difference signal within this time period shows fluctuations of 1-2 minutes within 10 minutes, identifying it as high-frequency fluctuation, representing the fluctuation frequency characteristic.

[0090] In step S22, based on the time-series change characteristics and the fluctuation frequency characteristics, fluctuation data and peak data are extracted to obtain the amplitude range.

[0091] It should be noted that, based on the periodic component with the highest energy proportion in the fluctuation frequency characteristics, such as by comparing the power spectral density amplitude, the main fluctuation period is determined to be 2 minutes.

[0092] It's worth noting that when extracting fluctuation data, the data is first divided into segments based on the main period of the fluctuation frequency characteristics. This ensures that each data segment contains a complete fluctuation period, avoiding data distortion caused by crossing periods. Then, combined with time-series variation characteristics, the maximum and minimum values ​​of the difference signal within each data segment are selected. The difference between these two values ​​represents the fluctuation amplitude of that data segment. Finally, the fluctuation amplitudes of all data segments are calculated, and the maximum value is taken as the upper limit of the fluctuation amplitude, and the minimum value as the lower limit, forming the overall fluctuation amplitude range. This range can completely cover the fluctuation of the difference signal, providing a basis for subsequent judgment on whether it exceeds a preset threshold.

[0093] For example, the difference signals are: 09:01:20, -67.5kW; 09:01:30, -70kW; 09:02:20, -68kW; 09:02:30, -66kW; 09:03:20, -65kW; 09:03:30, -67kW. Divided into two segments based on a 2-minute fluctuation period: 09:01-09:03 and 09:03-09:05, the first segment has a maximum value of -66kW and a minimum value of -70kW, with a fluctuation amplitude of 4kW; the second segment has a maximum value of -65kW and a minimum value of -67kW, with a fluctuation amplitude of 2kW. The statistically analyzed fluctuation amplitude range is 2kW to 4kW.

[0094] In step S23, the peak data of the fluctuation range is extracted to obtain the peak amplitude range.

[0095] It should be noted that peak data refers to the extreme values ​​of the difference signal within each data segment, including the maximum and minimum algebraic values ​​in each data segment. When extracting peak data, the data segments divided according to the fluctuation period in step S22 must be continued to ensure that the statistical basis of the peak source and the fluctuation amplitude range is consistent. The maximum value of all data segments is calculated, and the largest value is taken as the upper limit of the peak amplitude. The minimum value of all data segments is calculated, and the smallest value is taken as the lower limit of the peak amplitude. The interval formed by the upper and lower limits is the peak amplitude range. This range can completely cover the extreme fluctuations of the difference signal during the monitoring period, providing a basis for determining whether the power fluctuation exceeds the preset threshold.

[0096] For example, in step S22, the maximum value of the difference signal in the 09:01-09:03 data segment is -66kW and the minimum value is -70kW, and the maximum value of the 09:03-09:05 data segment is -65kW and the minimum value is -67kW. The maximum value of all maximum values ​​is -65kW, and the minimum value of all minimum values ​​is -70kW, resulting in a final peak amplitude range of -70kW to -65kW.

[0097] In step S3, if the amplitude range exceeds a preset amplitude range, historical power data sequences are extracted from a preset historical database and trend analysis is performed to obtain the fluctuation trend, including:

[0098] S31, compare the amplitude range of the fluctuation amplitude and peak amplitude with the preset amplitude range, and obtain a result that exceeds the preset range;

[0099] S32, based on the result of exceeding the preset range, extract the historical power data sequence from the preset historical database to obtain matching historical data;

[0100] S33, Analyze the fluctuation patterns of the matched historical data to obtain the fluctuation trend.

[0101] In step S31, the amplitude range of the fluctuation amplitude and the peak amplitude are compared with the preset amplitude range, and a result is obtained that the amplitude exceeds the preset range.

[0102] It should be noted that if the amplitude range exceeds the preset amplitude range, the system will retrieve scenarios with similar fluctuation characteristics from the preset historical database. However, its main purpose is not to directly predict trends, but to provide a set of historically verified, potentially better proportional and integral coefficient suggestions for real-time feedback controllers (such as PID controllers). The preset range needs to be set in conjunction with the photovoltaic system parameters and grid requirements. The preset fluctuation amplitude range is set to 0 to 3kW based on 3% of the system's rated output of 1000kW. The preset peak amplitude range is set to -50kW to +10kW with reference to the grid's allowable power deviation. The positive range is for photovoltaic power surplus scenarios. Both meet the industry's anti-reverse current control standards. During the comparison, the fluctuation amplitude range in step S22 is first compared with the preset fluctuation amplitude range, and then the peak amplitude range in step S23 is compared with the preset peak amplitude range. If either one exceeds the corresponding preset range, an out-of-preset-range result is generated. The result needs to record the time period of exceeding the range, the fluctuation amplitude deviation value, the peak deviation value, and the corresponding timestamp to facilitate subsequent location of unstable fluctuation periods.

[0103] For example, the fluctuation range in step S22 is 2kW to 4kW, and the peak amplitude range in step S23 is -70kW to -65kW. Comparison reveals that 4kW exceeds the preset upper limit of fluctuation amplitude by 3kW and 1kW; -70kW is below the preset lower limit of peak amplitude by -50kW, with a deviation of -20kW; -65kW is below -50kW, with a deviation of -15kW; and -65kW exceeds the preset lower limit of peak amplitude by -50kW and -15kW, resulting in a result exceeding the preset range. Records show that the fluctuation range exceeded 1kW and the peak amplitude of -70kW exceeded -20kW during the period from 09:01 to 09:03, and the peak amplitude of -65kW exceeded -15kW during the period from 09:03 to 09:05.

[0104] In step S32, based on the result of exceeding the preset range, historical power data sequences are extracted from the preset historical database to obtain matching historical data.

