Photovoltaic tracking algorithm gain evaluation method and related device

Abstract

The invention discloses a photovoltaic tracking algorithm gain evaluation method and a related device, and relates to the technical field of photovoltaic power generation, and the method comprises the steps: determining a reference difference coefficient between an experimental group and a control group according to the operation data of the experimental group and the control group in each analysis time period in a preset historical period, equipment performance drift caused by equipment attenuation and the like is effectively captured, so that power generation data is corrected according to the reference difference coefficient, the power generation gain of the experimental group relative to the control group is calculated, and the real-time accuracy of the reference of the experimental group and the control group is ensured. Meanwhile, interference factors including at least one of cloud layer non-uniformity interference, shadow shielding interference and power limiting interference are identified, and the power generation data influenced by the interference factors are corrected, so that the influence of the interference factors on the power generation gain evaluation result is reduced, and the accuracy and reliability of the power generation gain evaluation result are improved.

Description

Technical Field

[0001] This application relates to the field of photovoltaic power generation technology, and in particular to a photovoltaic tracking algorithm gain evaluation method and related apparatus. Background Technology

[0002] In photovoltaic (PV) power plants, the performance of PV tracking algorithms directly affects the power generation, making accurate evaluation of their gains crucial. Empirical testing of PV tracking algorithms typically employs a comparative matrix method, which involves setting up an experimental matrix running the PV tracking algorithm under evaluation and a control matrix running a conventional PV tracking algorithm. The gains of the intelligent algorithm are assessed by comparing the power generation of the two groups.

[0003] However, due to the presence of non-ideal factors in real power plant environments, such as uneven cloud distribution in different areas, shading interference, power plant curtailment, and equipment performance drift, the gain assessment results are affected by a variety of factors, resulting in inaccurate and unreliable gain assessment results. Summary of the Invention

[0004] In view of the above problems, this application provides a photovoltaic tracking algorithm gain evaluation method and related apparatus to reduce the impact of interference factors on the power generation gain evaluation results and improve the accuracy and reliability of the power generation gain evaluation results. The specific solution is as follows:

[0005] The first aspect of this application provides a method for evaluating the gain of a photovoltaic tracking algorithm, including:

[0006] The experimental group and the control group obtained operational data for each analysis period within a preset historical period. The photovoltaic array of the experimental group ran a first tracking algorithm, and the photovoltaic array of the control group ran a second tracking algorithm. The operational data included tracking angle and power generation data.

[0007] Based on the operational data corresponding to the experimental group and the control group, the baseline difference coefficient between the experimental group and the control group is determined;

[0008] Interference factors are identified in the operational data corresponding to the experimental group and the control group. The interference factors include at least one of the following: cloud unevenness interference, shadow occlusion interference, and power rationing interference.

[0009] The power generation data affected by the interference factors are corrected according to the benchmark difference coefficient to obtain the corrected power generation data for the experimental group and the control group.

[0010] Based on the baseline difference coefficient and the corrected power generation data of the experimental group and the control group, the power generation gain of the experimental group relative to the control group is calculated.

[0011] In one possible implementation, determining the baseline difference coefficient between the experimental group and the control group based on the operational data corresponding to the experimental group and the control group includes:

[0012] Extract the target analysis period where the tracking angle difference between the experimental group and the control group is within a preset angle range and the instantaneous power generation difference is within a preset power range. The power generation data includes: power generation and instantaneous power generation.

[0013] Based on the power generation of the experimental group and the control group during all the target analysis periods, a linear relationship model between the power generation of the experimental group and the power generation of the control group is established, and the baseline difference coefficient is calculated.

[0014] In one possible implementation, the step of establishing a linear relationship model between the power generation of the experimental group and the control group based on the power generation of the experimental group and the control group throughout all the target analysis periods, and calculating the benchmark difference coefficient, includes:

[0015] Based on the power generation of the experimental group and the control group within the sliding time window, a linear relationship model between the power generation of the experimental group and the power generation of the control group is established, and the reference benchmark difference coefficient corresponding to each sliding time window is calculated. The sliding time window includes multiple target analysis periods.

[0016] If the standard deviation of each of the reference benchmark difference coefficients is less than the standard deviation threshold, the average value of each of the reference benchmark difference coefficients is determined as the benchmark difference coefficient, or the linear relationship model is re-established based on the power generation of all the target analysis periods, and the benchmark difference coefficient is calculated.

[0017] Given that the reference difference coefficients exhibit a regular drift over time, a functional model of the reference difference coefficients and time parameters is established to obtain the functional expression of the reference difference coefficients with respect to time parameters.

[0018] In one possible implementation, the operational data corresponding to the experimental group and the control group are subjected to interference factor identification, including:

[0019] Obtain the power rationing periods of the preset historical period from the power rationing records, and remove the operating data of the experimental group and the control group during the power rationing periods;

[0020] or

[0021] The power curves of the experimental group and the control group within the preset historical period are obtained. Multiple consecutive analysis periods with plateau or step-like characteristics in the power curves are identified as the power restriction periods, and the operating data of the experimental group and the control group during the power restriction periods are removed.

[0022] In one possible implementation, the operational data corresponding to the experimental group and the control group are subjected to interference factor identification, including:

[0023] Extract the time periods when the tracking angle difference between the experimental group and the control group is consistent within a preset angle range, and the time periods when the tracking angle difference is inconsistent outside the preset angle range;

[0024] For each analysis period, the power generation of the experimental group is corrected using the benchmark difference coefficient, and the difference between the actual power generation of the corrected experimental group and the power generation of the control group is calculated, as well as the power generation difference ratio corresponding to the actual power generation difference. The power generation difference ratio is the ratio of the actual power generation difference to the minimum value between the corrected power generation of the experimental group and the power generation of the control group.

