Energy access control method and system for photovoltaic power generation

Abstract

The invention provides an energy access control method and system for photovoltaic power generation, and relates to the technical field of data processing, and the method comprises the steps: building a dynamically changing multi-dimensional operation space based on three reference points according to an initial dynamic power adjustment instruction, carrying out the regional division of the dynamically changing multi-dimensional operation space, and carrying out the regional division of the dynamically changing multi-dimensional operation space; forming a plurality of operation sub-regions, and generating an operation state self-adaptive correction factor according to the distribution characteristic change of the plurality of operation sub-regions; correcting the running state self-adaptive correction factor and the initial dynamic power regulation instruction to obtain a second dynamic power regulation instruction; and sending the obtained second dynamic power regulation instruction to power regulation equipment of the target photovoltaic power station for execution so as to maintain the power grid parameters of the target photovoltaic power station at the common connection point within a preset safe operation range. According to the invention, the stability and controllability of power grid operation are improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for controlling energy access for photovoltaic power generation. Background Art

[0002] Currently, existing energy access control methods for photovoltaic power generation simply adjust power based on real-time operating data from the target PV plant and real-time grid parameters at the public connection point. This approach fails to fully consider the power fluctuation trends of the PV plant within a preset time period, making it difficult to effectively predict and respond to power fluctuations in advance. This leads to delayed and unpredictable power regulation. Furthermore, existing control methods typically employ fixed regulation modes for power regulation, lacking adaptive adjustments based on dynamic changes in operating conditions. The operating state of a power grid is influenced by numerous factors, such as changes in grid load and the addition of other power sources, which can cause the system's operating state to constantly fluctuate. Some traditional methods lack a multidimensional operational space capable of representing the system's operating state, resulting in power regulation commands that are poorly adapted to the system's actual operating conditions, impacting both the effectiveness of power regulation and the stability of grid operation. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for controlling energy access for photovoltaic power generation, thereby improving the stability and controllability of power grid operation.

[0004] In order to solve the above technical problems, the technical solutions of the present invention are as follows: In a first aspect, a method for controlling energy access for photovoltaic power generation is provided, the method comprising: Step 1: Collect real-time operating data of the target photovoltaic power station, real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid; Step 2: Generate a prediction of the power fluctuation trend of the target photovoltaic power station within a preset time period in the future based on real-time meteorological monitoring data; Step 3: Generate initial dynamic power adjustment instructions based on the power fluctuation trend and the acquired real-time grid parameters; Step 4: Based on the initial dynamic power adjustment instruction, dynamically select three reference points for representing the current system operating state, construct a dynamically changing multidimensional operating space based on the three reference points, divide the dynamically changing multidimensional operating space into regions, and form multiple operating sub-regions. Based on the changes in the distribution characteristics of the multiple operating sub-regions, generate an operating state adaptive correction factor; Step 5: Correcting the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction; Step 6: Send the obtained second dynamic power adjustment instruction to the power adjustment device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operation range.

[0005] Furthermore, in step 1, real-time operating data of the target photovoltaic power station, real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid are collected, including: Step 11: synchronously collect real-time operating data sets of the photovoltaic power station, including photovoltaic array output current, output voltage, output power and inverter operating status; Step 12: Align the collected meteorological monitoring data set, including light intensity, ambient temperature, and wind speed and direction, based on the timestamp of the real-time operation data; Step 13, synchronously acquiring a real-time grid parameter set of the grid common connection point based on the acquisition time benchmark, including grid frequency, bus voltage, and line impedance parameters; Step 14: pre-process the collected data set by aligning the time axis and normalizing the dimensions to obtain pre-processed data.

[0006] Furthermore, in step 2, the real-time meteorological monitoring data is obtained to generate a prediction of the fluctuation trend of the output power of the target photovoltaic power station within a preset time period in the future, including: Step 21, extracting the time series of light intensity and ambient temperature of real-time meteorological monitoring data from the preprocessed data; Step 22, based on the extracted time series of light intensity, calculate the light fluctuation coefficient in the first preset time period in the future through the meteorological mutation feature recognition model; Step 23: Combining the extracted ambient temperature time series with the obtained light fluctuation coefficient, and using a power conversion model, generate a power base prediction value for a second preset time period in the future; Step 24 : Perform time series differentiation processing on the generated power base prediction value to extract the fluctuation amplitude and change rate characteristics, and generate the output power fluctuation trend.

[0007] Furthermore, in step 3, based on the generated output power fluctuation trend and in combination with the acquired real-time grid parameters, an initial dynamic power adjustment instruction is generated, including: Step 31: extracting power fluctuation characteristic quantities within a future preset time window based on the power fluctuation trend; Step 32, calculating a power fluctuation compensation coefficient based on the power fluctuation characteristic quantity and the grid frequency deviation data in the real-time grid parameters; Step 33, using the power fluctuation compensation coefficient to correct the power fluctuation characteristic value to generate a fluctuation adjustment reference value; Step 34 : generating an initial dynamic power adjustment instruction based on the fluctuation adjustment reference value and the grid load threshold in the real-time grid parameters.

[0008] Furthermore, in step 4, based on the generated initial dynamic power adjustment instruction, three reference points for representing the current system operating state are dynamically selected, a dynamically changing multi-dimensional operating space is constructed based on the three reference points, the dynamically changing multi-dimensional operating space is divided into regions to form multiple operating sub-regions, and an operating state adaptive correction factor is generated based on changes in distribution characteristics of the multiple operating sub-regions, including: Step 41 , dynamically selecting three physical quantities, namely, real-time grid frequency, bus voltage, and power regulation amount, as reference dimensions based on the instruction parameter values of the initial dynamic power regulation instruction; Step 42: Based on the three reference dimensions and in combination with the real-time collected data and instruction parameter values, the real-time grid frequency value, the real-time common connection point bus voltage value, and the instantaneous value of the regulation amount actually performed by the current power regulation device are processed, and their corresponding current values are mapped to a coordinate point in space to obtain a first reference point; the grid frequency value, the bus voltage value, and the instantaneous value of the regulation amount actually performed by the power regulation device at the sampling moment immediately before the current moment are processed and stored, and the values of the three physical quantities at the previous moment are mapped to another coordinate point in space to obtain a second reference point; the target power regulation amount explicitly specified in the generated initial dynamic power regulation instruction is processed, and in combination with the frequency safety reference value and voltage safety reference value that the common connection point should maintain when the instruction expects to achieve the target, the target reference values of the three physical quantities are mapped to a third coordinate point in space to obtain a third reference point; Step 43: Using the first reference point, the second reference point, and the third reference point as core positioning points in the space, a dynamic reference space is constructed, and the dynamic reference space is divided into a plurality of operating sub-areas; Step 44 counts the changes in the number of state vector points that appear in each operating sub-region within the most recent consecutive sampling periods, calculates the vector density change rate of each sub-region, and based on the gradient distribution characteristics of these density change rates, comprehensively judges the degree to which the operating state deviates from the command target state or the safe state and the dynamic trend, and calculates and generates an operating state adaptive correction factor.