[0105] It should be noted that the extraction should be based on the characteristics of the time period exceeding the preset range, including the fluctuation range and peak amplitude range of the out-of-range period, as well as the corresponding environmental conditions, such as light intensity and ambient temperature. Power data from the past year should be selected from a preset historical database. This time span should cover common seasonal and weather conditions for photovoltaic systems to ensure the representativeness of the matching data. The selection must meet two conditions: first, the fluctuation and peak amplitude deviations between the historical period and the current out-of-range period should be within 10%; second, the light intensity and ambient temperature of the historical period should deviate from the current period by ±5%, eliminating interference from environmental differences in data matching. The extracted matching historical data must include historical timestamps, historical smoothed power values, historical load values, and historical difference signals, consistent with the dimensions of the current data, providing a comparable basis for subsequent fluctuation trend extraction.

[0106] It is worth noting that the environmental condition deviation can be adjusted appropriately according to the actual application scenario. For example, when there are not enough data matches with a fluctuation characteristic deviation of ≤10%, the data with the highest fluctuation characteristic matching degree can be retained first, and the environmental condition deviation can be relaxed to ±10% to improve the matching success rate, or the data with a fluctuation characteristic deviation of ≤10% can be matched first, and the environmental condition can be used as a secondary filter.

[0107] For example, the current time period outside the preset range is 09:01-09:05, with a fluctuation range of 2-4kW, a peak value of -70 to -65kW, a light intensity of 810W / m², and a temperature of 27℃. Data from the past year's historical data for a specific day, 09:00-09:04, is selected. Its fluctuation range is 2.1-3.9kW, a peak value of -69.5 to -64.8kW, a light intensity of 808W / m², and a temperature of 26.8℃, meeting the deviation requirements. The resulting matching historical data are: 09:00:20, 815kW, 885kW, -70kW; 09:01:20, 818kW, 882.8kW, -64.8kW; 09:03:20, 816kW, 885.5kW, -69.5kW.

[0108] In step S33, the fluctuation pattern of the matched historical data is analyzed to obtain the fluctuation trend.

[0109] It should be noted that the analysis is based on matching historical difference signals in historical data, which directly reflect the supply and demand fluctuations between photovoltaic power and grid load. The system is divided into 15-minute time intervals, adapted to a sampling frequency of 10 seconds per interval. Each time interval contains 90 data points (corresponding to 15 minutes), and each interval contains 15 data points, allowing for precise capture of local fluctuations. The mean of the historical difference signal within each time interval is calculated and compared with the mean of adjacent time intervals. If the mean of a later time interval is higher than that of a previous time interval, the trend is considered upward; if the mean of a later time interval is lower, the trend is considered downward; if the mean deviation is less than 0.5kW, the trend is considered stable. Simultaneously, the overall change in the mean within the time interval is statistically analyzed, as well as key turning points, such as a shift from a continuous decline to an increase in the mean after a certain period. Combined with historical environmental conditions, such as a negative shrinkage of the mean when sunlight intensity increases, the final trend is summarized, covering the overall direction of change, the changes in the amplitude of fluctuations, and the timing of key turning points, providing a clear basis for subsequent trend matching.

[0110] For example, the historical difference signals matching historical data are: 09:00:20, -70kW; 09:00:30, -69.8kW; 09:00:40, -69.5kW; 09:00:50, -69.2kW; 09:01:00, -68.8kW; 09:01:10, -66.5kW; 09:01:20, -64.8kW; 09:01:30, -65.2kW; 0 The power consumption was as follows: 9:01:40, -65.5kW; 09:01:50, -65.8kW; 09:02:20, -67.5kW; 09:02:30, -68.2kW; 09:02:40, -68.8kW; 09:02:50, -69.0kW; 09:03:00, -69.2kW; 09:03:10, -69.4kW; 09:03:20, -69.5kW. Dividing the data into four time periods: 09:00-09:01, 09:01-09:02, 09:02-09:03, and 09:03-09:04, the average power consumption for each period was -69.46kW, -65.56kW, -68.54kW, and -69.24kW, respectively. The average value increased from 09:00 to 09:01 and then from 09:01 to 09:02, and gradually decreased in subsequent periods. The overall fluctuation range increased from 0.8kW to 1.7kW and then decreased to 0.6kW. The key turning point was in the period from 09:01 to 09:02, during which the average value increased from -69.46kW to -65.56kW, an increase of 3.9kW. The fluctuation trend was a slight increase in the early stage, a decline in the middle stage, and a stabilization in the later stage. The fluctuation range increased first and then decreased. The key turning point was around 09:01.

[0111] In step S4, the fluctuation trend is matched with a preset rule base to generate a dynamic signal, including:

[0112] S41, Match the fluctuation trend with the preset rule base to obtain the trend matching result;

[0113] S42, Based on the trend matching results, filter the adaptation rules in the preset rule base to obtain the target adaptation rule;

[0114] S43, Generate a dynamic signal of power fluctuation state according to the target adaptation rule.

[0115] In step S41, the fluctuation trend is matched with a preset rule base to obtain a trend matching result.

[0116] It should be noted that the preset rule base is constructed based on the historical fluctuation trend classification of photovoltaic systems over the past year. It includes three core rule types: slow rise with decreasing amplitude, rapid decline with increasing amplitude, and stable fluctuation. Each rule type specifies the overall direction of change, the range of fluctuation amplitude changes (the fluctuation amplitude change ratio is calculated based on the fluctuation amplitude of the initial short period and the fluctuation amplitude of the final short period; the formula for slow rise or stable fluctuation is the difference between the initial amplitude and the final amplitude divided by the initial amplitude multiplied by 100%, and the formula for rapid decline is the difference between the final amplitude and the initial amplitude divided by the initial amplitude multiplied by 100%), and the three core characteristics of the key turning point time interval.

[0117] It is worth noting that during the matching process, the fluctuation trend obtained in step S33, including the overall direction of change, fluctuation amplitude changes, and key turning point times, is compared one by one with each type of feature in the rule base. If the overall direction is completely consistent, the fluctuation amplitude change deviation is less than 10%, and the key turning point time is within the rule range, it is considered a match. The feature matching degree calculation logic is as follows: if the direction is consistent, it is 100%; otherwise, it is 0%. The amplitude matching degree is calculated by subtracting the absolute difference between the actual change ratio and the average value of the rule range from 100%, then dividing by the average value of the rule range, and finally multiplying by 100%. The time matching degree is calculated by subtracting the absolute difference between the actual time of the turning point and the time of the midpoint of the interval from 100%, then dividing by the interval duration, and finally multiplying by 100%. The generated trend matching result must include the matching rule type and the matching degree of each feature to provide a basis for the subsequent generation of dynamic reference signals.