[0025] For each time period with consistent angles, if the difference in power generation exceeds a threshold range, it is determined that the power generation data for the time period with consistent angles is affected by cloud unevenness.

[0026] For each time period with inconsistent angles, the theoretical power generation of the experimental group and the control group is calculated based on the photovoltaic power generation simulation model.

[0027] The theoretical power generation of the experimental group is corrected using the benchmark difference coefficient, and the difference between the corrected theoretical power generation of the experimental group and the theoretical power generation of the control group is calculated.

[0028] If the deviation of the actual power generation difference from the theoretical power generation difference is greater than a first threshold, or if the actual power generation difference and the theoretical power generation difference are opposite in sign, then it is determined that the power generation data during the period of inconsistent angles is affected by cloud unevenness.

[0029] In one possible implementation, the extraction of the tracking angle difference between the experimental group and the control group during periods when the angles are consistent within a preset angle range, and during periods when the tracking angle difference is inconsistent outside the preset angle range, includes:

[0030] Based on the instantaneous power generation or irradiance in the operational data corresponding to the experimental group and the control group, the weather type for each analysis period is determined;

[0031] In the analysis period where the weather type is fluctuating and cloudy, extract the periods with consistent angles and the periods with inconsistent angles.

[0032] In one possible implementation, identifying interference factors in the operational data corresponding to the experimental group and the control group further includes:

[0033] When the experimental group and the control group are located on flat terrain, multiple consecutive time periods with the same angle that have a stepped decrease feature in the power curve of the shadow occlusion period corresponding to the experimental group or the control group are identified as the analysis period affected by shadow occlusion. The shadow occlusion period is determined based on the simulated occlusion data of the experimental group and the control group by a fixed occluder in different analysis periods.

[0034] For the period of inconsistent angles during the period of shadow occlusion, the difference between the actual power generation and the difference between the theoretical power generation corresponding to the period of inconsistent angles are obtained. If the deviation of the difference between the actual power generation and the difference between the theoretical power generation is greater than a second threshold or the difference between the actual power generation and the theoretical power generation are opposite in sign, then it is determined that the power generation data of the period of inconsistent angles is affected by shadow occlusion.

[0035] In one possible implementation, identifying interference factors in the operational data corresponding to the experimental group and the control group further includes:

[0036] When the experimental group and the control group are located on uneven terrain, and the photovoltaic arrays of the experimental group and the control group are arranged in the same way and the slope difference between the photovoltaic arrays is the same, if the difference ratio of the power generation between the experimental group and the control group is less than the third threshold during the shadow shading period, the shadow shading interference identification method for flat terrain is executed.

[0037] In one possible implementation, correcting the power generation data affected by the interference factors based on the benchmark difference coefficient includes:

[0038] For analysis periods affected by cloud unevenness or shadow occlusion, if the power generation of the experimental group is greater than that of the control group, the ratio of the power generation of the control group to the baseline difference coefficient is determined as the corrected power generation of the experimental group.

[0039] For analysis periods affected by cloud unevenness or shadow occlusion, if the power generation of the experimental group is less than that of the control group, the product of the power generation of the experimental group and the baseline difference coefficient is determined as the corrected power generation of the control group.

[0040] In one possible implementation, after correcting the power generation data affected by the interference factors according to the benchmark difference coefficient, the method further includes:

[0041] For analysis periods affected by cloud unevenness or shadow occlusion, the theoretical power generation gain of the experimental group relative to the control group is calculated using a photovoltaic power generation simulation model.

[0042] The theoretical power generation gain is used to compensate for the power generation of the experimental group.

[0043] A second aspect of this application provides an electronic device, including at least one processor and a memory connected to the processor, wherein:

[0044] The memory is used to store computer programs;

[0045] The processor is used to execute the computer program so that the electronic device can implement the photovoltaic tracking algorithm gain evaluation method described in the first aspect or any implementation thereof.

[0046] A third aspect of this application provides a computer program product including computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the photovoltaic tracking algorithm gain evaluation method described in the first aspect or any implementation thereof.

[0047] The fourth aspect of this application provides a computer storage medium carrying one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the photovoltaic tracking algorithm gain evaluation method described in the first aspect or any implementation thereof.

[0048] By employing the above technical solution, this application provides a photovoltaic tracking algorithm gain evaluation method and related apparatus. Based on the operating data of the experimental group and the control group for each analysis period within a preset historical period, it determines the benchmark difference coefficient between the two groups, effectively capturing equipment performance drift caused by factors such as equipment degradation. Then, based on the benchmark difference coefficient, it corrects the power generation data and calculates the power generation gain of the experimental group relative to the control group, ensuring the real-time accuracy of the benchmark comparison between the experimental group and the control group. Simultaneously, by identifying at least one of the interference factors, including cloud inhomogeneity interference, shading interference, and power curtailment interference, and correcting the power generation data affected by these interference factors, it reduces the impact of these interference factors on the power generation gain evaluation results, improving the accuracy and reliability of the power generation gain evaluation results. Attached Figure Description

[0049] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0050] Figure 1 A flowchart illustrating a photovoltaic tracking algorithm gain evaluation method provided in an embodiment of this application;

[0051] Figure 2 A 5-minute power generation curve of an experimental group and a control group is provided for an embodiment of this application;

[0052] Figure 3 A modified 5-minute power generation curve of the experimental group and the control group is provided for an embodiment of this application;

[0053] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0054] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0055] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0056] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0057] This application provides a method for evaluating the gain of a photovoltaic tracking algorithm. The method for evaluating the gain of a photovoltaic tracking algorithm according to this application will be described in detail below with reference to the accompanying drawings.