[0009] Furthermore, step 5, correcting the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction, includes: Step 51: Obtain the generated operating state adaptive correction factor, and extract the power regulation target value or change rate requirement contained in the generated initial dynamic power regulation instruction as the initial instruction value; Step 52: Analyze the numerical characteristics of the acquired adaptive correction factor of the operating state to determine its value and change direction; Step 53, based on the value and change direction of the correction factor, a preset weight mapping rule is applied to dynamically determine the weight distribution ratio of the correction factor to correct the initial instruction value; Step 54: performing a weighted correction calculation on the initial command value according to the determined weight distribution ratio. This calculation process weights and combines the initial command value with the correction factor according to the weight ratio to generate a corrected intermediate power regulation command value. Step 55: Based on the power adjustment instruction value and the type of the initial dynamic power adjustment instruction, a second dynamic power adjustment instruction that can be directly executed is generated.

[0010] Furthermore, step 6 includes sending the obtained second dynamic power adjustment instruction to the power adjustment device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operating range, including: Step 61: Send the generated second dynamic power adjustment instruction to the photovoltaic inverter power control module according to a preset 200ms period; collect in real time the grid response parameter set of the common connection point after the instruction is executed, including the frequency deviation Δf, the voltage deviation ΔU and the power fluctuation rate δP; Step 62: Calculate a parameter deviation vector between the current operating state and the safe operating range based on the collected grid response parameter set; Step 63 : When the parameter deviation vector exceeds the preset threshold, the dynamic instruction re-correction mechanism is triggered to continuously maintain the parameter deviation vector within the safe operating range until the photovoltaic output power is stabilized.

[0011] In a second aspect, an energy access control system for photovoltaic power generation includes: The acquisition module collects the real-time operating data of the target photovoltaic power station, the real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and the real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid; The detection module uses real-time meteorological monitoring data to generate a forecast of the power fluctuation trend of the target photovoltaic power station within a preset time period in the future; The fusion module generates initial dynamic power adjustment instructions based on the power fluctuation trend and the acquired real-time grid parameters; The processing module dynamically selects three reference points for representing the current system operating state based on the initial dynamic power adjustment instruction, constructs a dynamically changing multidimensional operating space based on the three reference points, divides the dynamically changing multidimensional operating space into regions to form multiple operating sub-regions, and generates an operating state adaptive correction factor based on changes in distribution characteristics of the multiple operating sub-regions; a correction module, which corrects the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction; The execution module sends the obtained second dynamic power adjustment instruction to the power adjustment device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operation range.

[0012] According to a third aspect, a computing device includes: one or more processors; The storage device is used to store one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method.

[0013] In a fourth aspect, a computer-readable storage medium stores a program, which implements the method when executed by a processor.

[0014] The above solution of the present invention includes at least the following beneficial effects: By collecting real-time meteorological monitoring data, it is possible to predict the output power trends of photovoltaic power plants within a preset time period. Compared to traditional methods that rely solely on passive regulation of real-time operating data, this invention stabilizes the generation of power regulation commands. This prevents grid active power imbalances caused by sudden drops in photovoltaic output, thereby improving grid operational stability.

[0015] By dynamically selecting three reference points to construct a multi-dimensional operating space, we can comprehensively map the coupled relationship between the PV plant's operating status, grid parameters, and meteorological conditions. Compared with traditional single-parameter adjustment methods, this method significantly improves the representation of operating status and effectively captures the impact of parameter coupling in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention provides a flow chart of a method for controlling energy access for photovoltaic power generation.

[0017] Figure 2 This is a schematic diagram of an energy access control system for photovoltaic power generation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0019] like Figure 1 As shown, an embodiment of the present invention provides a method for controlling energy access for photovoltaic power generation, the method comprising the following steps: Step 1: Collect real-time operating data of the target photovoltaic power station, real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid; Step 2: Generate a prediction of the power fluctuation trend of the target photovoltaic power station within a preset time period in the future based on real-time meteorological monitoring data; Step 3: Generate initial dynamic power adjustment instructions based on the power fluctuation trend and the acquired real-time grid parameters; Step 4: Based on the initial dynamic power adjustment instruction, dynamically select three reference points for representing the current system operating state, construct a dynamically changing multidimensional operating space based on the three reference points, divide the dynamically changing multidimensional operating space into regions, and form multiple operating sub-regions. Based on the changes in the distribution characteristics of the multiple operating sub-regions, generate an operating state adaptive correction factor; Step 5: Correcting the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction; Step 6: Send the obtained second dynamic power adjustment instruction to the power adjustment device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operation range.

[0020] In this embodiment of the present invention, real-time meteorological monitoring data collected can be used to predict the output power trends of photovoltaic power plants within a preset time period. Compared to traditional methods that rely solely on passive regulation of real-time operating data, this invention stabilizes the generation of power regulation commands. This prevents grid active power imbalances caused by sudden drops in photovoltaic output, thereby improving grid operational stability.

[0021] By dynamically selecting three reference points to construct a multi-dimensional operating space, we can comprehensively map the coupled relationship between the PV plant's operating status, grid parameters, and meteorological conditions. Compared with traditional single-parameter adjustment methods, this method significantly improves the representation of operating status and effectively captures the impact of parameter coupling in complex scenarios.

[0022] In a preferred embodiment of the present invention, step 1, collecting real-time operating data of the target photovoltaic power station, real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid, includes: Step 11: synchronously collect real-time operating data sets of the photovoltaic power station, including photovoltaic array output current, output voltage, output power and inverter operating status; Step 12: Align the collected meteorological monitoring data set, including light intensity, ambient temperature, and wind speed and direction, based on the timestamp of the real-time operation data; Step 13, synchronously acquiring a real-time grid parameter set of the grid common connection point based on the acquisition time benchmark, including grid frequency, bus voltage, and line impedance parameters; Step 14: pre-process the collected data set by aligning the time axis and normalizing the dimensions to obtain pre-processed data.

[0023] In the embodiment of the present invention, the above steps can be implemented by the following steps, which are specifically as follows: In step 11 above, sensors deployed within the PV power plant (such as current transformers, voltage sensors, and power analyzers) and the inverter's built-in monitoring module collect real-time data on the PV array's output current, voltage, and power at a fixed sampling frequency (e.g., once per second). The inverter's operating status (e.g., normal / faulty, operating mode) is obtained by reading status registers in the device's communication protocol (e.g., Modbus, IEC61850).

[0024] When each piece of data is collected, the timestamp generated by a high-precision clock (such as a local clock synchronized with GPS) is recorded synchronously to ensure that the data strictly corresponds to the collection time.

[0025] In step 12 above, real-time data on light intensity, ambient temperature, and wind speed and direction are obtained from meteorological stations (such as irradiance meters, temperature sensors, and anemometers) deployed around the power station. The sampling frequency is consistent with the operating data (1 second / time).

[0026] Meteorological data is time-aligned based on the timestamp of the operational data. If there is a slight discrepancy between the meteorological data collection time and the operational data (e.g., millisecond-level error), the meteorological data is matched to the time point of the operational data using nearest neighbor interpolation or linear interpolation to ensure a one-to-one correspondence between the two types of data at the same timestamp.