[0118] For example, the fluctuation trend obtained in step S33 is a slight increase in the early stage, a decline in the middle stage, and a stabilization in the later stage. After adjustment, the core trend focuses on the slow increase after excluding the short-term fluctuations of the slight decline in the middle stage. The fluctuation range is 0.8kW in the initial short period from 09:00 to 09:01 and 0.6kW in the final short period from 09:03 to 09:04. The fluctuation range is reduced by 25%, and the key turning point is around 09:01.

[0119] The characteristics of the slow upward and decreasing amplitude rule in the preset rule base are: overall upward direction, fluctuation amplitude decrease ratio of 10%-30%, average range of 20%, key turning point time interval of 08:30-09:30, midpoint of interval of 09:00, duration of 60 minutes. The comparison revealed that the overall direction was consistent, with the fluctuation amplitude decreasing by 25%. Based on the initial 0.8kW and the final 0.6kW, the calculation was (0.8-0.6) / 0.8×100%=25%, which falls within the 10%-30% range. The turning point at 09:01 was within the rule range. The feature matching degree included: direction matching degree 100%, amplitude matching degree = 100%-|25-20| / 20×100%=75%, and time matching degree 100%-|09:01-09:00| / 60×100%≈98.3%. All feature matching degrees met the requirements, and the trend matching result was a match for a slowly rising amplitude decreasing rule, with an overall direction matching degree of 100%, an amplitude change matching degree of 75%, and a turning point time matching degree of approximately 98.3%.

[0120] In step S42, based on the trend matching results, the matching rules in the preset rule base are filtered to obtain the target matching rule.

[0121] It should be noted that each core rule type in the preset rule base—namely, the slow increase with decreasing amplitude, the rapid decrease with increasing amplitude, and the stable fluctuation type—includes three parts: feature description, applicable scenario, and control parameter reference. The applicable scenario clarifies the applicable lighting conditions, such as stable lighting, sudden changes in lighting, and periods of grid load. Low load periods are defined as times when regional grid electricity demand is low, such as 00:00-06:00 and 22:00-24:00. High load periods are defined as times when electricity demand is high, such as 08:00-12:00 and 18:00-22:00. The control parameter reference provides the basis for subsequent power adjustments.

[0122] It is worth noting that the selection process must be based on two conditions: first, the matching degree of each feature in the trend matching results of step S41 must exceed 85%. This threshold has been verified in multiple scenarios to balance matching accuracy and flexibility; second, the current actual scenario must be completely consistent with the rule's adaptation scenario, meaning that the actual lighting conditions (such as stable or sudden changes in lighting) must completely match the lighting conditions labeled by the rule, and the actual time must fall within the peak or trough period defined by the rule, excluding control deviations caused by scenario mismatch. The final target adaptation rule must fully retain the rule type, adaptation scenario description, and control parameter reference to provide specific parameter support for the subsequent generation of dynamic reference signals.

[0123] For example, the trend matching result in step S41 is a slowly increasing, decreasing amplitude rule. The overall direction matching degree is 100%, the amplitude change matching degree is 75%, and the inflection point time matching degree is approximately 98.3%, all exceeding 85%. The current actual scenario is stable lighting from 09:01 to 09:05 in the morning, which falls within the peak load period from 08:00 to 12:00. The applicable scenario for this rule is precisely stable lighting and peak load periods. The reference control parameters are a power adjustment step size of 0.5 kW / min and a fluctuation warning threshold of ±2 kW, which meet the screening conditions. Finally, the target applicable rule is a slowly increasing, decreasing amplitude rule, applicable to the scenario of stable lighting and peak load periods, with reference control parameters of a power adjustment step size of 0.5 kW / min and a fluctuation warning threshold of ±2 kW.

[0124] In step S43, a dynamic signal of power fluctuation state is generated according to the target adaptation rule.

[0125] It should be noted that the dynamic signal is not directly used as the power setpoint, but rather provides parameter adjustment suggestions and forward-looking warnings. The dynamic signal must include three core components: real-time power adjustment suggestions, fluctuation warning intervals, and timing synchronization indicators. All three must be generated based on the control parameters of the target adaptation rules to ensure matching with the operating rhythm of the photovoltaic system. The real-time power adjustment suggestions are based on the power adjustment step size in the rules, combined with the real-time value of the current difference signal. When the step size is 0.5 kW / min, an adjustment amount of 10 min multiplied by 0.5 kW / min is generated every 10 minutes to avoid frequent adjustments that could lead to system instability. The fluctuation warning interval refers to the fluctuation warning threshold in the rules, using the current difference signal as a benchmark, and defining the safe fluctuation boundary by the upper and lower fluctuation threshold range. The timing synchronization indicator must maintain the same timestamp as the previous difference signal and smoothed power signal to ensure that the control action accurately corresponds to the specific time period.

[0126] It is worth noting that the dynamic signals are ultimately presented in a structured form, with each timestamp corresponding to a set of adjustment suggestions and early warning interval data. The adjustment suggestions prioritize situations where photovoltaic power is excessive (the difference signal is positive, indicating a risk of backflow). By reducing photovoltaic output, the power backflow into the grid is prevented, providing a real-time and actionable basis for subsequent power regulation.

[0127] For example, the control parameters of the target adaptation rule are a power adjustment step size of 0.5 kW / min and a fluctuation warning threshold of ±2 kW. The current difference signal at 09:01 is +5 kW, indicating excess photovoltaic power and a risk of reverse current. The calculated power adjustment amount at 09:11 is the power adjustment step size of 0.5 kW / min multiplied by the adjustment interval of 10 min, resulting in 0.5 kW. The adjustment suggestion is to reduce the power output from +5 kW to +4.5 kW to reduce photovoltaic output and avoid the risk of reverse current. The fluctuation warning range is +5 kW ±2 kW, i.e., +3 kW to +7 kW. A difference exceeding +7 kW or falling below +3 kW requires close monitoring. The timing synchronization identifier is 09:11, which is linked to the timestamp of the current difference signal. The final generated dynamic signal segments are: 09:11, with the adjustment suggestion to reduce the power output from +5 kW to +4.5 kW, and a fluctuation warning range of +3 kW to +7 kW; 09:21, with the adjustment suggestion to reduce the power output from +4.5 kW to +4 kW, and a fluctuation warning range of +2.5 kW to +6.5 kW.