[0058] Reference Figure 1 , Figure 1 This is a flowchart illustrating a photovoltaic tracking algorithm gain evaluation method provided in an embodiment of this application, as shown below. Figure 1 As shown in the embodiment of this application, a photovoltaic tracking algorithm gain evaluation method may include steps 101 to 105, which are described in detail below.

[0059] 101: Obtain operational data for each analysis period of the experimental and control groups within a preset historical period.

[0060] The photovoltaic array in the experimental group ran the first tracking algorithm, which was the photovoltaic tracking algorithm to be evaluated, while the photovoltaic array in the control group ran the second tracking algorithm, which was the photovoltaic tracking algorithm that had been put into use.

[0061] The length of the preset historical period is set according to the climate of the region where the photovoltaic power station is located. For example, in regions with distinct seasons, the preset historical period can be the past year, while in regions with indistinct seasons, the preset historical period can be the past 6 months. The above are just examples, and the embodiments of this application do not impose specific limitations.

[0062] The analysis period can be set according to the required data processing accuracy, for example, it can be 5 minutes.

[0063] Operational data includes, but is not limited to: tracking angle and power generation data, including instantaneous power generation and power output.

[0064] For example, the runtime data includes the following:

[0065] Tracking angle: Records the real changes in the tracking angle with a time resolution of 1 minute, ensuring accurate recording of every rotation of the photovoltaic tracking bracket;

[0066] Power generation data: Power generation data is preferred, with a time resolution of 5 minutes or more. Compared to instantaneous power generation (kW), power generation is the energy integral over a period of time, which is less sensitive to instantaneous cloud fluctuations and sampling differences. It can more smoothly reflect the actual output of the photovoltaic array and is more suitable for correlation analysis with high stability requirements.

[0067] Supplementary meteorological data: When conditions permit, collect data such as irradiance and module temperature on the photovoltaic array plane. This data can be used for subsequent weather type classification (e.g., sunny, cloudy) to help determine the validity of the data.

[0068] The timestamps of all data sources in the above operational data are strictly synchronized, and outliers such as communication interruptions and data jumps have been removed. The operational data corresponding to the experimental group and the control group are aligned according to a unified timestamp, and the continuous time series data is divided into independent analysis periods according to the set time resolution of power generation statistics (i.e., the above analysis period). The operational data within each analysis period is analyzed as a whole.

[0069] 102: Determine the baseline difference coefficient between the experimental group and the control group based on the corresponding operational data of the experimental group and the control group.

[0070] The operational data of the experimental and control groups were initially screened to exclude severely distorted data significantly affected by cloud unevenness and shadow occlusion. A linear relationship model between the power generation of the experimental group and the power generation of the control group was then established.

[0071] ;

[0072] in, The power generation of the control group is shown. The power generation of the experimental group, The baseline difference coefficient.

[0073] 103: Identify interference factors in the operational data of the experimental and control groups. Interference factors include at least one of the following: cloud inhomogeneity interference, shadow occlusion interference, and power rationing interference.

[0074] Different types of interference factors can be identified simultaneously or sequentially.

[0075] In one possible implementation, the analysis periods affected by power curtailment interference are first identified (i.e., curtailment periods). Since the power generation of the photovoltaic array during curtailment periods is determined by the curtailment command and is unrelated to the performance of the first or second tracking algorithm, it cannot reflect the performance of the first and second tracking algorithms. To eliminate human interference from non-algorithmic factors, the operating data of the experimental and control groups during curtailment periods need to be removed. Then, the analysis periods affected by cloud unevenness are identified and corrected. Finally, the analysis periods affected by shading interference are identified and corrected.

[0076] 104: The power generation data affected by interference factors are corrected based on the baseline difference coefficient to obtain the corrected power generation data for the experimental group and the control group.

[0077] For analysis periods affected by cloud unevenness or shadow occlusion, a conservative correction strategy can be adopted to correct the power generation of groups with higher power generation to the lower power generation level affected by the interference factors. This effectively removes the uneven interference caused by differences in the external environment. Although it may temporarily "mask" the potential gain of the algorithm in that period, it makes the long-term statistical gain results more conservative, robust and reliable, avoiding artificially high values.

[0078] Furthermore, the theoretical power generation gain of the experimental group relative to the control group can be calculated using a photovoltaic power generation simulation model, and the power generation of the experimental group can be compensated using the theoretical power generation gain to make up for the limitations of the conservative correction strategy.

[0079] 105: Based on the baseline difference coefficient and the corrected power generation data of the experimental and control groups, calculate the power generation gain of the experimental group relative to the control group.

[0080] Specifically, the power generation gain of the experimental group relative to the control group can be calculated for the entire preset historical period, one or more analysis periods, etc. The basic calculation formula is as follows:

[0081] ;

[0082] The total power generation of the experimental group is the total power generation of the experimental group during the evaluation period, and the total power generation of the control group is the total power generation of the control group during the evaluation period.

[0083] When the analysis period is weekly, monthly, or quarterly, the weekly, monthly, and quarterly power generation gain change trends can be obtained, which facilitates the analysis of the stability of the first tracking algorithm's performance and whether there are seasonal patterns.

[0084] In addition, the power generation gain of the experimental group relative to the control group can be calculated for different weather types, which is convenient for evaluating the adaptability and advantage boundary of the first and second tracking algorithms under different weather types.

[0085] In one possible implementation, a photovoltaic power generation simulation model can be used to calculate the theoretical power generation gain within the evaluation period, and the actual power generation gain obtained from the above calculation can be compared with the theoretical power generation gain to verify the consistency between theory and reality. If the two trends are consistent and the numerical differences are within a reasonable range, the credibility of the evaluation results can be proven.