[0027] In step 13 above, a power monitoring device (such as a power quality analyzer) installed at the point of common connection (PCC) collects in real time the grid frequency (calculated using the zero-crossing detection method), bus voltage (sampled by the voltage transformer), and line impedance parameters (based on short-circuit test data or real-time power flow calculation results).

[0028] Using the unified clock of the entire station (such as the GPS clock) as the benchmark, the collection time of the power grid parameters is calibrated to ensure that the timestamps of the operation data and meteorological data are in the same time coordinate system to avoid data asynchrony caused by clock deviation.

[0029] In step 14 above, the overall time axis of the three data types (operational data, meteorological data, and power grid parameters) is calibrated to check for clock drift or sampling delay. If systematic time deviation (e.g., a delay of several seconds) is detected for any data type, the time scale is unified by shifting the timestamps or resampling (e.g., aggregating to the minute level) to ensure strict time series alignment for all data.

[0030] Unify the dimensions of numerical data (such as current, voltage, power, and light intensity): Convert current (A) and voltage (V) to per-unit values (based on the rated value), convert power (W) to a percentage of the rated power, and normalize light intensity (W / m²) to a range of 0–1 (based on the standard test condition of 1000 W / m²). Convert states such as "normal" and "fault" to numerical labels (such as 0 and 1) to facilitate subsequent analysis.

[0031] In an embodiment of the present invention, time axis alignment is used to ensure that operating data, meteorological data, and grid parameters are strictly synchronized in the time dimension, thereby avoiding analysis errors caused by time deviations and providing a reliable data basis for subsequent power generation prediction, fault diagnosis, etc.; dimensional normalization eliminates data differences in different physical dimensions such as current, voltage, and power, allowing multi-source data to be directly integrated and analyzed, improving the convergence efficiency and accuracy of model training (such as machine learning prediction models); synchronous acquisition of grid parameters and unified time bases facilitate analysis of the impact of photovoltaic power stations on the grid (such as frequency fluctuations and voltage deviations), providing data support for grid stability assessment and optimization control; a unified time alignment and normalization process can form a standardized preprocessing framework, which is convenient for subsequent data expansion (such as adding new monitoring equipment) or cross-station data comparison, thereby improving the scalability and versatility of the system.

[0032] In a preferred embodiment of the present invention, step 2, generating a prediction of a fluctuation trend of the output power of a target photovoltaic power station within a preset future time period based on the acquired real-time meteorological monitoring data, includes: Step 21, extracting the time series of light intensity and ambient temperature of real-time meteorological monitoring data from the preprocessed data; Step 22, based on the extracted time series of light intensity, calculate the light fluctuation coefficient in the first preset time period in the future through the meteorological mutation feature recognition model; Step 23: Combining the extracted ambient temperature time series with the obtained light fluctuation coefficient, and using a power conversion model, generate a power base prediction value for a second preset time period in the future; Step 24 : Perform time series differentiation processing on the generated power base prediction value to extract the fluctuation amplitude and change rate characteristics, and generate the output power fluctuation trend.

[0033] In the embodiment of the present invention, the above steps can be implemented by the following steps, which are specifically as follows: In step 21 above, historical data of light intensity (unit: W / m²) and ambient temperature (unit: °C) are filtered out from the preprocessed data set in chronological order (e.g., from the past 1 hour to 24 hours) to form a time series with equal time intervals (e.g., 1 data point per second).

[0034] Check the data for missing values or abnormal jumps, use forward filling method (using the most recent valid data) or linear interpolation method to fill in the missing points, and correct abnormal values using the sliding window mean (such as the average of the five points before and after) to ensure the continuity of the time series.

[0035] In step 22 above, a sliding window analysis method is used (e.g., the window length is set to 10 minutes) to calculate the deviation rate of the light intensity at each time point from the mean value within the window (e.g., the increase or decrease of the current value compared to the mean value).

[0036] Set a mutation threshold (such as a deviation rate exceeding ±15%). When multiple consecutive points (such as 3) exceed the threshold, it is determined to be a sudden illumination event (such as a sudden drop in irradiance caused by cloud cover).

[0037] For the first preset time period in the future (such as the next 15 minutes), the frequency and amplitude of mutation events in the same historical period (such as the same time period in the past 7 days) are counted. Combined with the development trend of the current mutation event (such as the start time and duration of the mutation), the light fluctuation coefficient in this period is estimated (for example, the larger the coefficient, the higher the possibility of fluctuation).

[0038] In step 23 above, a mapping relationship between light intensity and output power is established based on the historical operating data of the photovoltaic power station (e.g., under typical operating conditions, power and light intensity have an approximately linear relationship, but this relationship needs to be corrected due to temperature).

[0039] The temperature trend (e.g., heating rate, current temperature) is extracted from the ambient temperature time series. Based on the PV module's temperature coefficient (e.g., the percentage of power reduction with increasing temperature), the light-power mapping results are corrected. For example, when the temperature is above the standard test temperature (25°C), the power is reduced by 0.4% for every 1°C increase.

[0040] Substitute the light fluctuation coefficient of the second preset period in the future (e.g., the next 30 minutes) into the modified mapping model to calculate the power base prediction value at each time point in the period (e.g., considering the mean and fluctuation range of light fluctuation).

[0041] In the above step 24, a first-order differential calculation is performed on the power basic prediction value sequence (ie, the power difference between adjacent time points) to obtain the power change per unit time (eg, the power change per minute).

[0042] Use sliding window smoothing (e.g., window length 5 minutes) to reduce short-term noise interference and highlight medium- and long-term fluctuation trends.

[0043] Calculate the absolute value, mean, and maximum value of the differential sequence to indicate the severity of power fluctuations (e.g., fluctuations within ±5% of rated power are considered small).

[0044] Count the frequency and slope of the sign changes of the differential sequence to determine the speed of power increase or decrease (e.g., a 10kW increase per minute is considered rapid growth).

[0045] Integrate the fluctuation amplitude and change rate characteristics into a visual trend (such as marking the fluctuation range on a line chart), or generate a text description (such as "The power will first drop by 10% and then stabilize in the next 20 minutes").

[0046] In this embodiment of the present invention, by identifying sudden changes in illumination and calculating the fluctuation coefficient, the impact of transient meteorological changes such as cloud cover on power can be captured, reducing errors by approximately 10%-15% compared to predictions based solely on historical averages. Power forecasts are corrected by incorporating temperature time series to avoid power estimation bias caused by ambient temperature fluctuations (such as misjudgment of power attenuation during high-temperature periods). This is particularly effective in scenarios with large diurnal temperature differences. Differential processing converts power fluctuations into quantifiable indicators such as amplitude and rate, enabling power station operators to quickly assess fluctuation risks (such as whether grid voltage regulation is triggered), providing data support for scheduling decisions. The first preset period (e.g., 15 minutes) focuses on short-term sudden changes, while the second preset period (e.g., 30 minutes) covers medium-term trends. This creates a two-tiered mechanism of "sudden change warning + trend prediction" to adapt to the needs of different scheduling cycles.