[0128] In step S5, the dynamic signal is compared with the smoothed power signal to analyze the deviation range and determine the adjustment parameters, including:

[0129] S51, compare the dynamic signal with the smoothed power signal to obtain the signal comparison result;

[0130] S52, Analyze the deviation range based on the signal comparison results to obtain deviation range data;

[0131] S53. Based on the basic deviation range and the percentage of deviations exceeding the qualified threshold in the deviation range data, determine the adjustment parameters.

[0132] In step S51, the dynamic signal is compared with the smoothed power signal to obtain a signal comparison result.

[0133] It should be noted that the comparison should revolve around three core dimensions: timestamp synchronization, power value deviation, and warning interval matching, to ensure consistency in timing and data logic between the two. First, the timestamps must be verified to ensure a one-to-one correspondence between the timestamps of the dynamic signal and the smoothed power signal, avoiding comparison errors caused by timing misalignment. This synchronization is based on the timing calibration in previous steps S31-S43. Second, the power value deviation is calculated by subtracting the power adjustment recommendation value for the corresponding period in the dynamic signal from the real-time value of the smoothed power signal at the same timestamp. If the absolute value of the deviation is less than ±0.3kW, it is considered acceptable; otherwise, it is marked as abnormal. Finally, the warning interval matching is assessed by checking whether the real-time value of the smoothed power signal falls within the corresponding fluctuation warning interval of the dynamic signal. If it falls within the interval, the matching is acceptable; otherwise, it is marked as interval out of bounds.

[0134] It is worth noting that the signal comparison results need to be recorded one by one according to the timestamp, including the timestamp, power value deviation, deviation judgment result, and warning interval matching result, so as to provide a direct basis for subsequent judgment on whether control is needed.

[0135] For example, the dynamic signal 09:11 contains a power reference value of 817.9kW and a fluctuation warning range of +3kW to +7kW (corresponding to the difference signal range). The corresponding smoothed power signal at 09:11 has a value of 818kW. First, a timestamp synchronization check is performed. The calculated deviation is 818kW (smoothed power signal value) minus 817.9kW (dynamic signal power reference value) = +0.1kW, with an absolute value less than ±0.3kW, so the deviation is acceptable. Simultaneously, the real-time value of the difference signal corresponding to this smoothed power signal, +4.6kW, falls within the fluctuation warning range of +3kW to +7kW of the dynamic signal, so the warning matching is acceptable. The dynamic signal 09:21 contains a power reference value of 817.8kW and a fluctuation warning range of +2.5kW to +6.5kW. The corresponding smoothed power signal at 09:21 has a value of 817.9kW. The calculated deviation is 817.9kW minus 817.8kW = +0.1kW, so the deviation is acceptable. The corresponding real-time value of the difference signal, +4.1kW, falls within the warning range, so the warning matching is acceptable. At 09:26, the smoothed power signal value was 819.2kW, corresponding to a dynamic signal power reference value of 816kW for the period 09:21-09:31. The calculated deviation was 819.2kW - 816kW = +3.2kW, which exceeded ±0.3kW in absolute value, indicating an abnormal deviation. Furthermore, the real-time value of the difference signal corresponding to this smoothed power signal was +7.2kW, exceeding the upper limit of the dynamic signal's synchronous fluctuation warning range of +2.5kW to +6.5kW, indicating an abnormal warning match. The final signal comparison results show that the deviation and warning match were acceptable for the period 09:11-09:21, while the deviation and warning match were abnormal for the period 09:26.

[0136] In step S52, the deviation range is analyzed based on the signal comparison results to obtain deviation range data.

[0137] It should be noted that the analysis first requires filtering the effective power value deviation data from the signal comparison results, excluding invalid data with mismatched timestamps, to ensure the accuracy of the analysis sample. Then, the maximum and minimum values ​​of all effective deviations are calculated, forming the basic deviation range. Simultaneously, combined with the deviation acceptable threshold of ±0.3kW set in step S51, the proportion of deviation data exceeding the acceptable threshold within the range is marked. The final deviation range data must include the basic deviation range and the proportion of deviations exceeding the acceptable threshold, providing a quantitative basis for subsequent judgment on whether adjustment and correction are needed.

[0138] For example, the valid deviation data in the signal comparison results are +0.1kW, +0.1kW, +3.2kW, +0.2kW, and -0.2kW. The maximum deviation is +3.2kW, and the minimum is -0.2kW, with a basic deviation range of -0.2kW to +3.2kW. One deviation exceeds the acceptable threshold of ±0.3kW, at +3.2kW, totaling 1 valid data point, bringing the total to 5 valid data points, accounting for 20%. The final deviation range data record shows a basic deviation range of -0.2kW to +3.2kW, with deviations exceeding the acceptable threshold accounting for 20%.

[0139] In step S53, the adjustment parameters are determined by combining the basic deviation range and the percentage of deviations exceeding the qualified threshold in the deviation range data.

[0140] It should be noted that the core adjustment parameters include the adjustment step size correction value and the warning interval offset. Both are calculated based on the basic deviation range and the percentage of deviations exceeding the acceptable threshold in the deviation range data to ensure a close match with the actual deviation situation. If the percentage of deviations exceeding the acceptable threshold is less than 30%, the adjustment step size correction value is finely adjusted by 10% of the original adjustment step size of 0.5kW / min to avoid over-correction; if the percentage is greater than 30%, it is corrected by 20% to accelerate deviation convergence. At the same time, the warning interval is adjusted according to the overall bias of the basic deviation range. If the interval is generally negative, the warning interval is shifted downward by 0.2kW; if it is generally positive, it is shifted upward by 0.2kW to ensure that the warning boundary matches the actual power fluctuation. The final adjustment parameters must include the corrected adjustment step size and the upper and lower limits of the shifted warning interval to provide an execution standard for subsequent power regulation.