[0086] Furthermore, a photovoltaic tracking algorithm gain evaluation report can be generated. The evaluation report can also include an uncertainty analysis of the power generation gain. Based on the error introduced in the correction process, the gain values ​​of the conservative correction strategy and the correction based on the photovoltaic power generation simulation model are calculated respectively, and the possible fluctuation range of the gain value is given, making the test results more rigorous and reliable.

[0087] This embodiment provides a photovoltaic tracking algorithm gain evaluation method. Based on the operational data of the experimental and control groups for each analysis period within a preset historical period, a baseline difference coefficient is determined between the two groups. This effectively captures equipment performance drift caused by factors such as equipment degradation. The power generation data is then corrected based on the baseline difference coefficient, and the power generation gain of the experimental group relative to the control group is calculated, ensuring the real-time accuracy of the comparison between the experimental and control groups. Simultaneously, by identifying at least one of the interference factors, including cloud inhomogeneity, shading interference, and power curtailment interference, and correcting the power generation data affected by these interference factors, the impact of these interference factors on the power generation gain evaluation results is reduced, improving the accuracy and reliability of the power generation gain evaluation results.

[0088] In one possible implementation, step 102 of the above embodiment includes the following steps 1021-1022:

[0089] 1021: Extract the target analysis period where the difference in tracking angle between the experimental group and the control group is within a preset angle range and the difference in instantaneous power generation is within a preset power range. The power generation data includes: power generation and instantaneous power generation.

[0090] For example, the preset angle range can be ±0.5° or ±1°, so that the tracking angles of the experimental group and the control group are the same or the heights are similar during the target analysis period.

[0091] The preset power range is set based on the nominal installed capacity of the experimental group and the control group. For example, the preset power range can be that the instantaneous power difference exceeds the average expected difference by ±5%.

[0092] The purpose of extracting the target analysis period is to eliminate severely distorted data that is obviously affected by cloud unevenness and shadow occlusion.

[0093] 1022: Based on the power generation of the experimental group and the control group during all target analysis periods, establish a linear relationship model between the power generation of the experimental group and the power generation of the control group, and calculate the baseline difference coefficient.

[0094] To reduce the interference of outlier data, robust regression methods, such as robust least squares, can be used to fit the data and establish a linear relationship model between the power generation of the experimental group and the power generation of the control group.

[0095] ;

[0096] Because there is a strong linear correlation between the power generation of the experimental group and the power generation of the control group, the linear relationship model does not include an intercept term. At the same time, it is necessary to ensure that the coefficient of determination R² of the linear regression is greater than 0.999.

[0097] Considering the slow drift of equipment performance (such as degradation, dust accumulation) and seasonal environmental changes, the k value should be dynamic rather than fixed. This embodiment uses a sliding window mechanism (such as weekly or monthly) to repeatedly execute steps 1021-1022, dynamically calculate the baseline difference coefficient corresponding to each sliding time window, and arrange the calculated series of baseline difference coefficients in chronological order to analyze their changing trends.

[0098] If the benchmark difference coefficient remains stable over a long period of time, such as when the standard deviation of each reference benchmark difference coefficient is less than the standard deviation threshold, the average value of each reference benchmark difference coefficient is determined as the benchmark difference coefficient, or a linear relationship model is re-established based on the power generation of all target analysis periods, and the benchmark difference coefficient is calculated.

[0099] If the difference coefficients of each reference benchmark exhibit a regular drift over time, a functional model of the difference coefficients and time parameters can be established to obtain a functional expression of the difference coefficients with respect to time parameters, where the time parameter can be weeks.

[0100] For example, when the value of k exhibits a uniform and stable linear trend over time (e.g., the equipment shows no significant aging acceleration, seasonal effects are weak, or the drift rate is stable during long-term testing), a linear function model can be used, and the function expression can be:

[0101] k(t) = a × t + b;

[0102] t is a time parameter, and the unit is set according to the sliding time window. For example, if the sliding time window is "weekly", then t is the "number of weeks of operation"; if the sliding time window is "monthly", then t is the "number of months of operation", and t≥0.

[0103] a and b are coefficients, obtained by solving the function expression through linear fitting.

[0104] It should also be noted that if the k value changes abruptly, it is necessary to check whether there is equipment failure or a significant change in the performance of the photovoltaic array.

[0105] In one possible implementation, for power-limited interference, step 103 of the above embodiment can be implemented in either method one or method two.

[0106] Method 1

[0107] If power rationing records are available, obtain the power rationing periods for preset historical periods from the power rationing records, and directly remove the operating data of the experimental group and the control group during the power rationing periods.

[0108] Method 2

[0109] In cases where power rationing records cannot be directly obtained or quickly matched, the power curves of the experimental and control groups during a preset historical period are analyzed to identify the power rationing periods by determining multiple consecutive analysis periods with platform or step characteristics in the power curves. This enables automatic identification of power rationing periods and direct removal of the operating data of the experimental and control groups during the power rationing periods.

[0110] (1) Identification of power curve platform features

[0111] Phenomenon: The total output of the power plant or the power of the photovoltaic array remains stable at a certain fixed value or within a very narrow fluctuation range for a long period of time (e.g., more than 30 minutes continuously). This will form a distinct horizontal line on the power curve.

[0112] Identification method: A sliding time window can be used to calculate the standard deviation of power data within the sliding time window. If the standard deviation is consistently below a very small threshold, such as the standard deviation threshold, it can be determined that the period is under power curtailment. If irradiance data for the same period is available, the system can also automatically determine whether the photovoltaic array is under power curtailment based on the fluctuation of irradiance data for the same period or the theoretical power generation.

[0113] (2) Step-like characteristics of power curve

[0114] Phenomenon: The power plant's output experiences a sharp drop or rise within a short period (e.g., within minutes), suddenly shifting from one stable plateau to another and persisting for a period of time. This step-like change may occur multiple times throughout the power curtailment period.