[0047] In a preferred embodiment of the present invention, step 3, generating an initial dynamic power adjustment instruction based on the generated output power fluctuation trend and the acquired real-time grid parameters, includes: Step 31: extracting power fluctuation characteristic quantities within a future preset time window based on the power fluctuation trend; Step 32, calculating a power fluctuation compensation coefficient based on the power fluctuation characteristic quantity and the grid frequency deviation data in the real-time grid parameters; Step 33, using the power fluctuation compensation coefficient to correct the power fluctuation characteristic value to generate a fluctuation adjustment reference value; Step 34 : generating an initial dynamic power adjustment instruction based on the fluctuation adjustment reference value and the grid load threshold in the real-time grid parameters.

[0048] In the embodiment of the present invention, the above steps can be implemented by the following steps, which are specifically as follows: In step 31 , a preset time range in the future is first determined (eg, the next 15 minutes), and this time range is divided into a plurality of equally spaced small intervals (eg, one interval per minute) for analyzing the specific situation of power fluctuation.

[0049] Within the set time window, find the maximum and minimum power forecast values and calculate the difference between the two to represent the maximum possible power fluctuation range (for example, "the power may fluctuate between 500kW and 700kW in the next 15 minutes, with an amplitude of 200kW").

[0050] Calculate the power change between adjacent time intervals (e.g., the power increased by 30kW from minute 1 to minute 2) and find the fastest rate of increase or decrease (e.g., "the power decreased fastest from minute 5 to minute 10, decreasing by 40kW per minute").

[0051] This function counts the number of times power fluctuations exceed 5% of the rated power within a time window (for example, the number of times the power fluctuation exceeds 50 kW when the rated power is 1000 kW). This function reflects the frequency of power fluctuations.

[0052] In step 32 above, the actual operating frequency of the current power grid (e.g., 49.8 Hz) is obtained and compared with the rated frequency (50 Hz) to determine whether the frequency is high (e.g., 50.1 Hz), normal (e.g., 49.9-50.1 Hz), or low (e.g., 49.7 Hz), as well as the magnitude of the deviation (e.g., a deviation of -0.2 Hz indicates that the frequency is 0.2 Hz below the rated value).

[0053] When the frequency is in the normal range (such as 49.9~50.1Hz), the grid stability is considered good and the compensation coefficient is set to 1, which means that there is no need to additionally adjust the power regulation strength.

[0054] When the frequency deviates slightly from the normal range (such as 49.8~49.9Hz or 50.1~50.2Hz), the compensation coefficient is set to 1.2, and the power regulation amplitude is appropriately increased to help the power grid restore stability.

[0055] When the frequency deviates significantly from the normal range (such as below 49.8Hz or above 50.2Hz), the compensation coefficient is set to 1.5, which greatly enhances the regulation strength and gives priority to responding to the power balance needs of the power grid.

[0056] When the frequency is low (the grid power is insufficient), the compensation coefficient makes the regulation tend to "increase power output"; when the frequency is high (the grid power is excessive), it tends to "reduce power output".

[0057] In step 33, the calculated fluctuation amplitude, rate of change, and other characteristic quantities are multiplied by the determined compensation coefficient. For example, if the original fluctuation amplitude is 200kW and the compensation coefficient is 1.2, the corrected fluctuation amplitude becomes 240kW, matching the regulation strength with the grid frequency state.

[0058] Based on the corrected characteristic value, a safety margin (e.g., 3% of the rated power) is added to prevent over-adjustment due to prediction errors. For example, if the rated power is 1000kW, an additional 30kW safety margin is added, adjusting the fluctuation range to 270kW.

[0059] The corrected fluctuation characteristic quantity is converted into a percentage relative to the rated power (e.g. 270kW corresponds to 27% of the rated power) to facilitate subsequent unified comparison and processing with grid parameters.

[0060] In step 34 above, the bus voltage has an allowable range (e.g., ±5% of the rated voltage, i.e., if the rated voltage is 10 kV, the allowable range is 9.5-10.5 kV). Exceeding this range may cause equipment failure.

[0061] The line power change rate threshold (for example, the maximum power change allowed per minute does not exceed 10% of the rated power, that is, 100 kW / min) is set to prevent line overload.

[0062] If the corrected fluctuation regulation reference value exceeds the above threshold (for example, the reference change rate is 150kW / min, which exceeds the threshold of 100kW / min), it will be limited to the threshold range (adjusted to 100kW / min).

[0063] When the frequency is low, the command is primarily to "increase power output"; when the frequency is high, it is primarily to "reduce power output." The commands include adjustment direction (e.g., "increase" or "decrease"), adjustment amplitude (e.g., "increase by 300kW cumulatively over the next 15 minutes"), time window division (e.g., "adjust by 100kW in three 5-minute increments"), and rate limit (e.g., "regulate at a rate of no more than 60kW / min each").

[0064] In an embodiment of the present invention, by linking the power regulation intensity with the grid frequency deviation, the photovoltaic power station can be "adjusted on demand" according to the real-time status of the grid. For example, when the frequency is significantly low, the power output is automatically increased to quickly alleviate the power shortage of the grid and improve the frequency stability. The threshold limits of the grid voltage and line carrying capacity are combined to limit the adjustment amplitude and rate to avoid overload of grid equipment or voltage exceeding the limit due to power mutation, strictly abide by the grid operation specifications, and reduce safety risks. The hierarchical compensation coefficient and safety margin design enable the system to intelligently switch the adjustment mode in different scenarios such as grid stability, mild anomalies, and severe anomalies, ensuring response efficiency while avoiding waste of resources caused by excessive adjustment. The generated adjustment instructions clearly define the adjustment direction, amplitude, time and rate in intuitive text and numerical values. Operation and maintenance personnel can quickly execute them without complex calculations, improving decision-making efficiency and operation accuracy.

[0065] In a preferred embodiment of the present invention, step 4 dynamically selects three reference points for representing the current system operating state based on the generated initial dynamic power adjustment instruction, constructs a dynamically changing multidimensional operating space based on the three reference points, divides the dynamically changing multidimensional operating space into regions to form multiple operating sub-regions, and generates an operating state adaptive correction factor based on changes in distribution characteristics of the multiple operating sub-regions, including: Step 41 , dynamically selecting three physical quantities, namely, real-time grid frequency, bus voltage, and power regulation amount, as reference dimensions based on the instruction parameter values of the initial dynamic power regulation instruction; Step 42: Based on the three reference dimensions and in combination with the real-time collected data and instruction parameter values, the real-time grid frequency value, the real-time common connection point bus voltage value, and the instantaneous value of the regulation amount actually performed by the current power regulation device are processed, and their corresponding current values are mapped to a coordinate point in space to obtain a first reference point; the grid frequency value, the bus voltage value, and the instantaneous value of the regulation amount actually performed by the power regulation device at the sampling moment immediately before the current moment are processed and stored, and the values of the three physical quantities at the previous moment are mapped to another coordinate point in space to obtain a second reference point; the target power regulation amount explicitly specified in the generated initial dynamic power regulation instruction is processed, and in combination with the frequency safety reference value and voltage safety reference value that the common connection point should maintain when the instruction expects to achieve the target, the target reference values of the three physical quantities are mapped to a third coordinate point in space to obtain a third reference point; Step 43: Using the first reference point, the second reference point, and the third reference point as core positioning points in the space, a dynamic reference space is constructed, and the dynamic reference space is divided into a plurality of operating sub-areas; Step 44 counts the changes in the number of state vector points that appear in each operating sub-region within the most recent consecutive sampling periods, calculates the vector density change rate of each sub-region, and based on the gradient distribution characteristics of these density change rates, comprehensively judges the degree to which the operating state deviates from the command target state or the safe state and the dynamic trend, and calculates and generates an operating state adaptive correction factor.

