A photovoltaic time series data correction method and device, electronic equipment, storage medium and program product

Abstract

This invention relates to the field of photovoltaic power generation data processing technology, and discloses a method, device, electronic equipment, storage medium, and program product for photovoltaic time series data correction. This invention collects measured time series data from photovoltaic power plants, performs validity pre-screening and binarized boundary extraction to obtain the effective start and end boundaries of the time series and the median time within the interval. Time series misalignments are identified through unilateral boundary tolerance judgment, and time series correction is completed by combining the theoretical astronomical noon time of the power plant. This scheme abandons the traditional full-sequence numerical integration judgment logic, avoids the problem of miscorrection caused by artificial asymmetric power curtailment, and relies on unilateral tolerance judgment to identify normal time series offsets caused by terrain and local weather, preserving the original physical characteristics of the data. Simultaneously, it achieves bidirectional targeted correction using the theoretical astronomical noon time as an independent benchmark, accurately correcting equipment time series anomalies, and solving the shortcomings of traditional technologies that are easily affected by local shading and artificial power curtailment interference, effectively improving the physical fidelity and adaptability to complex operating conditions of photovoltaic time series preprocessing.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation data processing technology, specifically to a photovoltaic time-series data correction method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] As the construction of new power systems continues to advance, the installed capacity of photovoltaic power plants is increasing year by year. Photovoltaic power prediction is a core supporting technology for grid-connected dispatching, power generation verification, and energy storage synergy optimization. Its prediction accuracy directly affects the safe and stable operation of the power grid and the operating revenue of the power plant. Accurate alignment of the measured power generation time series data with the irradiance monitoring time series on the time axis is an indispensable preprocessing step for building high-precision prediction models and carrying out quantitative analysis of measured data. The reliability of data processing in the time series alignment stage directly determines the effectiveness of subsequent data analysis and model training.

[0003] Currently, mainstream time-series alignment methods in the industry generally rely on full-series numerical integration to achieve time-series matching. Typical solutions include the energy centroid method and dynamic time warping algorithms. These methods solve for the energy centroid of the power curve and irradiance curve by integrating the complete time-series data throughout the day. By comparing the time difference between the centroids of the two curves, a rigid overall translation is performed on one set of time series data until the correlation coefficient between the two sets of curves reaches its maximum value. Due to its advantages of simple implementation and fast calculation speed, it has been widely used in photovoltaic power plant data cleaning for a long time. However, this type of processing logic, which uses the global integration centroid as the basis for deviation judgment, has a fundamental shortcoming. The complete time-series integration calculation process will include the natural shift in illumination caused by local mountain shading and the distortion of the power curve caused by asymmetric scheduling power curtailment in the centroid calculation calculation. This causes the centroid position to be distorted and shifted by external forces. Even if the two sets of time series themselves do not have clock synchronization faults, the algorithm will incorrectly judge that there is a time series misalignment and perform forced translation correction, causing irreversible miscorrection problems, significantly losing the authenticity of the original monitoring data, and making it difficult to adapt to complex actual conditions such as mountain photovoltaic and frequent power curtailment. Summary of the Invention

[0004] This invention provides a photovoltaic time-series data correction method, apparatus, electronic device, storage medium, and program product to address the problem that existing technologies relying on full-sequence numerical integration for time alignment are easily distorted by local shading or artificial power rationing.

[0005] In a first aspect, the present invention provides a photovoltaic time-series data correction method, the method comprising: Obtain measured photovoltaic time-series data from photovoltaic power plants; Binarized boundary extraction is performed on the measured photovoltaic time series data that have passed the validity pre-screening to obtain the effective start and end boundaries of the time series and the median of the effective interval time. The effective start and end boundaries of the timing sequence are used to determine the tolerance of one-sided boundary. When the determination result indicates that there is a time series misalignment, the theoretical astronomical noon time of the photovoltaic power station is obtained, and the measured photovoltaic time series data is corrected by combining the median of the effective interval time.

[0006] First, measured photovoltaic time-series data from photovoltaic power plants is acquired. After pre-screening for validity, binarized boundary extraction is performed on qualified data to obtain the corresponding effective start and end boundaries and the median time of the effective interval. Then, the effective start and end boundaries are used to perform unilateral boundary tolerance judgment. When a time-series misalignment is confirmed, the theoretical astronomical noon time of the photovoltaic power plant is calculated, and the obtained median time of the effective interval is combined to complete the time-series correction of the measured photovoltaic time-series data. This invention abandons the traditional judgment logic of full-series numerical integration and instead uses binarized time-series boundary features to extract only the effective rising and falling edge times of the power and irradiance curves as comparison parameters. It no longer performs integration calculations on the complete time series of the whole day, which can directly avoid the problem of distortion of the integral centroid by invalid power generation data during artificial asymmetric power curtailment periods, thus eliminating the miscorrection caused by power curtailment conditions at the root. At the same time, a terrain self-consistency judgment mechanism is built based on unilateral boundary tolerance judgment. As long as the unilateral time difference of the rising or falling edge is less than the preset tolerance threshold, the lag is judged to be caused by... Real physical factors such as terrain and local weather obstruction cause and implement a zero-compensation strategy, without performing time-series translation operations, fully preserving the original terrain physical features inherent in the data. It will not misidentify normal time-series offsets caused by natural obstruction as clock synchronization faults. In scenarios where excessive misalignment occurs on both sides of the boundary and is confirmed as an equipment time-series anomaly, the theoretical astronomical noon time calculated from the latitude and longitude of the site is further introduced as an independent third-party reference benchmark. The deviation of the median time of the two sets of measured sequences from the theoretical noon is compared. For abnormal time series with larger deviations, dynamic time axis compensation is performed to achieve bidirectional alignment correction. This not only accurately identifies and preserves non-fault-related time-series deviations caused by local obstruction and human-induced power rationing, but also reliably corrects real acquisition time-series misalignments. It effectively solves the technical problem that existing technologies that rely on full-series numerical integration for time alignment are easily distorted by local obstruction or human-induced power rationing, significantly improving the physical fidelity of photovoltaic time-series preprocessing results and the adaptability to complex working conditions.