[0115] Identification method: Calculate the first-order difference (instantaneous rate of change) of the power sequences of the experimental group and the control group over a preset historical period, and set a relatively high threshold for change. When the difference value exceeds this threshold, it is marked as a potential "step point". Then check whether the power before and after the step point can form a new stable plateau; if so, it has a step-like characteristic.

[0116] In one possible implementation, to address cloud inhomogeneity interference, step 103 in the above embodiments includes the following steps A1-A6:

[0117] A1: Extract the time periods when the tracking angle difference between the experimental group and the control group is consistent within the preset angle range, and the time periods when the tracking angle difference is inconsistent outside the preset angle range;

[0118] A2: For each analysis period, the power generation of the experimental group is corrected using the baseline difference coefficient, and the difference between the actual power generation of the experimental group and the control group after correction is calculated, as well as the power generation difference ratio corresponding to the difference in actual power generation.

[0119] Actual power generation difference E diff The calculation formula is as follows:

[0120] ;

[0121] in, The power generation is obtained after correcting the power generation of the experimental group using the baseline difference coefficient.

[0122] The power generation difference ratio is the difference between actual power generation. Compared with the power generation of the revised experimental group Power generation compared to the control group The formula for calculating the ratio of the minimum to the maximum power generation difference is as follows:

[0123] .

[0124] A3: For each time period with consistent angles, if the difference in power generation exceeds the threshold range, it is determined that the power generation data for that time period with consistent angles is affected by cloud unevenness.

[0125] For example, the threshold range can be [-1%, 10%], that is... If the percentage is greater than 10% or less than -1%, it indicates that the power generation data for that period of time with the same angle is affected by cloud inhomogeneity.

[0126] A4: For each time period with inconsistent angles, calculate the theoretical power generation of the experimental group and the control group based on the photovoltaic power generation simulation model;

[0127] For periods with inconsistent angles, it is necessary to determine whether the difference in power generation between the experimental and control groups conforms to the power generation difference pattern under the current environment. If the difference pattern deviates significantly, it can be determined that the period was significantly affected by cloud inhomogeneity. To address this, this embodiment uses a photovoltaic power generation simulation model to calculate theoretical power generation to isolate the differences in algorithm strategies and environmental influences.

[0128] Specifically, by inputting the operational data (including tracking angle sequence, irradiance, temperature, and other environmental data) of the experimental and control groups during periods of inconsistent angles into any existing photovoltaic power generation simulation model (such as PVsyst, SAM, or a self-developed model), the theoretical power generation of the experimental group under ideal conditions can be calculated. Theoretical power generation compared to the control group .

[0129] A5: The theoretical power generation of the experimental group is corrected using the benchmark difference coefficient, and the difference between the theoretical power generation of the experimental group and the theoretical power generation of the control group is calculated.

[0130] Difference in theoretical power generation The calculation formula is as follows:

[0131] ;

[0132] in, The theoretical power generation is obtained by correcting the theoretical power generation of the experimental group using the benchmark difference coefficient.

[0133] A6: If the actual power generation difference is E diff Difference relative to theoretical power generation The deviation is greater than the first threshold or the difference between the actual power generation and E. diff Difference from theoretical power generation If the positive and negative values ​​are opposite, it indicates that the power generation data during the period of inconsistent angles is affected by cloud inhomogeneity.

[0134] Among them, the difference in actual power generation E diff Difference relative to theoretical power generation The formula for calculating the deviation is as follows:

[0135] .

[0136] The first threshold can be 30%.

[0137] Understandably, if there is no interference, the actual power generation difference E diff Difference from theoretical power generation The difference should be small and the positive and negative values ​​should be the same. Therefore, if the actual power generation difference E diff Difference relative to theoretical power generation The deviation is large or the difference between the actual power generation and E is significant. diff Difference from theoretical power generation If the positive and negative values ​​are opposite, it indicates that the power generation data during the period of inconsistent angles is affected by cloud inhomogeneity.

[0138] Weather types can include: stable sunny, fluctuating cloudy, and uniformly overcast. On stable sunny days, the tracking angles of the experimental and control groups are generally the same, and the difference in power generation between the experimental and control groups after correction using the baseline difference coefficient is very small, generally unaffected by cloud unevenness. On uniformly overcast days, the cloud distribution is relatively uniform, resulting in uniform scattered light, and similarly, there is no interference from cloud unevenness. Therefore, cloud unevenness interference identification is not necessary on stable sunny and uniformly overcast days.

[0139] To reduce the data processing workload for identifying cloud unevenness interference, one possible implementation is to determine the weather type for each analysis period based on the instantaneous power generation or irradiance in the operational data corresponding to the experimental and control groups. In the analysis period where the weather type is fluctuating cloudy, the periods with consistent and inconsistent angles mentioned in steps A1-A6 above are extracted; that is, cloud unevenness interference identification is performed only for fluctuating cloudy days. It should be noted that since fluctuating cloudy days are likely to increase the power generation gain of the experimental group and increase the difference in actual power generation between the experimental and control groups, the threshold range can be widened, such as [-2%, 20%]. Furthermore, the method for determining the weather type for the analysis period in this embodiment can be any existing weather type identification method; this embodiment does not impose specific limitations.

[0140] To address shadow occlusion interference, the shadows cast by fixed obstructions (such as buildings, prefabricated substations, and mountains) exhibit a regularity in specific seasons and time periods. Before identifying shadow occlusion interference, detailed site mapping can be conducted, and photovoltaic 3D simulation software (such as PVsyst) can be used to simulate the shadow occlusion caused by these fixed obstructions on the experimental and control groups at different analysis periods, thereby obtaining the shadow occlusion periods and assisting in subsequent shadow occlusion interference identification.