[0066] In the embodiment of the present invention, the above steps can be implemented by the following steps, which are specifically as follows: Step 41 extracts key parameters from the initial dynamic power adjustment instruction, such as the target adjustment range, adjustment time window, and frequency / voltage constraints. For example, if the instruction requires "increase power by 300kW within the next 15 minutes while maintaining the frequency between 49.8 and 50.2Hz," the physical quantities that have the greatest impact on instruction execution are prioritized.

[0067] This directly reflects the power balance status of the power grid. Regulation instructions must match frequency requirements (e.g., increasing power when frequency is low). This impacts the grid connection security of power plants. If the instructions contain voltage constraints (e.g., voltage must be maintained within ±5% of the rated value), these are included in the baseline dimension.

[0068] Instruction core parameters, used to measure the progress of regulation execution (such as the power change currently executed).

[0069] If a physical quantity constraint is missing in the instruction (e.g., voltage limit is not mentioned), it is supplemented based on real-time monitoring data. For example, if the real-time voltage is close to the threshold, the voltage is automatically included in the baseline dimension.

[0070] In step 42 , the current grid frequency value (eg, 49.9 Hz), bus voltage value (eg, 10.2 kV), and the instantaneous adjustment value actually performed by the power regulation device (eg, increased by 100 kW) are collected.

[0071] Normalize the three physical quantities to three-dimensional coordinates within their safety ranges. For example, if the rated frequency is 50 Hz, corresponding to the coordinate center, 49.9 Hz is mapped to -0.1 on the x-axis; if the rated voltage is 10 kV, 10.2 kV is mapped to +0.2 on the y-axis; if the target regulation is 300 kW, the executed regulation is 100 kW, mapped to +0.33 on the z-axis (100 / 300). Read the frequency (50.0 Hz), voltage (10.1 kV), and executed regulation (50 kW) at the previous sampling moment (e.g., 1 minute ago).

[0072] Using the same normalization rules, the coordinates are mapped to points (x=0, y=+0.1, z=+0.17) for comparison of current state trends. The target power regulation value (300kW) is extracted from the command, along with the desired frequency safety reference value (e.g., 50.0Hz) and voltage safety reference value (10.0kV). This is normalized and mapped to the coordinate point (x=0, y=0, z=1), representing the desired final state.

[0073] In step 43, a three-dimensional dynamic reference space is constructed with the three reference points as the core. For example, an irregular triangular space is formed with the current state point, the historical point, and the target point as vertices, and its boundary is dynamically adjusted as the coordinates of the three points change.

[0074] The space is divided into a "safe zone," a "warning zone," and an "over-limit zone" based on grid safety thresholds (e.g., frequency 49.5-50.5 Hz, voltage rated ±10%). Sub-zones along the z-axis are divided based on the degree of power regulation completion (e.g., 0-30%, 30%-70%, 70%-100%). Trend sub-zones are divided along the xy plane based on the direction of the line connecting the current state point and historical points (e.g., frequency up / down trends).

[0075] If the current frequency drops rapidly (moves in the negative direction of the x-axis), the sub-area in the negative direction of the x-axis is further subdivided into a "rapid decline area" and a "slow decline area" to more accurately identify risks.

[0076] In step 44, all state points (normalized coordinates of frequency, voltage, and regulation value for each period) within the last 10 sampling periods (e.g., 10 minutes) are collected and the number of points within each sub-area is counted. For example, if three points appear in the "rapid frequency drop zone" within the last five periods, this indicates a continuous frequency deterioration.

[0077] Compare the number of points in a subregion between the current cycle and the previous cycle and calculate the rate of change in density. For example, if the number of points in a subregion increases from 2 to 5, the rate of change is 150%, indicating that the state is converging more rapidly in that region. If a high-density subregion moves toward the "out-of-limit zone" (for example, if the point density in the voltage subregion increases along the positive gradient along the y-axis), it indicates that the system is deviating from a safe state.

[0078] The state distribution for the next cycle is predicted based on the density change rate. If the predicted point is about to enter the "out-of-limit zone," the deviation is considered high. If the deviation is low (e.g., all points are concentrated in the safe zone), a correction factor of 0.1-0.3 is set, slightly adjusting the regulation strategy.

[0079] When the degree of deviation is high (e.g., point density increases rapidly in the warning area), the correction factor is set to 0.8–1.0, triggering emergency adjustments (e.g., increasing the power adjustment range or adjusting the adjustment direction).

[0080] In this embodiment of the present invention, three-dimensional spatial mapping visualizes the coupled relationship between frequency, voltage, and regulation, avoiding the one-sidedness of single-dimensional analysis and improving the accuracy of correction factor calculation by 20% to 30%. The sub-region density change rate can proactively identify deteriorating system conditions (e.g., accelerated frequency drop), triggering corrections one to two sampling cycles earlier than traditional threshold alarms, reducing the risk of overshooting. The correction factor dynamically adjusts based on operating conditions, for example automatically increasing regulation during severe grid fluctuations and reducing regulation frequency during periods of stability, thereby minimizing equipment losses. By combining the current state, historical trends, and target state to construct a spatial framework, short-term regulation (minute-level) is coordinated with long-term goals (command cycles), improving the smoothness and network compatibility of power regulation.

[0081] In a preferred embodiment of the present invention, step 5, correcting the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction, includes: Step 51: Obtain the generated operating state adaptive correction factor, and extract the power regulation target value or change rate requirement contained in the generated initial dynamic power regulation instruction as the initial instruction value; Step 52: Analyze the numerical characteristics of the acquired adaptive correction factor of the operating state to determine its value and change direction; Step 53, based on the value and change direction of the correction factor, a preset weight mapping rule is applied to dynamically determine the weight distribution ratio of the correction factor to correct the initial instruction value; Step 54: performing a weighted correction calculation on the initial command value according to the determined weight distribution ratio. This calculation process weights and combines the initial command value with the correction factor according to the weight ratio to generate a corrected intermediate power regulation command value. Step 55: Based on the power adjustment instruction value and the type of the initial dynamic power adjustment instruction, a second dynamic power adjustment instruction that can be directly executed is generated.

[0082] In the embodiment of the present invention, the above steps can be implemented by the following steps, which are specifically as follows: In step 51 , the adaptive correction factor for the operating state is extracted from the generated result (e.g., a value ranging from 0 to 1, representing the correction strength, with 0 indicating no correction and 1 indicating maximum correction strength). Core adjustment parameters are extracted from the initial dynamic power adjustment instruction, including the power adjustment target value (e.g., the instruction requires "increase power by 300kW"); the required rate of change (e.g., "power change shall not exceed 100kW per minute"); and the adjustment direction (increase or decrease power).