[0007] In one optional implementation, the measured photovoltaic time-series data includes a measured total irradiance time series and a measured power time series, and the effectiveness pre-screening specifically involves: The maximum measured irradiance is extracted from the measured total irradiance time series, and the maximum measured power is extracted from the measured power time series; Obtain the installed capacity of the photovoltaic power station; The installed capacity is multiplied by a preset power generation coefficient threshold to obtain the invalid power data threshold. If the measured maximum power value is greater than or equal to the power data invalid threshold, and the measured maximum irradiance value is greater than or equal to the preset weak light threshold, then the measured photovoltaic time series data is determined to have passed the validity pre-screening. If the maximum measured power is less than the invalid power data threshold, or the maximum measured irradiance is less than the preset weak light threshold, then the measured photovoltaic time series data is determined to have failed the validity pre-screening.

[0008] By combining the installed capacity of the power plant, preset power generation coefficient thresholds, and preset low light thresholds to pre-screen data validity, invalid time-series data with no effective power generation behavior, such as extremely weak lighting and near-shutdown of the unit, can be quickly identified and directly intercepted from subsequent processing. On the one hand, this reduces the amount of computation on invalid data, lowers the overall algorithm's computing power consumption, and improves data processing efficiency. On the other hand, it avoids meaningless data from interfering with subsequent boundary extraction, deviation judgment, and time-series correction processes from the source, ensuring that the entire processing logic only works on valid power generation time series with analysis and alignment value, thereby improving the overall operational stability and result validity of the algorithm.

[0009] In one optional implementation, the effective start and end boundaries of the time series include irradiance start and end boundaries and power start and end boundaries, and the median time of the effective interval includes the midpoint of the irradiance interval and the midpoint of the power interval. The step of performing binarized boundary extraction on the measured photovoltaic time series data that has passed the effectiveness pre-screening to obtain the effective start and end boundaries and the median time of the effective interval includes: Extract the start time and end time of the measured total irradiance time series to obtain the start and end boundaries of the irradiance; Extract the power start time and power end time from the measured power time series to obtain the power start and end boundaries; The midpoint of the irradiance interval is calculated using the start time and end time of the irradiance. The midpoint of the power interval is calculated using the power start time and power end time.

[0010] Abandoning the traditional approach of full-sequence numerical integration, this method extracts the start and end boundaries and calculates the midpoint of the effective power generation intervals for two sets of time series. The core computational scope focuses on the effective time period of actual photovoltaic output, and is not affected by time series curve distortion caused by human-induced asymmetric power curvature or local mountain shading. This fundamentally avoids the problem of the integration center of gravity being distorted. At the same time, using the time series boundaries and the midpoint of the interval as core feature parameters, the feature dimensions are simple and the physical meaning is clear. It can accurately characterize the time distribution characteristics of the effective power generation period, providing stable and reliable basic data support for subsequent single-sided boundary tolerance determination and time series deviation calculation.

[0011] In one optional implementation, the step of using the effective start and end boundaries of the timing sequence to determine the single-sided boundary tolerance includes: Calculate the absolute time difference between the power start time and the irradiance start time to obtain the left boundary difference value; Calculate the absolute time difference between the power end time and the irradiance end time to obtain the right boundary difference value; If the difference at the left boundary is less than or equal to a preset tolerance threshold, or if the difference at the right boundary is less than or equal to the preset tolerance threshold, then it is determined that there is no temporal misalignment between the measured total irradiance time series and the measured power time series. When both the difference between the left boundary and the difference between the right boundary are greater than the preset tolerance threshold, it is determined that there is a temporal misalignment between the measured total irradiance time series and the measured power time series.

[0012] By calculating the time difference between the left and right boundaries separately and combining it with a preset tolerance threshold, the cause of the deviation can be effectively distinguished. For small time offsets on one side of the boundary caused by natural factors such as partial shading by mountains or short-term weather changes, they are directly judged as normal operating conditions without performing time-series translation operations. The inherent physical characteristics of the site's terrain and environmental response carried by the measured data are fully preserved, eliminating the problem of traditional solutions misjudging natural deviations as time-series faults and forcibly correcting them. Only when the offsets on both sides exceed the tolerance range are they judged as true time-series misalignments. The judgment rules are in line with the actual operating scenarios of photovoltaic power plants, with high recognition accuracy. At the same time, the overall judgment logic is simple and has a fast response speed, which can meet the application needs of real-time processing of massive time-series data of the power plant.

[0013] In one optional implementation, when the determination result indicates a timing misalignment, obtaining the theoretical astronomical noon time of the photovoltaic power station and performing timing correction on the measured photovoltaic time series data by combining the median of the effective interval time includes: When the determination result indicates that there is a timing misalignment, the theoretical astronomical noon time of the photovoltaic power station is obtained; The absolute difference between the midpoint of the irradiance interval and the theoretical astronomical noon time is calculated to obtain the irradiance deviation. The absolute difference between the midpoint of the power range and the theoretical astronomical noon time is calculated to obtain the power deviation. Calculate the absolute difference between the midpoint of the irradiance interval and the midpoint of the power interval to obtain the relative translation compensation amount of the sequence. When the power deviation is less than the irradiance deviation, the measured total irradiance time series is time-shifted according to the relative shift compensation amount until it is aligned with the measured power time series. When the irradiance deviation is less than the power deviation, the measured power time series is time-shifted according to the relative shift compensation amount until it is aligned with the measured total irradiance time series.

[0014] By using the theoretical astronomical noon time, independent of the on-site acquisition equipment, as a third-party objective benchmark, and comparing the deviation of the midpoints of the two time series intervals relative to the benchmark, the acquisition sequences with clock anomalies and transmission delays can be accurately located. This breaks the limitation of traditional time series alignment schemes, which can only adjust fixed sequences in one direction, and achieves bidirectional targeted correction. Based on the relative translation compensation amount calculated from the midpoints of the two intervals, the time step of the time series translation can be accurately determined, ensuring the accuracy of translation correction. Ultimately, reliable alignment of the measured total irradiance time series and the measured power time series is achieved, effectively correcting the time series misalignment problem caused by the asynchrony of the acquisition terminals, significantly improving the synchronization accuracy of time series data, and providing a high-quality time series data source for subsequent photovoltaic power prediction, data analysis, and other upper-level applications.

[0015] In one optional implementation, the step of obtaining the theoretical astronomical noon time of the photovoltaic power station when the determination result indicates a timing misalignment includes: When the judgment result indicates that there is a time sequence misalignment, the station is simulated under clear sky conditions using a preset clear sky model to obtain the continuous time sequence curves of solar altitude angle and zenith angle. The timestamp corresponding to the maximum value of the solar altitude angle in the continuous time series curve of the solar altitude angle, or the timestamp corresponding to the minimum value of the zenith angle in the continuous time series curve of the zenith angle, is extracted and used as the theoretical astronomical noon time of the photovoltaic power station.