[0141] For example, the deviation range data is a basic deviation interval of -0.2kW to +3.2kW, with an overall positive deviation and a deviation exceeding the acceptable threshold accounting for 20%, which is less than 30%. The calculated adjustment step size correction value is 0.5kW / min multiplied by 10%, which equals 0.05kW / min, resulting in a corrected step size of 0.55kW / min. The warning interval is shifted upward by 0.2kW, adjusting the original interval of +3kW to +7kW to +3.2kW to +7.2kW. The final determined adjustment parameters are an adjustment step size of 0.55kW / min and a warning interval of +3.2kW to +7.2kW.

[0142] In step S6, if the deviation range does not exceed a preset deviation range, the adjustment parameters and the load signal are fused to generate a control command, including:

[0143] S61, if the deviation range does not exceed the preset deviation range, then a range compliance result is obtained;

[0144] S62, based on the range compliance result, the adjustment parameters and the load signal are fused to obtain fused data;

[0145] S63, match the fused data with the preset adjustment response threshold, determine the initial power allocation ratio, and calibrate the response speed parameter in combination with the preset dynamic monitoring data to obtain the adaptation parameter;

[0146] S64, based on the adaptation parameters, generate control commands for the final power allocation ratio and adjust the response speed parameters.

[0147] In step S61, if the deviation range does not exceed the preset deviation range, a range compliance result is obtained.

[0148] It should be noted that the preset range here must be consistent with the deviation compliance standard in step S51, which includes two core indicators. First, the basic deviation range must fall entirely within ±0.3kW. This threshold is determined through system stability testing to ensure power control accuracy. Second, the proportion of deviations exceeding the compliance threshold must not exceed 30% to avoid a small number of abnormal deviations affecting the overall compliance judgment. Both indicators must be met simultaneously during the judgment. If the basic deviation range does not exceed ±0.3kW and the proportion exceeding the threshold does not exceed 30%, it can be judged as not exceeding the preset range. The generated range compliance result must record complete information, including the timestamp range of the compliance period, the actual basic deviation range, and the actual proportion of deviations exceeding the compliance threshold, to ensure clear verification of the compliance basis during subsequent traceability.

[0149] For example, the deviation range data is a basic deviation interval of -0.2kW to +0.2kW. If it is completely within ±0.3kW and the deviation exceeds the qualified threshold by 10%, it does not exceed 30%. Both indicators meet the preset range requirements, and the range compliance result is obtained. The compliance period is recorded as 09:11-09:31. The actual basic deviation interval is -0.2kW to +0.2kW, and the actual excess is 10%. The judgment result is that the range is compliant.

[0150] In step S62, based on the range compliance result, the adjustment parameters and the load signal are fused to obtain fused data.

[0151] It should be noted that the fusion adopts a timestamp anchoring and association technology. Using the compliant time period timestamps in the compliance results as a benchmark, the adjustment parameters (corrected step size, offset warning interval) are first matched to the corresponding timestamps according to time periods. Then, the load signal data (real-time load value, load change trend) under the same timestamp are associated with the adjustment parameters to form a complete data set under a single timestamp. This is not a weighted average; multi-dimensional data integration is achieved solely through timestamp alignment. The load change trend judgment criteria are: a slight increase refers to a load value increase of less than or equal to 2% between adjacent timestamps, and a sharp drop refers to a load value decrease of greater than or equal to 5%. The logic of combining the adjustment step size with the load trend is that the step size remains unchanged when the load slightly increases, and temporarily decreases by 20% when the load sharply drops. Actual load adaptability is defined as the ability of the fused data to dynamically adjust the step size according to the real-time load trend, avoiding excessively large step sizes during sharp load drops that could lead to power imbalance, and ensuring that the control actions closely match the actual load changes. Finally, the fused data is arranged in timestamp order, with each timestamp corresponding to a complete set of data, providing one-stop data support for subsequent power control execution.

[0152] For example, the compliance period for the scope compliance results is 09:11-09:31, with adjustment parameters of 0.55kW / min and a warning range of +3.2kW to +7.2kW. The load signals are: 09:11 real-time load 882kW with a stable trend; 09:21 real-time load 883.8kW with a slight increase (2%); 09:31 real-time load 883.8kW with a stable trend. The fused data is: 09:11, adjustment step 0.55kW / min, warning range +3.2kW to +7.2kW, real-time load 882kW, stable load trend; 09:21, adjustment step 0.55kW / min, warning range +3.2kW to +7.2kW, real-time load 883.8kW, slightly increasing load trend; 09:31, adjustment step 0.55kW / min, warning range +3.2kW to +7.2kW, real-time load 883.8kW, stable load trend.

[0153] In step S63, the fused data is matched with a preset adjustment response threshold to determine the initial power allocation ratio. The response speed parameter is calibrated by combining the preset dynamic monitoring data to obtain the adaptation parameter.

[0154] It should be noted that the timestamp anchoring association technology is adopted. Based on the timestamp of the fused data, the real-time load value, adjustment step size and load trend are first extracted from the fused data and compared with the preset adjustment response thresholds (load range threshold 800-900kW, step size adaptation threshold 0.4-0.6kW / min). If the threshold conditions are met, the initial power allocation ratio is determined according to the preset rules: 60% for the main line and 40% for the auxiliary line. Then, the preset dynamic monitoring data (power fluctuation frequency, load deviation value) under the same timestamp are associated and calibrated with the adjustment response speed parameters. The integration is achieved by matching data features without weighted calculation. The calibration logic is that if the fluctuation frequency exceeds 2 times / minute, the response speed is increased by 15% and if it is less than 1 time / minute, it is decreased by 10%. When the load deviation value exceeds 1.5%, it is adjusted synchronously. The actual adaptability is defined as the parameters after calibration conforming to the real-time load trend to avoid response lag or over-adjustment.