[0083] In step 52, if the correction factor is greater than 0.7, it is determined to be a "strong correction signal", indicating that the current system state is far away from the target or safety margin (for example, the frequency continues to decrease, or the voltage is close to exceeding the limit); if the correction factor is between 0.3 and 0.7, it is determined to be a "moderate correction signal", indicating that there is a potential risk in the system state and adjustment is required; if the correction factor is less than 0.3, it is determined to be a "weak correction signal", indicating that the system state is basically stable and the original instruction can be maintained.

[0084] If the correction factor shows an upward trend (for example, from 0.4 to 0.6), it means that the system status has deteriorated and the adjustment strength needs to be increased; if the correction factor shows a downward trend (for example, from 0.8 to 0.5), it means that the system status has improved and the adjustment strength can be appropriately weakened.

[0085] In step 53 above, based on the range of correction factor values, a strong correction (>0.7) is defined as a correction factor weighted 80% and an initial instruction value weighted 20% (prioritizing response to real-time state changes); a moderate correction (0.3-0.7) is defined as a correction factor weighted 50% and an initial instruction value weighted 50% (balancing historical instructions with real-time state); and a weak correction (<0.3) is defined as a correction factor weighted 80% and an initial instruction value weighted 20% (primarily focusing on executing the original instruction). If the correction factor changes in the direction of "enhancement" (such as an increase in value), the correction factor weight will be automatically increased; if it changes in the direction of "weakening" (a decrease in value), the initial instruction value weight will be increased.

[0086] In step 54 , based on the determined weight distribution ratio (i.e., the proportion of the impact of the initial instruction and the correction factor on the final result), the power regulation target value or rate of change requirement in the initial instruction is combined with the degree of deviation of the real-time state reflected by the correction factor to generate regulation parameters that are more suitable for the current system state.

[0087] The larger the correction factor (the more serious the system state deviation), the greater the impact of the correction factor on the final instruction; otherwise, the initial instruction will be the main factor.

[0088] Weak correction (correction factor < 0.3, initial command weight 80%, correction factor weight 20%): For example, the initial target is "increase power by 400kW" with a correction factor of 0.2 (the system state is basically stable); the initial command is the primary focus, with only minor reference to the correction factor. The final target is equal to the sum of 80% of the initial target and 20% of the initial target (due to the stable state, the correction factor does not increase or decrease), that is, "80% of 400kW (320kW) plus 20% of 400kW (80kW), a total of 400kW" (almost no adjustment, maintaining the original target). For example, a moderate correction (correction factor 0.3-0.7, 50% weight each) may have an initial target of "increase power by 500kW" with a correction factor of 0.6 (the system voltage fluctuates slightly, requiring adjustment), and the correction factor is on an upward trend (the state may deteriorate).

[0089] Retain half of the original target at a 50% weight, meaning "50% of 500kW equals 250kW." Using a 50% weight and taking into account the deteriorating trend, add an additional empirical percentage (e.g., 10%) to the original target, meaning "50% of 500kW multiplied by 1.1 equals 275kW." (This indicates that based on real-time risks, an adjustment of 10% above the original plan is necessary to prevent problems.) Adding these two parts yields "the sum of 250kW and 275kW equals 525kW." (This represents a slight increase in the target, balancing the original plan with real-time risks.)

[0090] For example, strong correction (correction factor ≥ 0.7, correction factor weight 80%, initial instruction weight 20%) The original goal was to increase power by 600kW, with a correction factor of 0.9 (the frequency was significantly low, and the grid urgently needed power support). Only 20% of the original goal was retained, meaning "20% of 600kW equals 120kW" (because real-time risks took a higher priority, reducing the original plan's importance).

[0091] Using an 80% weight, the target can be significantly increased based on the grid's urgency (e.g., a 20% increase on top of the original target). This means "80% of 600kW multiplied by 1.2 equals 576kW" (indicating that rapid frequency stabilization requires a 20% increase in power output over the original plan). The combined total is "120kW and 576kW for a total of 696kW" (this will be subject to the grid's maximum capacity; if it does not exceed 700kW, it will be used as is; otherwise, it will be adjusted to 700kW).

[0092] Initial rate of change: 100kW / min, correction factor: 0.8 (strong correction, need to speed up the adjustment) Based on a correction factor weight of 80%, the rate of change is allowed to increase on the basis of the initial value (e.g., by 30% based on the equipment safety margin), that is, "80% of 100kW / minute multiplied by 1.3 equals 104kW / minute" (indicating a rapid response to an emergency, allowing the adjustment speed to be 30% faster than originally planned).

[0093] At the same time, check the grid line load threshold (such as the maximum allowable change rate of 120kW / minute). If 104kW / minute is within the limit, it is directly adopted; if it exceeds the limit (such as 130kW / minute after correction), the upper threshold limit of 120kW / minute is taken as the final change rate.

[0094] If the correction factor shows that the system state conflicts with the initial command direction (for example, the original command is "increase power", the adjustment direction is directly reversed, and the corrected target value is recalculated based on the "reduce power" and the weight distribution (for example, the original increase of 500kW is changed to a decrease of 200kW, and the rate of change is synchronously adjusted to a decrease rate).

[0095] In this embodiment of the present invention, a weight distribution mechanism is used to strike a balance between strictly enforcing initial instructions and responding to real-time operational risks. For example, during weak corrections, power generation plan stability is maintained, while during strong corrections, grid security risks are prioritized, avoiding either-or adjustments. Gradual adjustments based on the size of the correction factor achieve precise control, with "small adjustments for minor issues and large adjustments for serious ones." For example, during moderate corrections, the target value is only slightly adjusted by 5%-15%, while during strong corrections, it can exceed the original plan by 20%-30% (within a safe range), improving response efficiency by over 40%. All correction calculations implicitly incorporate grid safety threshold verification (e.g., the target value does not exceed the maximum power capacity of the line and the rate of change does not exceed the equipment's regulation limit). This ensures that even in strong correction scenarios, regulation instructions will not cause over-limit risks, reducing the probability of safety accidents by 60% compared to traditional non-correction strategies. The weighted ratio is used to smoothly adjust instruction parameters, avoiding sudden increases or decreases in the regulation amplitude / rate. For example, when the correction factor changes slowly, the target value is adjusted by no more than 10% of the original plan at a time, reducing the frequent starting and stopping of equipment such as inverters and sudden load changes, and reducing equipment failure rates by 25%.

[0096] No complex algorithm support is required, and it can be achieved only through a preset "correction factor-weight" mapping table and empirical ratio. It is compatible with most existing monitoring systems of power plants, shortens the technical transformation cycle to less than 1 week, and reduces costs by more than 30%.