[0016] The theoretical astronomical noon time is obtained by simulation calculations based on clear sky models and site geographic information. This reference time is generated in accordance with astronomical laws and is completely free from interference from on-site data acquisition hardware, communication links, and human operation. It has strong objectivity and universality, and can effectively ensure the reliability of subsequent deviation calculations, time sequence determination, and correction results. At the same time, the timestamps corresponding to the maximum solar altitude angle or the minimum zenith angle are selected as the theoretical astronomical noon time. The selection rules are highly consistent with astronomical principles, the calculation logic is mature and stable, and the reproducibility is strong. It can accurately generate standard time references, further enhancing the adaptability of the entire correction scheme to different operating conditions in different regions and different types of photovoltaic sites.

[0017] Secondly, the present invention provides a photovoltaic time-series data correction device, the device comprising: The acquisition module is used to acquire measured photovoltaic time-series data from photovoltaic power plants. The pre-screening module is used to perform binarized boundary extraction on the measured photovoltaic time series data that has passed the validity pre-screening, and to obtain the effective start and end boundaries of the time series and the median of the effective interval time. The determination module is used to determine the tolerance of one-sided boundary using the effective start and end boundaries of the timing sequence. The timing correction module is used to obtain the theoretical astronomical noon time of the photovoltaic power station when the determination result is that there is a timing misalignment, and to perform timing correction on the measured photovoltaic time series data in combination with the median of the effective interval time.

[0018] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the photovoltaic timing data correction method of the first aspect or any corresponding embodiment described above.

[0019] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the photovoltaic timing data correction method of the first aspect or any corresponding embodiment thereof.

[0020] Fifthly, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the photovoltaic time-series data correction method of the first aspect or any corresponding embodiment described above. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the first process of the photovoltaic time-series data correction method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the second process of the photovoltaic time-series data correction method according to an embodiment of the present invention; Figure 3 This is a structural block diagram of a photovoltaic time-series data correction device according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0024] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0025] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0026] This invention provides a photovoltaic time-series data correction method, device, electronic device, storage medium, and program product. Most existing related technologies rely on full-series numerical integration to solve for the energy centroid and complete the alignment of power and irradiance time series. During the full intraday time-series integration calculation, natural shifts in sunlight access time caused by localized shading from mountains and distortions in the power curve caused by asymmetric power curvature directly interfere with the energy centroid calculation results, causing abnormal shifts in the centroid position. Even if the two time series themselves do not have clock synchronization anomalies, they will incorrectly determine that the time series are misaligned and forcibly implement overall rigid translation correction. On the one hand, this irreversibly erases the unique topographical physical characteristics of the photovoltaic site; on the other hand, it is highly prone to misalignment under artificial power curvature conditions. Furthermore, this type of time-series alignment mode only adjusts the power-side time series, lacking an independent third-party time reference for bidirectional verification, and cannot identify whether the irradiance acquisition end has a time zone configuration or other issues. Clock synchronization failures have significant limitations in practical engineering applications. This invention acquires measured photovoltaic time-series data from photovoltaic power plants sequentially, performs binarization boundary extraction on data that has passed the validity pre-screening to obtain the effective start and end boundaries of the time series and the median time of the effective interval. Based on the extracted boundary parameters, a one-sided boundary tolerance judgment is performed. Only when a time series misalignment is determined, the theoretical astronomical noon time of the power plant is calculated and combined with the median time of the effective interval to complete the complete processing logic for time series correction. This eliminates the need for full-time integral calculations for deviation judgment, avoiding the problems of local shading and artificial power curtailment distorting the judgment results. At the same time, the one-sided boundary tolerance judgment achieves terrain self-consistency verification, and reasonable time series deviations caused by the natural environment are no longer forcibly corrected, fully preserving the physical attributes of the original data. It can also achieve bidirectional time series verification and dynamic compensation with the independent benchmark of theoretical astronomical noon. Ultimately, it achieves the technical effect of accurately distinguishing non-fault time series fluctuations from equipment time series anomalies and completing high-quality time series alignment preprocessing.

[0027] This embodiment provides a photovoltaic time-series data correction method. Figure 1 This is a flowchart of a photovoltaic time-series data correction method according to an embodiment of the present invention, as follows: Figure 1 As shown, the process includes the following steps: Step S101: Obtain the measured photovoltaic time series data of the photovoltaic power station.

[0028] In this embodiment of the invention, the local SCADA monitoring system and the supporting meteorological monitoring subsystem of the target photovoltaic power station are connected. A fixed, equally spaced sampling frequency preset by the power station is used to collect complete daily continuous monitoring data. On one hand, the measured output power time series corresponding to each timestamp is read from the photovoltaic grid-connected acquisition terminal, with each sampling point bound to a unique time identifier and the measured real-time power generation value. On the other hand, the total horizontal irradiance time series collected by the power station's irradiance sensor at the same time axis is simultaneously retrieved, ensuring that the two time series use completely consistent sampling time markers. At the same time, the inherent static geographical parameters of the power station are simultaneously recorded, including the actual longitude, latitude, and rated installed capacity of the power station.

[0029] Step S102: Binarize the measured photovoltaic time series data that has passed the validity pre-screening and extract the boundaries to obtain the effective start and end boundaries of the time series and the median of the effective interval time.

[0030] In this embodiment of the invention, the received regularized measured photovoltaic time-series data is first pre-screened for validity. Data without correction significance is filtered according to preset judgment rules, and the measured photovoltaic time-series data that have passed the validity pre-screening are retained. Binarization boundary extraction is performed on the filtered valid data, and the corresponding valid start and end boundaries of the time series are extracted for power time series and irradiance time series respectively. Based on the valid start and end boundaries of each time series, midpoint calculation is performed to calculate the median of the valid interval time for each time series. All extracted valid start and end boundaries of the time series and the calculated median of the valid interval time are output completely.

[0031] Step S103: Use the effective start and end boundaries of the time sequence to determine the tolerance of one-sided boundary.

[0032] In this embodiment of the invention, the effective start and end boundaries of the received timing sequence are obtained, and the start and end boundary times corresponding to the power timing sequence and irradiance timing sequence are extracted respectively. The start-side time difference and the end-side time difference are calculated in sequence, and the two single-side time difference values ​​are compared with the preset tolerance threshold to complete the single-side boundary tolerance determination and generate the corresponding determination result.

[0033] Step S104: When the determination result is that there is a time series misalignment, obtain the theoretical astronomical noon time of the photovoltaic power station, and perform time series correction on the measured photovoltaic time series data by combining the median of the effective interval time.