[0141] In one possible implementation, the corresponding shadow occlusion interference identification methods can differ when the experimental group and the control group are located in different terrains.

[0142] When the experimental and control groups are located on flat terrain, they are generally only shaded by fixed obstructions. In the above embodiment, one implementation of step 103 includes the following steps B1-B2:

[0143] B1: The analysis period affected by shadow occlusion is determined by identifying multiple consecutive angles with a stepped decrease characteristic in the power curve of the shadow occlusion period corresponding to the experimental group or control group.

[0144] Among them, the power curves of the experimental group or control group during the shadow occlusion period have a step-like decline characteristic, specifically: the instantaneous power generation during multiple consecutive periods with the same angle continuously decreases, the decline is greater than the threshold, and the instantaneous power generation during multiple consecutive periods with the same angle after the decline is continuously lower than the set value, that is, the power curve drops sharply and remains low for a period of time.

[0145] B2: For periods of inconsistent angles during periods of shading, obtain the difference between actual power generation and theoretical power generation corresponding to the periods of inconsistent angles. If the deviation between the actual power generation difference and the theoretical power generation difference is greater than the second threshold, or if the actual power generation difference and the theoretical power generation are opposite in sign, then it is determined that the power generation data during the periods of inconsistent angles is affected by shading.

[0146] The period of inconsistent angles usually occurs when the experimental group is severely shaded. The intelligent tracking bracket algorithm driven by inverter power data (i.e., the first tracking algorithm mentioned above) will identify the tracking status as cloudy and rotate at a small angle. In this case, it is necessary to use a photovoltaic power generation simulation model to calculate the theoretical power generation to separate the differences in algorithm strategy and environmental influences.

[0147] Specifically, by inputting the operational data (including tracking angle sequence, irradiance, temperature, and other environmental data) of the experimental and control groups during periods of inconsistent angles into any existing photovoltaic power generation simulation model (such as PVsyst, SAM, or a self-developed model), the theoretical power generation of the experimental group under ideal conditions can be calculated. Theoretical power generation compared to the control group The theoretical power generation of the experimental group was corrected using a baseline difference coefficient, and the difference between the corrected theoretical power generation of the experimental group and the theoretical power generation of the control group was calculated. If the deviation between the actual power generation difference and the theoretical power generation difference is greater than the second threshold, or if the actual power generation difference and the theoretical power generation difference are opposite in sign, then the power generation data for the period with inconsistent angles is affected by shadow occlusion.

[0148] For the division of time periods with consistent angles and time periods with inconsistent angles in the above shadow occlusion interference identification, please refer to the division method of steps A1-A6 in the above cloud unevenness interference identification, which will not be repeated here.

[0149] When the experimental and control groups are located on uneven terrain, the inverse tracking optimization function of the first tracking algorithm running in the experimental group is generally required for testing. This function mainly identifies shading caused by uneven terrain between arrays, optimizes the inverse tracking angle, and reduces power generation loss caused by shading between photovoltaic arrays. For uneven terrain, the power generation difference between the experimental and control groups is generally required to be less than a third threshold, which can be 1%. The impact of shading between photovoltaic arrays on their power generation patterns is complex; therefore, the photovoltaic arrays in the experimental and control groups must be arranged identically with consistent slope differences to ensure nearly identical shading patterns and accurately test the true effectiveness of the inverse tracking optimization function. When the photovoltaic arrays in the experimental and control groups are arranged identically with consistent slope differences, during shading periods, if the power generation difference between the experimental and control groups is less than the third threshold, steps A1-A2 above are used to identify shading interference from fixed obstructions.

[0150] The following describes a method for correcting power generation data during analysis periods that are affected by cloud inhomogeneity or shadow occlusion.

[0151] In one possible implementation, a conservative correction strategy is employed to correct power generation data for analysis periods that are affected by cloud inhomogeneity or shadowing.

[0152] During periods affected by uneven cloud cover, the same cloud may only cover one photovoltaic array in the experimental and control groups. This results in the photovoltaic array with higher power generation not being affected by cloud unevenness. A conservative correction strategy is adopted to uniformly correct the power generation of the experimental and control groups to a lower level affected by cloud cover, effectively eliminating the uneven interference caused by differences in the external environment. Although this may temporarily "mask" the potential gain of the algorithm during that period, it makes the long-term statistical gain results more conservative, robust and reliable, avoiding artificially high values.

[0153] Similarly, during the analysis period affected by shadowing, a conservative correction strategy was adopted to uniformly correct the power generation of the experimental and control groups to a lower level affected by shadowing.

[0154] Specifically, the conservative correction strategy is implemented using the following steps 1041-1042:

[0155] 1041: If the power generation of the experimental group is greater than that of the control group, then the ratio of the power generation of the control group to the baseline difference coefficient shall be determined as the corrected power generation of the experimental group.

[0156] The formula for correcting the power generation of the experimental group is:

[0157] .

[0158] 1042: If the power generation of the experimental group is less than that of the control group, the product of the power generation of the experimental group and the baseline difference coefficient shall be determined as the corrected power generation of the control group.

[0159] The correction formula for the power generation of the control group is:

[0160] .

[0161] Furthermore, the theoretical power generation gain of the experimental group relative to the control group can be calculated using a photovoltaic power generation simulation model, and the power generation of the experimental group can be compensated using the theoretical power generation gain to make up for the limitations of the conservative correction strategy.

[0162] Specifically, for analysis periods affected by cloud unevenness or shading, the operational data of the experimental and control groups (including tracking angle sequences, irradiance, temperature, and other environmental data) are input into the existing photovoltaic power generation simulation model. This allows the calculation of the theoretical power generation of the experimental group under ideal conditions. Theoretical power generation compared to the control group .