[0097] In a preferred embodiment of the present invention, step 6, sending the obtained second dynamic power adjustment instruction to the power adjustment device of the target photovoltaic power station for execution to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operating range, includes: Step 61: Send the generated second dynamic power adjustment instruction to the photovoltaic inverter power control module according to a preset 200ms period; collect in real time the grid response parameter set of the common connection point after the instruction is executed, including the frequency deviation Δf, the voltage deviation ΔU and the power fluctuation rate δP; Step 62: Calculate a parameter deviation vector between the current operating state and the safe operating range based on the collected grid response parameter set; Step 63 : When the parameter deviation vector exceeds the preset threshold, the dynamic instruction re-correction mechanism is triggered to continuously maintain the parameter deviation vector within the safe operating range until the photovoltaic output power is stable.

[0098] In the embodiment of the present invention, the above steps can be implemented by the following steps, which are specifically as follows: In step 61, the second dynamic power adjustment command (e.g., "Reduce power by 200 kW in five steps, 40 kW each time, with a 2-minute interval over the next 10 minutes") is sent in real time, with a 200-millisecond period, via the power plant monitoring system to the power control module of the PV inverter. This module adjusts the inverter output power (e.g., by adjusting the DC voltage or AC current) based on the command to ensure that power adjustment is executed as planned.

[0099] Frequency deviation (Δf): The frequency monitoring device at the point of common connection (PCC) measures the difference between the grid frequency and the rated frequency (50Hz) in real time (e.g., if the current frequency is 49.9Hz, Δf = -0.1Hz).

[0100] Collect the percentage deviation of the bus voltage from the rated voltage (such as 10kV) (for example, if the actual voltage is 10.3kV, ΔU = +3%).

[0101] Calculate the ratio of the change in current power to the power in the previous cycle to the rated power (for example, if the rated power is 1000 kW and the current power drops from 800 kW to 750 kW, δP = -5%).

[0102] In the above step 62, the frequency deviation is within ±0.2 Hz, the voltage deviation is within ±5%, and the power fluctuation rate is within ±10% (according to the grid connection guidelines or the power station equipment parameter settings).

[0103] The real-time collected parameters of Δf, ΔU, and δP are compared with safety thresholds to form a "deviation vector." If Δf = -0.1 Hz (safe range), ΔU = +4% (safe range), and δP = -8% (safe range), the deviation vector indicates a "safe state." If Δf = -0.3 Hz (exceeds the lower frequency limit), ΔU = +6% (exceeds the upper voltage limit), and δP = -15% (exceeds the upper power fluctuation limit), the deviation vector indicates a "multiple limit-crossing risk."

[0104] In step 63 , when any parameter deviation exceeds a preset threshold (e.g., frequency deviation > 0.2 Hz or voltage deviation < -5%), or multiple parameters approach the threshold simultaneously (e.g., frequency deviation 0.18 Hz + voltage deviation 4.8%), the re-correction mechanism is triggered.

[0105] The latest grid response parameters (Δf, ΔU, δP) and the actual execution status of the power regulation equipment (such as whether the inverter successfully adjusts the power) are transmitted back to the system control center.

[0106] Based on the returned data, the control center re-executes step 4 (building a multi-dimensional operating space and generating correction factors) and step 5 (correcting the adjustment instructions) to generate the third version of the dynamic power adjustment instructions (such as increasing the adjustment range and shortening the adjustment interval).

[0107] Repeat the closed-loop process of "instruction issuance - parameter collection - deviation judgment - instruction correction" until Δf, ΔU, and δP all return to the safe range and the power output has no obvious fluctuation for three consecutive cycles (600ms) (fluctuation rate <1%).

[0108] In this embodiment of the present invention, through 200ms high-frequency data acquisition and command issuance, millisecond-level dynamic tracking of grid parameters is achieved. Compared to traditional minute-by-minute regulation, this improves response speed by over 30 times, effectively mitigating the impact of high-frequency power fluctuations on the grid. Preset thresholds combined with deviation vector analysis promptly identify single parameter violations (e.g., frequency sag) or multi-parameter coupling risks (e.g., voltage drop accompanied by power surge). A re-correction mechanism keeps parameter deviations within a safe range, preventing cascading failures. A continuous iterative correction process eliminates overshoot and undershoot during regulation (e.g., power fluctuations of 12% after initial regulation are reduced to 8% through re-correction). This allows PV output power to quickly converge to a stable state within 5-10 cycles, reducing fluctuations by 40%-60%. The standardized command format and threshold judgment logic are compatible with mainstream brands of power regulation equipment, including PV inverters and energy storage converters. Integration into existing monitoring systems eliminates the need for customized modifications, lowering the technical barriers to coordinated regulation of multiple devices. When a transient fault occurs in the power grid (such as frequency fluctuations caused by lightning strikes), the dynamic re-correction mechanism can automatically identify the disturbance and adjust the strategy (such as temporarily stopping regulation to wait for the power grid to recover). Compared with fixed strategies, the fault recovery time is shortened by more than 50%, improving the power station's grid support capabilities.

[0109] like Figure 2 As shown, an embodiment of the present invention further provides an energy access control system for photovoltaic power generation, comprising: The acquisition module is used to collect real-time operating data of the target photovoltaic power station, real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid; The detection module is used to generate a prediction of the power fluctuation trend of the target photovoltaic power station within a preset time period in the future based on real-time meteorological monitoring data; The fusion module is used to generate initial dynamic power adjustment instructions based on the power fluctuation trend and the acquired real-time grid parameters; a processing module configured to dynamically select three reference points for representing the current system operating state based on the initial dynamic power adjustment instruction, construct a dynamically changing multidimensional operating space based on the three reference points, divide the dynamically changing multidimensional operating space into regions to form multiple operating sub-regions, and generate an operating state adaptive correction factor based on changes in distribution characteristics of the multiple operating sub-regions; A correction module, configured to correct the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction; The execution module is configured to send the obtained second dynamic power regulation instruction to the power regulation device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operation range.

[0110] It should be noted that this system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0111] An embodiment of the present invention further provides a computing device comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the above-described method. All implementations in the above-described method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0112] The embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, causes the computer to execute the above-described method. All implementations in the above-described method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0113] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for controlling energy access for photovoltaic power generation, characterized in that: The method comprises: Step 1: Collect real-time operating data of the target photovoltaic power station, real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid; Step 2: Generate a prediction of the power fluctuation trend of the target photovoltaic power station within a preset time period in the future based on real-time meteorological monitoring data; Step 3: Generate initial dynamic power adjustment instructions based on the power fluctuation trend and the acquired real-time grid parameters; Step 4: Based on the initial dynamic power adjustment instruction, dynamically select three reference points for representing the current system operating state, construct a dynamically changing multidimensional operating space based on the three reference points, divide the dynamically changing multidimensional operating space into regions, and form multiple operating sub-regions. Based on the changes in the distribution characteristics of the multiple operating sub-regions, generate an operating state adaptive correction factor; Step 5: Correcting the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction; Step 6: Send the obtained second dynamic power adjustment instruction to the power adjustment device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operation range.