[0034] In this embodiment of the invention, if the received judgment result indicates that there is a time series misalignment, the corresponding theoretical astronomical noon time is obtained based on the latitude and longitude information of the photovoltaic power station. The median of the effective interval time of each of the two time series is compared with the theoretical astronomical noon time. The abnormal time series channel is located based on the comparison result. The corresponding time axis compensation operation is performed on the measured photovoltaic time series data based on the deviation difference to complete the time series correction processing and output the corrected data.

[0035] The measured photovoltaic time-series data refers to the original monitoring time-series dataset synchronously collected by on-site monitoring equipment at photovoltaic power plants according to a unified sampling timestamp. It includes two core monitoring time series: irradiance and power generation. Validity pre-screening involves performing a preliminary verification on the original measured photovoltaic time-series data, removing invalid daily data without correction value, and retaining only valid time-series data necessary for time-series alignment. Binarization boundary extraction involves performing binarization screening on the complete time series based on a preset judgment threshold, removing low-value invalid sampling points, and extracting the boundaries of the start and end times of valid power generation periods. The valid start and end boundaries of the time series are the start and end sampling times corresponding to the effective power generation working interval after binarization screening. The median time of the effective interval is the center time representation value of the effective power generation period calculated based on the valid start and end boundaries of the time series, used to represent the entire effective time interval. The time sequence is centered; the single-sided boundary tolerance judgment refers to the verification logic that compares the time difference values ​​of the start and end sides of the power time sequence and the irradiance time sequence respectively, and judges the cause of the time sequence deviation by combining the preset threshold; the time sequence misalignment is an abnormal state in which the time offset of the power time sequence and the irradiance time sequence exceeds the allowable threshold due to equipment failures such as the clock of the acquisition terminal being out of sync or transmission delay, and the time axis of the two time sequences cannot be naturally aligned; the theoretical astronomical noon time is the standard timestamp corresponding to the maximum solar altitude angle under ideal working conditions without clouds, calculated by combining the site's geographical location and the date of the day through the astronomical radiation model, and can be used as an independent third-party time reference; the time sequence correction is to use the third-party standard time reference as a reference, calculate the time sequence offset, and perform an overall time axis translation on the abnormal time sequence to achieve a correction operation for precise time alignment of the two time sequences of power and irradiance.

[0036] The photovoltaic time-series data correction method provided in this embodiment includes the following steps: Step S101: Standardized acquisition of synchronously aligned measured power and irradiance time series data, as well as site geographical parameters, providing a standardized data source with a unified format and time axis matching for the entire process calculation; Step S102: First, invalid data that does not require correction is removed through validity pre-screening, and then the boundary and median of the effective power generation interval are locked using a binarized boundary extraction method. This method breaks out of the calculation framework of full-series numerical integration and will not change the judgment benchmark due to power curve distortion caused by asymmetric power curtailment, thus avoiding erroneous judgments under power curtailment conditions; Step S103: Directly use... The extracted effective start and end boundaries of the time series are used to determine the tolerance of one-sided boundaries. This can quickly identify small one-sided time deviations originating from real physical scenarios such as mountain shading and local meteorological disturbances, without the need for forced time series translation, thus fully preserving the inherent characteristics of the site terrain carried by the measured data. If a real time series misalignment is determined, step S104 introduces the theoretical astronomical noon time of the photovoltaic site as an independent reference standard, and compares the deviation with the median of the effective interval time of the two sets of time series. This accurately locates the abnormal time series and performs corresponding time compensation operations to correct the deviation, overcoming the limitation of unidirectional correction by simply adjusting the power time series. The coordinated efforts of these steps effectively solve the technical pain points of the relevant technologies being easily affected by local shading and human-induced power rationing interference in the judgment results. This approach not only fully preserves the physical attributes of the original measured data but also accurately corrects time series misalignments caused by equipment clock anomalies, effectively improving the robustness and physical fidelity of the time series alignment results.

[0037] This embodiment provides a photovoltaic time-series data correction method. Figure 2 This is a flowchart of a photovoltaic time-series data correction method according to an embodiment of the present invention, as follows: Figure 2 As shown, the process includes the following steps: Step S201: Obtain the measured photovoltaic time-series data of the photovoltaic power station. For details, please refer to [link to relevant documentation]. Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0038] Step S202: Binarize the measured photovoltaic time series data that has passed the validity pre-screening and extract the boundaries to obtain the effective start and end boundaries of the time series and the median of the effective interval time.

[0039] Specifically, the measured photovoltaic time-series data includes the measured total irradiance time series and the measured power time series, and the effectiveness pre-screening is as follows: The maximum measured irradiance is extracted from the measured total irradiance time series, and the maximum measured power is extracted from the measured power time series; Obtain the installed capacity of photovoltaic power plants; The invalid power data threshold is obtained by multiplying the installed capacity with a preset power generation coefficient threshold. If the maximum measured power value is greater than or equal to the power data invalid threshold, and the maximum measured irradiance value is greater than or equal to the preset weak light threshold, then the measured photovoltaic time series data is determined to have passed the validity pre-screening. If the maximum measured power is less than the invalid power data threshold, or the maximum measured irradiance is less than the preset weak light threshold, then the measured photovoltaic time series data is determined to have failed the validity pre-screening.

[0040] The measured total irradiance time series is a continuous time series formed by irradiance sensors continuously collecting total irradiance values ​​on the horizontal plane according to a fixed sampling period, with each sampling point bound to a unique timestamp. The measured power time series is a continuous time series formed by photovoltaic grid-connected metering terminals continuously collecting actual output active power values ​​of the power station using a sampling period synchronized with irradiance and binding corresponding timestamps. The maximum measured irradiance value is the maximum irradiance monitoring value among all sampling points in the daily measured total irradiance time series; the maximum measured power value is the power generated among all sampling points in the daily measured power time series. The maximum value of the monitored values; installed capacity refers to the rated grid-connected total installed capacity of the photovoltaic power station, which is also the theoretical maximum power generation of the station; the preset power generation coefficient threshold is a pre-set minimum power generation ratio coefficient used to determine whether the power generation of the day is close to shutdown; the power data invalid threshold is the critical power value obtained by multiplying the installed capacity of the station by the preset power generation coefficient threshold. If it is lower than this value, it is determined that the day is close to shutdown and there is no value in processing; the preset weak light threshold is a pre-set critical irradiance value. If the maximum irradiance of the day is lower than this threshold, it is determined that the light is extremely weak and there is no effective power generation period.