[0155] For example, the fused data is as follows: 09:11 real-time load 882kW, step size 0.55kW / min, stable trend; 09:21 real-time load 883.8kW, step size 0.55kW / min, slightly increasing trend; 09:31 real-time load 883.8kW, step size 0.55kW / min, stable trend; preset dynamic monitoring data: 09:11 fluctuation 1 time / minute, deviation 1%; 09:21 fluctuation 2 times / minute, deviation 1.2%; 09:31 fluctuation 1 time / minute, deviation 0.8%; the corresponding adaptation parameters are: 09:11 initial ratio 60% main line / 40% auxiliary line, response speed 1.0 times; 09:21 initial ratio 60% main line / 40% auxiliary line, response speed 1.15 times; 09:31 initial ratio 60% main line / 40% auxiliary line, response speed 1.0 times.

[0156] In step S64, control commands for the final power allocation ratio and the adjustment of the response speed parameters are generated based on the adaptation parameters.

[0157] It should be noted that the parameter analysis and integration technology is used to first extract the initial power allocation ratio (main line / auxiliary line) and the calibrated adjustment response speed from the adaptation parameters. Then, combined with the real-time load value and smoothed photovoltaic power value in the fused data, the specific power allocation value of photovoltaic and energy storage is calculated (photovoltaics are matched according to the real-time load value to avoid the risk of reverse current, and energy storage is the smoothed photovoltaic power value minus the photovoltaic allocation value). The adjustment response speed directly uses the calibrated parameters. Among them, the power allocation value must be in line with the anti-reverse current target to avoid excessive photovoltaic power. The adjustment response speed must be adapted to the system sampling interval and execution delay in the subsequent step S71. The control command must clearly specify the photovoltaic or energy storage power allocation value and response speed to provide a clear parameter basis for system adjustment setting updates.

[0158] For example, with adaptation parameter 09:11, the initial ratio is 60% for the main line and 40% for the auxiliary line, and the response speed is 1.0 times (corresponding to 10s / time). The fused data 09:11 shows a real-time load of 882kW and a smoothed photovoltaic power of 886.5kW, resulting in a calculated photovoltaic allocation of 882kW and energy storage of 4.5kW. With adaptation parameter 09:21, the initial ratio is 60% for the main line and 40% for the auxiliary line, and the response speed is 1.15 times (corresponding to 8s / time). The fused data 09:21 shows a real-time load of 883.8kW and a smoothed photovoltaic power of 887.8kW, resulting in a calculated photovoltaic allocation of 883.8kW and energy storage of 4kW. The generated control commands are: 09:11 power allocation: photovoltaic 882kW, energy storage 4.5kW, adjustment response speed 10s / time; 09:21 power allocation: photovoltaic 883.8kW, energy storage 4kW, adjustment response speed 8s / time.

[0159] In step S7, the adjustment settings of the photovoltaic system are updated according to the control command, and the operating status is monitored in conjunction with the preset overload protection mechanism and preset stability index to obtain a stable power output result, including:

[0160] S71, Update the adjustment settings of the photovoltaic system according to the control command to obtain the system adjustment parameters;

[0161] S72, Based on the system adjustment parameters, and combined with the preset overload protection mechanism, an overload risk verification is performed to obtain the verification result;

[0162] S73. Based on the verification results and combined with the preset stable state indicators, monitor and analyze the operating status of the photovoltaic system to obtain a stable power output result.

[0163] In step S71, the adjustment settings of the photovoltaic system are updated according to the control command to obtain the system adjustment parameters.

[0164] It should be noted that updates must revolve around two core pieces of information in the control commands: the power allocation value and the adjustment response speed. This must be combined with the actual photovoltaic power generation capacity (smoothing the photovoltaic power value) and the anti-reverse current target to ensure the system settings closely match actual operating conditions. The power allocation module sets the output threshold according to the specific power values ​​of photovoltaic and energy storage in the commands. When the difference signal is positive, indicating photovoltaic excess and a risk of reverse current, the photovoltaic output threshold is set to the real-time load value to prevent excess power from flowing back into the grid. The energy storage threshold equals the smoothed photovoltaic power value minus the real-time load value (accommodating excess power). When the difference signal is normal, the photovoltaic output threshold is set to the smoothed photovoltaic power value (maximizing power generation). The response speed module adjusts the system sampling interval and execution delay according to the response speed in the commands. For example, if the command response speed is 10 seconds / time, the sampling interval is set to 10 seconds, and the execution delay is controlled within 1 second to avoid command execution lag. The final system adjustment parameters must include the photovoltaic and energy storage output thresholds, sampling and delay settings, and can be directly imported into the system to complete the configuration update.

[0165] For example, control command 09:11 allocates power to 882kW of photovoltaic power and 4.5kW of energy storage, with an adjustment response speed of 10s / time, corresponding to a smoothed photovoltaic power value of 886.5kW and a real-time load value of 882kW. After the update, the system adjustment parameters are: 09:11, photovoltaic output threshold 882kW (equal to the real-time load value), energy storage threshold equal to 886.5kW minus 882kW equals 4.5kW, sampling interval 10s, execution delay 1s; command 09:21 allocates power to 883.8kW of photovoltaic power and 4kW of energy storage, with an adjustment response speed of 8s / time, corresponding to a smoothed photovoltaic power value of 887.8kW and a real-time load value of 883.8kW. After the update, the parameters are: 09:21, photovoltaic output threshold 883.8kW (equal to the real-time load value), energy storage threshold equal to 887.8kW minus 883.8kW equals 4kW, sampling interval 8s, execution delay 1s.

[0166] In step S72, based on the system adjustment parameters and in conjunction with the preset overload protection mechanism, an overload risk verification is performed to obtain the verification result.