[0163] Calculate theoretical power generation gain :

[0164] ;

[0165] Using theoretical power generation gain The power generation of the experimental group was compensated, and the power generation of the experimental group after compensation was... as follows:

[0166] .

[0167] This implementation adopts a differentiated correction strategy, which combines a conservative correction strategy with correction of the power generation simulation model to ensure the accuracy of the correction results.

[0168] The following example demonstrates how to evaluate the performance of the first tracking algorithm under fluctuating and cloudy conditions using weather type identification and a conservative correction strategy.

[0169] On a specific test day at a test power plant, significant uneven cloud distribution was observed based on instantaneous power generation or irradiance, indicating periods of fluctuating cloudy weather. The 5-minute power generation curves for the experimental and control groups are shown below. Figure 2As shown, there are two significant instances of uneven cloud distribution throughout the day. For example, at 9:00 AM, the experimental group generated 41.34 kWh of electricity in 5 minutes, while the test group generated 73.26 kWh in 5 minutes. The test group's electricity generation was 77% higher than that of the experimental group. At this time, the tracking angle of the experimental group was 35°, while that of the test group was 40°. Such a huge difference in electricity generation is clearly not caused by the difference in tracking angle, but rather by the uneven cloud distribution, which resulted in a large difference in the amount of light received by the experimental and control groups.

[0170] Using the photovoltaic tracking algorithm gain evaluation method provided in this application, cloud inhomogeneity interference identification and a conservative correction strategy were performed to correct power generation. The corrected 5-minute power generation curves of the experimental group and the control group are shown below. Figure 3 As shown, a conservative correction strategy is used to correct power generation, which effectively avoids the disturbance of uneven cloud distribution on the algorithm gain evaluation.

[0171] Further analysis of relevant data revealed that the baseline difference coefficient k for that day was 1.0085. The experimental group generated 6547.82 kWh of electricity, while the control group generated 6536.06 kWh. The corrected generation of the control group was 6453.86 kWh. The generation gain of the experimental group was calculated, showing gains of -0.664% and 0.601% compared to the control group before and after correction, respectively. This day was typically mostly sunny, with a combination of some drastically fluctuating cloudy weather and stable overcast skies in the afternoon. The algorithm's main gain occurred during the stable overcast period in the afternoon, and the corrected gain was closer to the theoretical expectation, demonstrating the effectiveness of this method in removing interference from cloudy weather.

[0172] This application also provides an electronic device in its embodiments. (See reference...) Figure 4 The diagram illustrates a structural schematic suitable for implementing the electronic devices in the embodiments of this application. The electronic devices in these embodiments can be terminal devices deployed in photovoltaic power plants, such as smartphones, tablets, laptops, desktop computers, etc.; the electronic devices can also be network-side devices deployed outside the photovoltaic power plant, such as servers or server clusters, in the cloud or elsewhere. Figure 4 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0173] like Figure 4As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage device 408 into a random access memory (RAM) 403. When the electronic device is powered on, the RAM 403 also stores various programs and data required for the operation of the electronic device. The processing unit 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.

[0174] Typically, the following devices can be connected to I / O interface 405: input devices 606 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.; storage devices 408 including, for example, memory cards, hard drives, 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. More or fewer devices may be implemented or have alternatively.

[0175] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the photovoltaic tracking algorithm gain evaluation methods provided in this application.

[0176] This application also provides a computer-readable storage medium that carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any of the photovoltaic tracking algorithm gain evaluation methods provided in this application.

[0177] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0178] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0179] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0180] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

Claims

1. A method for evaluating the gain of a photovoltaic tracking algorithm, characterized in that, include: The experimental group and the control group obtained operational data for each analysis period within a preset historical period. The photovoltaic array of the experimental group ran a first tracking algorithm, and the photovoltaic array of the control group ran a second tracking algorithm. The operational data included tracking angle and power generation data. Based on the operational data corresponding to the experimental group and the control group, the baseline difference coefficient between the experimental group and the control group is determined; Interference factors are identified in the operational data corresponding to the experimental group and the control group. The interference factors include at least one of the following: cloud unevenness interference, shadow occlusion interference, and power rationing interference. The power generation data affected by the interference factors are corrected according to the benchmark difference coefficient to obtain the corrected power generation data for the experimental group and the control group. Based on the baseline difference coefficient and the corrected power generation data of the experimental group and the control group, the power generation gain of the experimental group relative to the control group is calculated.

2. The photovoltaic tracking algorithm gain evaluation method according to claim 1, characterized in that, The step of determining the baseline difference coefficient between the experimental group and the control group based on the corresponding operational data includes: Extract the target analysis period where the tracking angle difference between the experimental group and the control group is within a preset angle range and the instantaneous power generation difference is within a preset power range. The power generation data includes: power generation and instantaneous power generation. Based on the power generation of the experimental group and the control group during all the target analysis periods, a linear relationship model between the power generation of the experimental group and the power generation of the control group is established, and the baseline difference coefficient is calculated.

3. The photovoltaic tracking algorithm gain evaluation method according to claim 2, characterized in that, The step involves establishing a linear relationship model between the power generation of the experimental group and the control group during all target analysis periods, based on their respective power generation data across all target analysis periods, and calculating the baseline difference coefficient, including: Based on the power generation of the experimental group and the control group within the sliding time window, a linear relationship model between the power generation of the experimental group and the power generation of the control group is established, and the reference benchmark difference coefficient corresponding to each sliding time window is calculated. The sliding time window includes multiple target analysis periods. If the standard deviation of each of the reference benchmark difference coefficients is less than the standard deviation threshold, the average value of each of the reference benchmark difference coefficients is determined as the benchmark difference coefficient, or the linear relationship model is re-established based on the power generation of all the target analysis periods, and the benchmark difference coefficient is calculated. Given that the reference difference coefficients exhibit a regular drift over time, a functional model of the reference difference coefficients and time parameters is established to obtain the functional expression of the reference difference coefficients with respect to time parameters.