[0041] In this embodiment of the invention, the maximum measured irradiance is extracted from the measured total irradiance time series, and the maximum measured power is extracted from the measured power time series; the installed capacity of the photovoltaic power station is obtained; the installed capacity is multiplied by a preset minimum power generation coefficient threshold to obtain an invalid power data threshold, wherein the minimum power generation coefficient threshold can be exemplarily set to 0.05; when the maximum measured power is less than the invalid power data threshold, or the maximum measured irradiance is less than the preset weak light threshold, the day is determined to be an extreme shutdown or extremely dark invalid day, and the measured photovoltaic time series data is determined to have failed the validity pre-screening. The system does not perform time series shifting operation, directly outputs the original sequence, and does not enter the subsequent binarization boundary extraction stage; when the maximum measured power is greater than or equal to the invalid power data threshold, and the maximum measured irradiance is greater than or equal to the preset weak light threshold, the measured photovoltaic time series data is determined to have passed the validity pre-screening, indicating that the photovoltaic power generation is normal and the irradiance conditions are normal on that day, and the data has correction value.

[0042] Specifically, the effective start and end boundaries of the time series include the start and end boundaries of irradiance and the start and end boundaries of power, and the median of the effective interval time includes the midpoint of the irradiance interval and the midpoint of the power interval. The above step S202 includes: Step S2021: Extract the start time and end time of the measured total irradiance time series to obtain the start and end boundaries of irradiance.

[0043] In this embodiment of the invention, the discrete-time sampling point or timestamp is set as follows: t The measured total irradiance time series is denoted as I(t) First, extract the maximum daily irradiance value of the sequence. max(I) Configure preset percentage thresholds α Exemplary examples are acceptable. α =0.05; Perform binarization activation determination on each sample point of the sequence, and filter all samples that meet the condition. The timestamp with the smallest value in the set is taken as the irradiance start time. Start rad The timestamp with the largest value is taken as the end time of irradiance. End rad ,combination Start rad and End rad The complete irradiance start and end boundaries are obtained. Start rad , End rad ].

[0044] Step S2022: Extract the power start time and power end time of the measured power time series to obtain the power start and end boundaries.

[0045] In this embodiment of the invention, the measured power time series is denoted as P(t), and the daily maximum power value of the series is first extracted. max(P) The preset percentage threshold is used and is consistent with the irradiance boundary. α ; Traverse all discrete-time sampling points t Filter all that meet the criteria The minimum value among the timestamps is taken as the power start time. Start pow The maximum value among them is taken as the power end time. End pow ,combination Start pow and End pow The complete power start and end boundaries are obtained. Start pow , End pow ].

[0046] Step S2023: Calculate the midpoint of the irradiance interval using the start time and end time of the irradiance.

[0047] In this embodiment of the invention, the midpoint calculation formula is used. Mid rad = Start rad +( End rad Start rad Substituting the previously calculated irradiance start time and irradiance end time into the formula, the midpoint of the irradiance interval is calculated. Mid rad .

[0048] Step S2024: Calculate the midpoint of the power interval using the power start time and power end time.

[0049] In this embodiment of the invention, the midpoint calculation formula is used. Mid pow = Start pow +( End pow Start pow Substituting the obtained power start time and power end time into the formula, the midpoint of the power interval is calculated. Mid pow .

[0050] The irradiance start and end boundaries are the complete boundary intervals formed by the effective start time and effective end time after the measured total irradiance time series is extracted through binarization; the power start and end boundaries are the complete boundary intervals formed by the effective start time and effective end time after the measured power time series is extracted through binarization; the irradiance start time is the first sampling timestamp in the irradiance time series that meets the effective judgment threshold; the irradiance end time is the last sampling timestamp in the irradiance time series that meets the effective judgment threshold; the power start time is the first sampling timestamp in the power time series that meets the effective judgment threshold; the power end time is the last sampling timestamp in the power time series that meets the effective judgment threshold; the irradiance interval midpoint is the center time point of the effective power generation period obtained based on the irradiance start and end boundaries; the power interval midpoint is the center time point of the effective power generation period obtained based on the power start and end boundaries.

[0051] In one optional embodiment, the original embodiment uses a preset percentage threshold α multiplied by the daily maximum values ​​of irradiance and power sequences to obtain a dynamic threshold for filtering valid sampling times. This alternative solution abandons the dynamic percentage threshold and replaces it with a fixed absolute value threshold preset based on the installed capacity and physical limits of the power station for boundary filtering. An example configurable fixed threshold for power determination is 2MW, and the fixed threshold for irradiance determination is 30W / m².2 Traversing the time series of measured total irradiance, we selected those with measured irradiance values ​​greater than 30 W / m². 2 All discrete timestamps are used to extract the minimum timestamp as the irradiance start time and the maximum timestamp as the irradiance end time, thus obtaining the irradiance start and end boundaries. Simultaneously, the measured power time series is traversed, and all timestamps with measured power values ​​greater than 2MW are selected to extract the power start and end boundaries. The subsequent operation logic after boundary extraction is completely consistent with the original embodiment.

[0052] In another alternative embodiment, the original embodiment uses the effective start and end boundaries to solve for the midpoint of the irradiance interval and the midpoint of the power interval using the arithmetic midpoint formula; this alternative solution no longer uses the arithmetic midpoint calculation method. Within the effective start and end time intervals corresponding to irradiance and power, the duration corresponding to the non-zero effective sampling points within the interval is statistically analyzed, and the median time value corresponding to this duration is solved. This median time point represents the median time of the effective interval of the corresponding time series; the median time point is used to complete the subsequent deviation calculation, single-sided boundary tolerance verification, and bidirectional time series compensation calculation.

[0053] Step S203: Use the effective start and end boundaries of the time sequence to determine the tolerance of one-sided boundary.

[0054] Specifically, step S203 includes: Step S2031: Calculate the absolute time difference between the power start time and the irradiance start time to obtain the left boundary difference value.

[0055] In this embodiment of the invention, according to the calculation formula Diff left =∣ Start pow Start rad | Calculate the difference at the left boundary, which is the absolute value of the time difference between the power start time and the irradiance start time.

[0056] Step S2032: Calculate the absolute time difference between the power end time and the irradiance end time to obtain the right boundary difference value.

[0057] In this embodiment of the invention, according to the calculation formula Diff right =∣ End pow End rad | Calculate the difference at the right boundary, which is the absolute value of the time difference between the power end time and the irradiance end time.