[0167] It should be noted that the verification revolves around the power threshold and response speed in the system adjustment parameters, combined with the standard for the preset overload protection mechanism. The core focus is on whether two indicators are compliant. The preset overload protection mechanism includes two requirements: first, the power limit should not exceed 105% of the system's rated power. This ratio is to reserve safety redundancy, avoid long-term full-load operation leading to component damage, and cope with short-term load fluctuations; second, the response delay limit should not exceed 2 seconds. This duration is derived by balancing the equipment's execution speed and stability. Too long a delay will prevent timely intervention in overload situations, while too short a delay may cause frequent false triggers. First, the power threshold is verified by calculating the sum of the output thresholds of photovoltaic and energy storage. If the sum does not exceed the power limit, it is qualified; otherwise, it is marked as a power overload risk. Next, the response speed is verified by checking whether the execution delay in the system adjustment parameters exceeds the delay limit. If it does not exceed the limit, it is qualified; otherwise, it is marked as a response lag risk. The verification results must fully record the timestamp, the comparison between the total power and the limit, the comparison between the response delay and the limit, and the final risk assessment, clearly reflecting whether the system has any potential overload risks.

[0168] For example, if the system adjustment parameter 09:11 is 882kW for photovoltaic and 4.5kW for energy storage, with an execution delay of 1s, and the system rated power is 1000kW, then the power limit is 1000kW multiplied by 105%, which equals 1050kW. The calculated total power is 882kW plus 4.5kW, which equals 886.5kW, not exceeding 1050kW; the execution delay of 1s does not exceed 2s. The verification result at 09:11 shows no overload risk, and both power and response are qualified. At 09:21, with parameters of 883.8kW for photovoltaic and 4kW for energy storage, and an execution delay of 1s, the total power is 883.8kW plus 4kW, which equals 887.8kW, not exceeding 1050kW. The execution delay is qualified, and the verification result at 09:21 shows no overload risk. At 09:26, assuming parameters of 1000kW for photovoltaic and 60kW for energy storage, and an execution delay of 1s, the total power is 1000kW plus 60kW, which equals 1060kW, exceeding 1050kW. The verification result at 09:26 indicates a power overload risk, but the response delay is qualified.

[0169] In step S73, based on the verification results and combined with preset stability indicators, the operating status of the photovoltaic system is monitored and analyzed to obtain a stable power output result.

[0170] It should be noted that the analysis must simultaneously consider the risk assessment results and the preset stability state indicator monitoring standards. Both must be combined to determine whether the system is in a stable output state. The preset stability state indicators include two core indicators: first, the power fluctuation amplitude should not exceed ±1%, ensuring that there are no significant voltage or current fluctuations when power is output to the load; second, the response delay should consistently meet the requirement of not exceeding 2 seconds to avoid affecting the overload protection effect due to delay fluctuations. If the verification results show no overload risk, and both power fluctuation and response delay meet the stability indicators, the system can be judged as having a stable power output result. If the verification results show an overload risk, or any stability indicator fails to meet the standard, the system is marked as needing optimization, and the system adjustment parameters need to be adjusted. The final stable power output result must record the timestamp, risk assessment result, and stability indicator compliance status to clearly indicate whether the system currently has stable power supply conditions.

[0171] For example, if the verification result at 09:11 shows no overload risk, the monitored power fluctuation is 0.6%, which is within the ±1% range, and the response delay is 1s, which is within 2s, then the system is considered to be in a stable power output state at 09:11. If the verification result at 09:21 shows no overload risk, the power fluctuation is 0.7%, which meets the requirements, and the response delay is 1s, then the system is considered to be in a stable power output state at 09:21. If the verification result at 09:26 shows a power overload risk, even if the monitored power fluctuation is 0.8%, which meets the standard, and the response delay is 1s, then the system is still considered to be in a state that needs optimization and needs to be returned to adjust the system adjustment parameters and re-verified.

[0172] In summary, this invention discloses a reverse current prevention control method for photovoltaic inverters based on power signal coupling, including steps such as acquiring and smoothing power and load signals, analyzing deviation-generated dynamic signals, selecting and adjusting parameters according to adaptation rules, fusing data to generate control commands, updating system parameters and performing overload verification, and monitoring stable state to obtain output results. This invention achieves closed-loop control of photovoltaic inverter reverse current prevention from signal processing to stable output through a dynamic regulation and overload protection mechanism driven by power signal coupling, effectively improving the accuracy of reverse current prevention control and the stability of system operation, and providing reliable technical support for the safety of photovoltaic grid connection.

[0173] Reference Figure 2 The second embodiment of the present invention provides a photovoltaic inverter anti-reverse current control system based on power signal coupling, comprising:

[0174] The data acquisition and processing module is used to acquire real-time power data and grid load data of the photovoltaic system and perform noise reduction processing to obtain smooth power signal and load signal. The load signal is subtracted from the smooth power signal to obtain the difference signal.

[0175] The fluctuation amplitude analysis module is used to analyze the signal timing changes and fluctuation frequency of the difference signal, and determine the amplitude range of the fluctuation amplitude and peak amplitude.

[0176] The historical trend extraction module extracts historical power data sequences from a preset historical database and performs trend analysis if the amplitude range exceeds a preset amplitude range, thereby obtaining the fluctuation trend.

[0177] The rule base matching module matches the fluctuation trend with a preset rule base to generate a dynamic signal;

[0178] The deviation analysis and adjustment module is used to compare the dynamic signal with the smoothed power signal, analyze the deviation range, and determine the adjustment parameters.

[0179] A control command generation module is used to generate a control command by integrating the adjustment parameters and the load signal if the deviation range does not exceed a preset deviation range.

[0180] The system regulation and monitoring module is used to update the regulation settings of the photovoltaic system according to the control command, and monitor the operating status in combination with the preset overload protection mechanism and preset stability index to obtain stable power output results.

[0181] It should be noted that the photovoltaic inverter anti-reverse current control system based on power signal coupling provided in this embodiment of the invention is used to execute all the process steps of the photovoltaic inverter anti-reverse current control method based on power signal coupling in the above embodiment. The working principle and beneficial effect of the two are one-to-one, so they will not be described again.

[0182] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a photovoltaic inverter anti-reverse current control program. When the processor executes the computer program, it implements the steps in the various photovoltaic inverter anti-reverse current control method embodiments described above, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, such as the data acquisition and processing module.