2. A photovoltaic power generation energy access control method according to claim 1, characterized in that: Step 1: Collect real-time operating data of the target PV power station, real-time meteorological monitoring data of the area where the target PV power station is located, and real-time grid parameters of the public connection point where the target PV power station is connected to the grid, including: Step 11: synchronously collect real-time operating data sets of the photovoltaic power station, including photovoltaic array output current, output voltage, output power and inverter operating status; Step 12: Align the collected meteorological monitoring data set, including light intensity, ambient temperature, and wind speed and direction, based on the timestamp of the real-time operation data; Step 13, synchronously acquiring a real-time grid parameter set of the grid common connection point based on the acquisition time benchmark, including grid frequency, bus voltage, and line impedance parameters; Step 14: pre-process the collected data set by aligning the time axis and normalizing the dimensions to obtain pre-processed data.

3. A photovoltaic power generation energy access control method according to claim 2, characterized in that: Step 2: Generate a forecast of the output power fluctuation trend of the target photovoltaic power station within a preset time period based on the acquired real-time meteorological monitoring data, including: Step 21, extracting the time series of light intensity and ambient temperature of real-time meteorological monitoring data from the preprocessed data; Step 22, based on the extracted time series of light intensity, calculate the light fluctuation coefficient in the first preset time period in the future through the meteorological mutation feature recognition model; Step 23: Combining the extracted ambient temperature time series with the obtained light fluctuation coefficient, and using a power conversion model, generate a power base prediction value for a second preset time period in the future; Step 24 : Perform time series differentiation processing on the generated power base prediction value to extract the fluctuation amplitude and change rate characteristics, and generate the output power fluctuation trend.

4. A photovoltaic power generation energy access control method according to claim 3, characterized in that: Step 3: Generate an initial dynamic power adjustment instruction based on the generated output power fluctuation trend and the acquired real-time grid parameters, including: Step 31: extracting power fluctuation characteristic quantities within a future preset time window based on the power fluctuation trend; Step 32, calculating a power fluctuation compensation coefficient based on the power fluctuation characteristic quantity and the grid frequency deviation data in the real-time grid parameters; Step 33, using the power fluctuation compensation coefficient to correct the power fluctuation characteristic value to generate a fluctuation adjustment reference value; Step 34 : generating an initial dynamic power adjustment instruction based on the fluctuation adjustment reference value and the grid load threshold in the real-time grid parameters.

5. A photovoltaic power generation energy access control method according to claim 4, characterized in that: Step 4: Based on the generated initial dynamic power adjustment instruction, dynamically select three reference points for representing the current system operating state, construct a dynamically changing multi-dimensional operating space based on the three reference points, divide the dynamically changing multi-dimensional operating space into regions, and form multiple operating sub-regions. Based on the changes in the distribution characteristics of the multiple operating sub-regions, generate an operating state adaptive correction factor, including: Step 41 , dynamically selecting three physical quantities, namely, real-time grid frequency, bus voltage, and power regulation amount, as reference dimensions based on the instruction parameter values of the initial dynamic power regulation instruction; Step 42: Based on the three reference dimensions and in combination with the real-time collected data and instruction parameter values, the real-time grid frequency value, the real-time common connection point bus voltage value, and the instantaneous value of the regulation amount actually performed by the current power regulation device are processed, and their corresponding current values are mapped to a coordinate point in space to obtain a first reference point; the grid frequency value, the bus voltage value, and the instantaneous value of the regulation amount actually performed by the power regulation device at the sampling moment immediately before the current moment are processed and stored, and the values of the three physical quantities at the previous moment are mapped to another coordinate point in space to obtain a second reference point; the target power regulation amount explicitly specified in the generated initial dynamic power regulation instruction is processed, and in combination with the frequency safety reference value and voltage safety reference value that the common connection point should maintain when the instruction expects to achieve the target, the target reference values of the three physical quantities are mapped to a third coordinate point in space to obtain a third reference point; Step 43: Using the first reference point, the second reference point, and the third reference point as core positioning points in the space, a dynamic reference space is constructed, and the dynamic reference space is divided into a plurality of operating sub-areas; Step 44 counts the changes in the number of state vector points that appear in each operating sub-region within the most recent consecutive sampling periods, calculates the vector density change rate of each sub-region, and based on the gradient distribution characteristics of these density change rates, comprehensively judges the degree to which the operating state deviates from the command target state or the safe state and the dynamic trend, and calculates and generates an operating state adaptive correction factor.

6. A photovoltaic power generation energy access control method according to claim 5, characterized in that: Step 5, correcting the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction, including: Step 51: Obtain the generated operating state adaptive correction factor, and extract the power regulation target value or change rate requirement contained in the generated initial dynamic power regulation instruction as the initial instruction value; Step 52: Analyze the numerical characteristics of the acquired adaptive correction factor of the operating state to determine its value and change direction; Step 53, based on the value and change direction of the correction factor, a preset weight mapping rule is applied to dynamically determine the weight distribution ratio of the correction factor to correct the initial instruction value; Step 54: performing a weighted correction calculation on the initial command value according to the determined weight distribution ratio. This calculation process weights and combines the initial command value with the correction factor according to the weight ratio to generate a corrected intermediate power regulation command value. Step 55: Based on the power adjustment instruction value and the type of the initial dynamic power adjustment instruction, a second dynamic power adjustment instruction that can be directly executed is generated.

7. A photovoltaic power generation energy access control method according to claim 6, characterized in that: Step 6, sending the obtained second dynamic power adjustment instruction to the power adjustment device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operating range, including: Step 61: Send the generated second dynamic power adjustment instruction to the photovoltaic inverter power control module according to a preset 200ms period; collect in real time the grid response parameter set of the common connection point after the instruction is executed, including the frequency deviation Δf, the voltage deviation ΔU and the power fluctuation rate δP; Step 62: Calculate a parameter deviation vector between the current operating state and the safe operating range based on the collected grid response parameter set; Step 63 : When the parameter deviation vector exceeds the preset threshold, the dynamic instruction re-correction mechanism is triggered to continuously maintain the parameter deviation vector within the safe operating range until the photovoltaic output power is stabilized.

8. An energy access control system for photovoltaic power generation, the system implementing the method according to any one of claims 1 to 7, characterized in that: include: The acquisition module is used to collect real-time operating data of the target photovoltaic power station, real-time meteorological monitoring data of the area where the target photovoltaic power station is located, and real-time grid parameters of the public connection point where the target photovoltaic power station is connected to the grid; The detection module is used to generate a prediction of the power fluctuation trend of the target photovoltaic power station within a preset time period in the future based on real-time meteorological monitoring data; The fusion module is used to generate initial dynamic power adjustment instructions based on the power fluctuation trend and the acquired real-time grid parameters; a processing module configured to dynamically select three reference points for representing the current system operating state based on the initial dynamic power adjustment instruction, construct a dynamically changing multidimensional operating space based on the three reference points, divide the dynamically changing multidimensional operating space into regions to form multiple operating sub-regions, and generate an operating state adaptive correction factor based on changes in distribution characteristics of the multiple operating sub-regions; A correction module, configured to correct the operating state adaptive correction factor and the initial dynamic power adjustment instruction to obtain a second dynamic power adjustment instruction; The execution module is configured to send the obtained second dynamic power regulation instruction to the power regulation device of the target photovoltaic power station for execution, so as to maintain the grid parameters of the target photovoltaic power station at the common connection point within a preset safe operation range.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images