[0058] Step S2033: When the difference at the left boundary is less than or equal to the preset tolerance threshold, or the difference at the right boundary is less than or equal to the preset tolerance threshold, it is determined that there is no temporal misalignment between the measured total irradiance time series and the measured power time series.

[0059] In this embodiment of the invention, the preset tolerance threshold is denoted as... T tol For example, it can be set to 1.5 hours; if the conditions are met... Diff left ≤ T tol or Diff right ≤ T tol The time deviation between the two time series was determined to originate from reasonable physical deviations caused by local terrain shading and sudden weather changes at the photovoltaic power plant. A zero-compensation strategy was then triggered, setting the relative time shift correction between the power series and the irradiance series to zero. Shift steps =0, no longer perform time axis translation operation, directly output the original time series, and completely retain the inherent environmental response characteristics of the photovoltaic power station.

[0060] Step S2034: When both the difference between the left and right boundaries are greater than the preset tolerance threshold, it is determined that there is a time misalignment between the measured total irradiance time series and the measured power time series.

[0061] In this embodiment of the invention, if simultaneously satisfying Diff left > T tol as well as Diff right > T tol Excluding normal physical offset factors such as terrain and weather, it is determined that the two sets of time series are indeed misaligned, and the determination result is transmitted to the next correction step.

[0062] The left boundary difference is the absolute value of the time difference between the power start time and the irradiance start time, used to characterize the unilateral time offset of the two rising edges of the timing sequence; the right boundary difference is the absolute value of the time difference between the power end time and the irradiance end time, used to characterize the unilateral time offset of the two falling edges of the timing sequence; the preset tolerance threshold is the maximum duration of natural offset allowed by the unilateral boundary, which can be used to distinguish between normal offset caused by terrain shading and timing faults caused by equipment.

[0063] Step S204: When the determination result is that there is a time series misalignment, obtain the theoretical astronomical noon time of the photovoltaic power station, and perform time series correction on the measured photovoltaic time series data by combining the median of the effective interval time.

[0064] Specifically, step S204 includes: Step S2041: When the determination result is that there is a timing misalignment, obtain the theoretical astronomical noon time of the photovoltaic power station.

[0065] In an optional implementation, step S2041 includes: Step a1: When the determination result is that there is a time misalignment, the station is simulated under clear sky using a preset clear sky model to obtain the continuous time series curves of solar altitude angle and zenith angle.

[0066] In this embodiment of the invention, under the premise that there is a time sequence misalignment between the two sets of time series, the longitude (Lon), latitude (Lat), and date information of the photovoltaic power station obtained in the previous period are retrieved; a preset clear sky model is called, which is a general solar radiation theory model, and can be any one of the Bird model, Ineichen model, Haurwitz model, and REST2 model. Based on the geographical coordinates of the power station and the date information, a clear sky simulation calculation under the ideal cloudless state is carried out to solve and generate the theoretical zenith angle time series curve and solar altitude angle time series curve for the entire day at that geographical location.

[0067] Step a2: Extract the timestamp corresponding to the maximum value of the solar altitude angle in the continuous time series curve of solar altitude angle, or the timestamp corresponding to the minimum value of the zenith angle in the continuous time series curve of zenith angle, as the theoretical astronomical noon time of the photovoltaic power station.

[0068] In this embodiment of the invention, the obtained continuous time-series curve of solar altitude angle is traversed to locate the timestamp corresponding to the maximum value of the solar altitude angle; or the continuous time-series curve of zenith angle is traversed to locate the timestamp corresponding to the minimum value of the zenith angle, and this timestamp is assigned as the theoretical astronomical noon time of the photovoltaic power station on that day. T noon The reference time is output for subsequent timing correction calculations.

[0069] The preset clear-sky model is a general solar radiation theory clear-sky simulation model, which can be selected from the Bird model, Ineichen model, Haurwitz model, and REST2 model, for the simulation calculation of solar angle time series under ideal cloudless conditions; the clear-sky simulation of the photovoltaic power station is the simulation calculation process of inputting the longitude and latitude of the photovoltaic power station and the date information, and calling the clear-sky model to solve the solar angle time series curve for the whole day; the continuous time series curve of solar altitude angle is a continuous time series curve composed of the solar altitude angle values ​​at each moment of the whole day generated after the clear-sky simulation of the power station; the continuous time series curve of zenith angle is a continuous time series curve composed of the solar zenith angle values ​​at each moment of the whole day generated after the clear-sky simulation of the power station; the timestamp is a unique time identifier bound to each discrete time series sampling moment, which can realize the precise time positioning of each sampling point.

[0070] Step S2042: Calculate the absolute difference between the midpoint of the irradiance interval and the theoretical astronomical noon time to obtain the irradiance deviation.

[0071] In this embodiment of the invention, according to the calculation formula Dist rad =∣ Mid rad T noon | Perform calculations to determine the midpoint of the irradiance interval and the theoretical astronomical noon time. T noon The absolute value of the time is the irradiance deviation.

[0072] Step S2043: Calculate the absolute difference between the midpoint of the power range and the theoretical astronomical noon time to obtain the power deviation.

[0073] In this embodiment of the invention, according to the calculation formula Dist pow =∣ Mid pow T noon | Perform calculations to determine the midpoint of the power interval and the theoretical astronomical noon time. T noon The absolute value of the time is the power deviation.

[0074] Step S2044: Calculate the absolute difference between the midpoint of the irradiance interval and the midpoint of the power interval to obtain the relative translation compensation amount of the sequence.

[0075] In this embodiment of the invention, according to the calculation formula Shift steps =∣ Mid pow Mid rad | Perform the calculation to find the absolute value of the time difference between the midpoints of the two intervals, and use this value as the relative translation compensation amount of the sequence.

[0076] Step S2045: When the power deviation is less than the irradiance deviation, the time series of the measured total irradiance is time-shifted according to the relative shift compensation amount until it is aligned with the time series of the measured power.

[0077] In this embodiment of the invention, if it is determined Dist pow < Dist rad This will determine if there is an anomaly in the time configuration of the weather station system, and will allow for the determination of the measured total irradiance time series. I(t)And its associated meteorological data, according to Shift steps The compensation amount is used to perform an overall translation operation to align the irradiance sequence with the power sequence on the time axis.

[0078] Step S2046: When the irradiance deviation is less than the power deviation, the measured power time series is time-shifted according to the relative shift compensation amount until it is aligned with the measured total irradiance time series.