[0183] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0184] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0185] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0186] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0187] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0188] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0189] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for preventing reverse current in a photovoltaic inverter based on power signal coupling, characterized in that, include: Real-time power data and grid load data of the photovoltaic system are collected and noise-reduced to obtain smooth power signals and load signals. The load signals are subtracted from the smooth power signals to obtain the difference signals. Analyze the signal timing changes and fluctuation frequency of the difference signal to determine the amplitude range of the fluctuation amplitude and peak amplitude; If the amplitude range exceeds the preset amplitude range, historical power data sequences are extracted from the preset historical database and trend analysis is performed to obtain the fluctuation trend. The fluctuation trend is matched with a preset rule base to generate a dynamic signal; By comparing the dynamic signal with the smoothed power signal, the deviation range is analyzed, and the adjustment parameters are determined. If the deviation range does not exceed the preset deviation range, the adjustment parameters and the load signal are combined to generate a control command; The photovoltaic system's adjustment settings are updated according to the control commands, and the operating status is monitored in conjunction with the preset overload protection mechanism and preset stable state indicators to obtain a stable power output result.

2. The photovoltaic inverter anti-reverse current control method based on power signal coupling according to claim 1, characterized in that, The process involves collecting real-time power data and grid load data from the photovoltaic system, performing noise reduction processing to obtain smoothed power and load signals, and subtracting the load signal from the smoothed power signal to obtain a difference signal, including: Real-time power data and grid load data of photovoltaic systems are collected and integrated to obtain raw collected data. The original acquired data is subjected to noise reduction processing, outliers are screened and corrected, and the processed signal is obtained; The time alignment deviation of the processed signal is calibrated to obtain a smoothed power signal and load signal; The difference between the smoothed power signal and the load signal is calculated to obtain the difference signal.

3. The photovoltaic inverter anti-reverse current control method based on power signal coupling according to claim 1, characterized in that, The analysis of the signal timing changes and fluctuation frequency of the difference signal, and the determination of the amplitude range of the fluctuation amplitude and peak amplitude, includes: By analyzing the signal timing changes and fluctuation frequencies of the difference signal, the timing change characteristics and fluctuation frequency characteristics are obtained. Based on the time-series variation characteristics and the fluctuation frequency characteristics, fluctuation data and peak data are extracted to obtain the amplitude range.

4. The photovoltaic inverter anti-reverse current control method based on power signal coupling according to claim 1, characterized in that, If the amplitude range exceeds a preset amplitude range, historical power data sequences are extracted from a preset historical database and trend analysis is performed to obtain the fluctuation trend, including: The amplitude range is compared with a preset amplitude range, and a result indicating that the amplitude exceeds the preset range is obtained. Based on the results that exceed the preset range, historical power data sequences are extracted from a preset historical database to obtain matching historical data; By analyzing the fluctuation patterns of the historical matching data, the trend of fluctuations can be obtained.

5. The photovoltaic inverter anti-reverse current control method based on power signal coupling according to claim 1, characterized in that, The step of matching the fluctuation trend with a preset rule base to generate a dynamic signal includes: The fluctuation trend is matched with a preset rule base to obtain the trend matching result; Based on the trend matching results, the matching rules in the preset rule base are filtered to obtain the target matching rules; Based on the target adaptation rules, a dynamic signal for power fluctuation status is generated.

6. The photovoltaic inverter anti-reverse current control method based on power signal coupling according to claim 1, characterized in that, The process of comparing the dynamic signal with the smoothed power signal, analyzing the deviation range, and determining the adjustment parameters includes: By comparing the dynamic signal with the smoothed power signal, a signal comparison result is obtained; Based on the signal comparison results, the deviation range is analyzed to obtain deviation range data; The adjustment parameters are determined by combining the basic deviation range and the percentage of deviations exceeding the acceptable threshold in the deviation range data.

7. The photovoltaic inverter anti-reverse current control method based on power signal coupling according to claim 6, characterized in that, If the deviation range does not exceed a preset deviation range, the adjustment parameters and the load signal are integrated to generate a control command, including: If the deviation range does not exceed the preset deviation range, then the range compliance result is obtained; Based on the range compliance results, the adjustment parameters and the load signal are fused to obtain fused data; The fused data is matched with a preset adjustment response threshold to determine the initial power allocation ratio. The response speed parameters are then calibrated using preset dynamic monitoring data to obtain the adaptation parameters. Based on the adaptation parameters, control commands are generated to determine the final power allocation ratio and adjust the response speed parameters.

8. The photovoltaic inverter anti-reverse current control method based on power signal coupling according to claim 1, characterized in that, The process of updating the adjustment settings of the photovoltaic system according to the control command, and monitoring the operating status in conjunction with the preset overload protection mechanism and preset stability index to obtain a stable power output result includes: The system adjustment settings of the photovoltaic system are updated according to the control command to obtain the system adjustment parameters; Based on the system adjustment parameters, and combined with the preset overload protection mechanism, an overload risk verification is performed to obtain the verification result. Based on the verification results and combined with preset stability indicators, the operating status of the photovoltaic system is monitored and analyzed to obtain stable power output results.

9. A photovoltaic inverter anti-reverse current control system based on power signal coupling, characterized in that, include: The data acquisition and processing module is used to acquire real-time power data and grid load data of the photovoltaic system and perform noise reduction processing to obtain smooth power signal and load signal. The load signal is subtracted from the smooth power signal to obtain the difference signal. The fluctuation amplitude analysis module is used to analyze the signal timing changes and fluctuation frequency of the difference signal, and determine the amplitude range of the fluctuation amplitude and peak amplitude. The historical trend extraction module extracts historical power data sequences from a preset historical database and performs trend analysis if the amplitude range exceeds a preset amplitude range, thereby obtaining the fluctuation trend. The rule base matching module matches the fluctuation trend with a preset rule base to generate a dynamic signal; The deviation analysis and adjustment module is used to compare the dynamic signal with the smoothed power signal, analyze the deviation range, and determine the adjustment parameters. A control command generation module is used to generate a control command by integrating the adjustment parameters and the load signal if the deviation range does not exceed a preset deviation range. The system regulation and monitoring module is used to update the regulation settings of the photovoltaic system according to the control command, and monitor the operating status in combination with the preset overload protection mechanism and preset stability index to obtain stable power output results.

Images