4. The photovoltaic tracking algorithm gain evaluation method according to claim 1, characterized in that, Identify interference factors in the operational data corresponding to the experimental group and the control group, including: Obtain the power rationing periods of the preset historical period from the power rationing records, and remove the operating data of the experimental group and the control group during the power rationing periods; or The power curves of the experimental group and the control group within the preset historical period are obtained. Multiple consecutive analysis periods with plateau or step-like characteristics in the power curves are identified as the power restriction periods, and the operating data of the experimental group and the control group during the power restriction periods are removed.

5. The photovoltaic tracking algorithm gain evaluation method according to claim 1 or 4, characterized in that, Identify interference factors in the operational data corresponding to the experimental group and the control group, including: Extract the time periods when the tracking angle difference between the experimental group and the control group is consistent within a preset angle range, and the time periods when the tracking angle difference is inconsistent outside the preset angle range; For each analysis period, the power generation of the experimental group is corrected using the benchmark difference coefficient, and the difference between the actual power generation of the corrected experimental group and the power generation of the control group is calculated, as well as the power generation difference ratio corresponding to the actual power generation difference. The power generation difference ratio is the ratio of the actual power generation difference to the minimum value between the corrected power generation of the experimental group and the power generation of the control group. For each time period with consistent angles, if the difference in power generation exceeds a threshold range, it is determined that the power generation data for the time period with consistent angles is affected by cloud unevenness. For each time period with inconsistent angles, the theoretical power generation of the experimental group and the control group is calculated based on the photovoltaic power generation simulation model. The theoretical power generation of the experimental group is corrected using the benchmark difference coefficient, and the difference between the corrected theoretical power generation of the experimental group and the theoretical power generation of the control group is calculated. If the deviation of the actual power generation difference from the theoretical power generation difference is greater than a first threshold, or if the actual power generation difference and the theoretical power generation difference are opposite in sign, then it is determined that the power generation data during the period of inconsistent angles is affected by cloud unevenness.

6. The photovoltaic tracking algorithm gain evaluation method according to claim 5, characterized in that, The extraction of the tracking angle difference between the experimental group and the control group within a preset angle range during time periods when the angles are consistent, and the extraction of the tracking angle difference outside the preset angle range during time periods when the angles are inconsistent, includes: Based on the instantaneous power generation or irradiance in the operational data corresponding to the experimental group and the control group, the weather type for each analysis period is determined; In the analysis period where the weather type is fluctuating and cloudy, extract the periods with consistent angles and the periods with inconsistent angles.

7. The photovoltaic tracking algorithm gain evaluation method according to claim 5, characterized in that, The process of identifying interference factors in the operational data corresponding to the experimental group and the control group also includes: When the experimental group and the control group are located on flat terrain, multiple consecutive time periods with the same angle that have a stepped decrease feature in the power curve of the shadow occlusion period corresponding to the experimental group or the control group are identified as the analysis period affected by shadow occlusion. The shadow occlusion period is determined based on the simulated occlusion data of the experimental group and the control group by a fixed occluder in different analysis periods. For the period of inconsistent angles during the period of shadow occlusion, the difference between the actual power generation and the difference between the theoretical power generation corresponding to the period of inconsistent angles are obtained. If the deviation of the difference between the actual power generation and the difference between the theoretical power generation is greater than a second threshold or the difference between the actual power generation and the theoretical power generation are opposite in sign, then it is determined that the power generation data of the period of inconsistent angles is affected by shadow occlusion.

8. The photovoltaic tracking algorithm gain evaluation method according to claim 7, characterized in that, The process of identifying interference factors in the operational data corresponding to the experimental group and the control group also includes: When the experimental group and the control group are located on uneven terrain, and the photovoltaic arrays of the experimental group and the control group are arranged in the same way and the slope difference between the photovoltaic arrays is the same, if the difference ratio of the power generation between the experimental group and the control group is less than the third threshold during the shadow shading period, the shadow shading interference identification method for flat terrain is executed.

9. The photovoltaic tracking algorithm gain evaluation method according to claim 1, characterized in that, The step of correcting the power generation data affected by the interference factors based on the benchmark difference coefficient includes: For analysis periods affected by cloud unevenness or shadow occlusion, if the power generation of the experimental group is greater than that of the control group, the ratio of the power generation of the control group to the baseline difference coefficient is determined as the corrected power generation of the experimental group. For analysis periods affected by cloud unevenness or shadow occlusion, if the power generation of the experimental group is less than that of the control group, the product of the power generation of the experimental group and the baseline difference coefficient is determined as the corrected power generation of the control group.

10. The photovoltaic tracking algorithm gain evaluation method according to claim 9, characterized in that, After correcting the power generation data affected by the interference factors according to the benchmark difference coefficient, the method further includes: For analysis periods affected by cloud unevenness or shadow occlusion, the theoretical power generation gain of the experimental group relative to the control group is calculated using a photovoltaic power generation simulation model. The theoretical power generation gain is used to compensate for the power generation of the experimental group.

11. An electronic device, characterized in that, It includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program to enable the electronic device to implement the photovoltaic tracking algorithm gain evaluation method as described in any one of claims 1 to 10.

12. A computer program product, characterized in that, It includes computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the photovoltaic tracking algorithm gain evaluation method as described in any one of claims 1 to 10.

13. A computer storage medium, characterized in that, The storage medium carries one or more computer programs that, when executed by an electronic device, enable the electronic device to implement the photovoltaic tracking algorithm gain evaluation method as described in any one of claims 1 to 10.

Images