[0079] In this embodiment of the invention, if it is determined Dist rad < Dist pow This confirms that there is an abnormal data delay in the power acquisition gateway, and the measured power time series can be used to determine this. P(t) according to Shift steps The compensation amount is used to perform an overall translation operation to align the power sequence with the irradiance sequence on the time axis.

[0080] Irradiance deviation is the absolute time difference between the midpoint of the irradiance interval and the theoretical astronomical noon, which can characterize the degree of deviation of the irradiance time series relative to a third-party reference; power deviation is the absolute time difference between the midpoint of the power interval and the theoretical astronomical noon, which can characterize the degree of deviation of the power time series relative to a third-party reference; the sequence relative translation compensation is the absolute time difference between the midpoint of the power interval and the midpoint of the irradiance interval, and it is also the time step required for time series translation correction.

[0081] This invention no longer relies on full-sequence numerical integration to determine timing deviations. Instead, it uses binarized boundary extraction to perform subsequent determinations based solely on the start and end boundaries of the effective power generation intervals for power and irradiance. This completely avoids the technical defects of distorting the integral centroid and causing erroneous corrections during invalid power generation periods under asymmetric scheduling and power curtailment conditions. Simultaneously, it utilizes a one-sided boundary tolerance determination mechanism to achieve one-sided boundary alignment without inspection. As long as the time difference of one-sided boundaries is less than a preset tolerance threshold, it determines that the time lag originates from real physical factors such as mountain terrain obstruction or local extreme weather, triggering a zero-compensation strategy and ceasing timing shift operations. This fully preserves the reasonable lag characteristics caused by the site terrain and environmental disturbances carried by the measured data. To ensure the physical fidelity of measured data, this invention uses a theoretical astronomical noon time, which is completely physically isolated from the on-site acquisition equipment, as a third-party absolute benchmark. By comparing the deviation of the median time of the effective time interval of two sets of time series from this benchmark, it can automatically and accurately distinguish between abnormal meteorological irradiance acquisition equipment and abnormal power SCADA acquisition gateway. For abnormal time series, it performs targeted translation compensation to complete bidirectional targeted time series repair, breaking the application blind spot of traditional algorithms that can only blindly align in one direction. It realizes a system-level upgrade of the correction logic from one-way alignment to bidirectional arbitration repair, which greatly improves the algorithm robustness and adaptability to complex working conditions of the entire process of photovoltaic time series preprocessing and data cleaning.

[0082] This embodiment also provides a photovoltaic time-series data correction device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0083] This embodiment provides a photovoltaic time-series data correction device, such as... Figure 3 As shown, it includes: The acquisition module 301 is used to acquire measured photovoltaic time-series data of photovoltaic power plants; The pre-screening module 302 is used to perform binarized boundary extraction on the measured photovoltaic time series data that has passed the validity pre-screening, and to obtain the effective start and end boundaries of the time series and the median of the effective interval time. The determination module 303 is used to determine the tolerance of one-sided boundary using the effective start and end boundaries of the timing sequence; The correction module 304 is used to obtain the theoretical astronomical noon time of the photovoltaic power station when the judgment result is that there is a time misalignment, and to perform time misalignment correction on the measured photovoltaic time series data by combining the median of the effective interval time.

[0084] In one optional implementation, the measured photovoltaic time-series data includes a measured total irradiance time series and a measured power time series, and the effectiveness pre-screening specifically involves: Extraction sub-units are used to extract the maximum measured irradiance from the measured total irradiance time series and the maximum measured power from the measured power time series. Acquire sub-units to obtain the installed capacity of photovoltaic power stations; The calculation subunit is used to perform a multiplication operation between the installed capacity and a preset power generation coefficient threshold to obtain an invalid power data threshold. The first determination subunit is used to determine that the measured photovoltaic time series data passes the validity pre-screening when the maximum measured power value is greater than or equal to the power data invalid threshold and the maximum measured irradiance value is greater than or equal to the preset weak light threshold. The second determination subunit is used to determine that the measured photovoltaic time series data has not passed the validity pre-screening when the maximum measured power is less than the power data invalid threshold or the maximum measured irradiance is less than the preset weak light threshold.

[0085] In one optional implementation, the effective start and end boundaries of the time series include irradiance start and end boundaries and power start and end boundaries, the median time of the effective interval includes the midpoint of the irradiance interval and the midpoint of the power interval, and the pre-screening module 302 includes: The first extraction unit is used to extract the start time and end time of the measured total irradiance time series to obtain the start and end boundaries of the irradiance. The second extraction unit is used to extract the power start time and power end time of the measured power time series to obtain the power start and end boundaries; The first calculation unit is used to calculate the midpoint of the irradiance interval using the start time and end time of irradiance. The second calculation unit is used to calculate the midpoint of the power interval using the power start time and power end time.

[0086] In one optional implementation, the determination module 303 includes: The third calculation unit is used to calculate the absolute time difference between the power start time and the irradiance start time, and obtain the left boundary difference value. The fourth calculation unit is used to calculate the absolute time difference between the power end time and the irradiance end time, and obtain the right boundary difference value; The third determination unit is used to determine that there is no time misalignment between the measured total irradiance time series and the measured power time series when the difference between the left boundary and the right boundary is less than or equal to the preset tolerance threshold. The fourth determination unit is used to determine that there is a time misalignment between the measured total irradiance time series and the measured power time series when both the difference between the left boundary and the difference between the right boundary are greater than the preset tolerance threshold.

[0087] In one alternative implementation, the correction module 304 includes: The acquisition unit is used to obtain the theoretical astronomical noon time of the photovoltaic power station when the judgment result is that there is a timing misalignment; The irradiance deviation unit is used to calculate the absolute difference between the midpoint of the irradiance interval and the theoretical astronomical noon time, thus obtaining the irradiance deviation. The power deviation calculation unit is used to calculate the absolute difference between the midpoint of the power interval and the theoretical astronomical noon time, and obtain the power deviation. The compensation unit is used to calculate the absolute difference between the midpoint of the irradiance interval and the midpoint of the power interval, and to obtain the relative translation compensation of the sequence. The alignment unit is used to time-shift the measured total irradiance time series according to the relative translation compensation amount of the sequence when the power deviation is less than the irradiance deviation, until it is aligned with the measured power time series. The translation unit is used to time-shift the measured power time series according to the relative translation compensation amount of the sequence when the irradiance deviation is less than the power deviation, until it is aligned with the measured total irradiance time series.

[0088] In one optional implementation, the acquisition unit includes: The third judgment subunit is used to simulate the clear sky of the station using a preset clear sky model when the judgment result is that there is a time misalignment, and to obtain the continuous time series curve of the solar altitude angle and the continuous time series curve of the zenith angle. The timestamp extraction sub-unit is used to extract the timestamp corresponding to the maximum value of the solar altitude angle in the continuous time series curve of solar altitude angle, or the timestamp corresponding to the minimum value of the zenith angle in the continuous time series curve of zenith angle, as the theoretical astronomical noon time of the photovoltaic power station.

[0089] The photovoltaic time-series data correction device provided in this embodiment of the invention can execute the photovoltaic time-series data correction method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0090] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0091] The following is a detailed reference. Figure 4This diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 401, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 402 or a program loaded from memory 408 into random access memory (RAM) 403. The RAM 403 also stores various programs and data required for the operation of the electronic device. The processor 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0092] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0093] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 409, or installed from a memory 408, or installed from a ROM 402. When the computer program is executed by the processor 401, it performs the functions defined in the photovoltaic time-series data correction method of the embodiments of the present invention.

[0094] Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0095] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the photovoltaic timing data correction method shown in the above embodiments is implemented.

[0096] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0097] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A photovoltaic timing data rectification method, characterized in that, The method includes: Obtain measured photovoltaic time-series data from photovoltaic power plants; Binarized boundary extraction is performed on the measured photovoltaic time series data that have passed the validity pre-screening to obtain the effective start and end boundaries of the time series and the median of the effective interval time. The effective start and end boundaries of the timing sequence are used to determine the tolerance of one-sided boundary. When the determination result indicates that there is a time series misalignment, the theoretical astronomical noon time of the photovoltaic power station is obtained, and the measured photovoltaic time series data is corrected by combining the median of the effective interval time.

2. The method of claim 1, wherein, The measured photovoltaic time-series data includes the measured total irradiance time series and the measured power time series. The effectiveness pre-screening specifically involves: The maximum measured irradiance is extracted from the measured total irradiance time series, and the maximum measured power is extracted from the measured power time series; Obtain the installed capacity of the photovoltaic power station; The installed capacity is multiplied by a preset power generation coefficient threshold to obtain the invalid power data threshold. If the measured maximum power value is greater than or equal to the power data invalid threshold, and the measured maximum irradiance value is greater than or equal to the preset weak light threshold, then the measured photovoltaic time series data is determined to have passed the validity pre-screening. If the maximum measured power is less than the invalid power data threshold, or the maximum measured irradiance is less than the preset weak light threshold, then the measured photovoltaic time series data is determined to have failed the validity pre-screening.

3. The method of claim 2, wherein, The effective start and end boundaries of the time series include irradiance start and end boundaries and power start and end boundaries. The median time of the effective interval includes the midpoint of the irradiance interval and the midpoint of the power interval. The step of performing binarized boundary extraction on the measured photovoltaic time series data that has passed the validity pre-screening to obtain the effective start and end boundaries of the time series and the median time of the effective interval includes: Extract the start time and end time of the measured total irradiance time series to obtain the start and end boundaries of the irradiance; Extract the power start time and power end time from the measured power time series to obtain the power start and end boundaries; The midpoint of the irradiance interval is calculated using the start time and end time of the irradiance. The midpoint of the power interval is calculated using the power start time and power end time.

4. The method of claim 3, wherein, The step of using the effective start and end boundaries of the time series to determine the tolerance of one-sided boundaries includes: Calculate the absolute time difference between the power start time and the irradiance start time to obtain the left boundary difference value; Calculate the absolute time difference between the power end time and the irradiance end time to obtain the right boundary difference value; If the difference at the left boundary is less than or equal to a preset tolerance threshold, or if the difference at the right boundary is less than or equal to the preset tolerance threshold, then it is determined that there is no temporal misalignment between the measured total irradiance time series and the measured power time series. When both the difference between the left boundary and the difference between the right boundary are greater than the preset tolerance threshold, it is determined that there is a temporal misalignment between the measured total irradiance time series and the measured power time series.

5. The method of claim 3, wherein, When the determination result indicates a timing misalignment, the theoretical astronomical noon time of the photovoltaic power station is obtained, and the measured photovoltaic time series data is corrected by combining the median of the effective interval time, including: When the determination result indicates that there is a timing misalignment, the theoretical astronomical noon time of the photovoltaic power station is obtained; The absolute difference between the midpoint of the irradiance interval and the theoretical astronomical noon time is calculated to obtain the irradiance deviation. The absolute difference between the midpoint of the power range and the theoretical astronomical noon time is calculated to obtain the power deviation. Calculate the absolute difference between the midpoint of the irradiance interval and the midpoint of the power interval to obtain the relative translation compensation amount of the sequence. When the power deviation is less than the irradiance deviation, the measured total irradiance time series is time-shifted according to the relative shift compensation amount until it is aligned with the measured power time series. When the irradiance deviation is less than the power deviation, the measured power time series is time-shifted according to the relative shift compensation amount until it is aligned with the measured total irradiance time series.

6. The method according to claim 1 or 5, characterized in that, When the determination result indicates a timing misalignment, obtaining the theoretical astronomical noon time of the photovoltaic power station includes: When the judgment result indicates that there is a time sequence misalignment, the station is simulated under clear sky conditions using a preset clear sky model to obtain the continuous time sequence curves of solar altitude angle and zenith angle. The timestamp corresponding to the maximum value of the solar altitude angle in the continuous time series curve of the solar altitude angle, or the timestamp corresponding to the minimum value of the zenith angle in the continuous time series curve of the zenith angle, is extracted and used as the theoretical astronomical noon time of the photovoltaic power station.

7. A photovoltaic time-series data correction device, characterized in that, The device includes: The acquisition module is used to acquire measured photovoltaic time-series data from photovoltaic power plants. The pre-screening module is used to perform binarized boundary extraction on the measured photovoltaic time series data that has passed the validity pre-screening, and to obtain the effective start and end boundaries of the time series and the median of the effective interval time. The determination module is used to determine the tolerance of one-sided boundary using the effective start and end boundaries of the timing sequence. The timing correction module is used to obtain the theoretical astronomical noon time of the photovoltaic power station when the determination result is that there is a timing misalignment, and to perform timing correction on the measured photovoltaic time series data in combination with the median of the effective interval time.

8. An electronic device, characterized in that, include: A memory and a processor are interconnected, the memory stores computer instructions, and the processor executes the photovoltaic time-series data correction method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the photovoltaic time-series data correction method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes computer instructions for causing a computer to execute the photovoltaic time-series data correction method according to any one of claims 1 to 6.

Images