Photovoltaic power station model determination method and system, and medium

Abstract

The invention relates to the technical field of new energy, in particular to a photovoltaic power station model determination method and system and a medium. The method comprises the following steps: acquiring photovoltaic power station information; performing meteorological resource evaluation according to the photovoltaic power station information to obtain meteorological resource data; predicting generating capacity according to the meteorological resource data to obtain generating capacity data; performing component energy loss analysis according to the generating capacity data to obtain energy loss data; performing power generation performance degradation detection based on the energy loss data to obtain power generation performance degradation data; performing component aging evaluation according to the power generation performance degradation data to obtain component aging data; and performing power generation capacity prediction correction on the power generation capacity data according to the assembly aging data to obtain corrected power generation data. According to the method, the power generation capacity is dynamically corrected and the photoelectric conversion efficiency is accurately evaluated based on the new energy technology, so that the accuracy and reliability of determining the photovoltaic power station model are improved.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, and in particular to a method, system and medium for determining a photovoltaic power plant model. Background Technology

[0002] Traditional methods for determining photovoltaic (PV) power plant models typically rely on static model parameters and idealized weather conditions to predict power generation, neglecting the impact of performance degradation and material aging of PV modules on model accuracy over long-term operation. Establishing fixed models based on initial design parameters lacks dynamic perception and feedback adjustments to factors such as energy loss, inverter efficiency changes, encapsulation material degradation, and light transmittance decline during actual operation. This leads to significant discrepancies between modeling results and actual power generation performance, hindering refined prediction and performance evaluation. Furthermore, the lack of multi-scale modeling capabilities fails to incorporate microscopic influences such as changes in the optical response of encapsulation materials and thermal stress aging paths, making it difficult to meet the emerging demands of intelligent operation and maintenance, performance prediction, and lifespan management of PV systems. This limits its application in the full-lifecycle digital management of large-scale PV power plants. Summary of the Invention

[0003] Therefore, the present invention needs to provide a method, system and medium for determining a photovoltaic power plant model to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for determining a photovoltaic power plant model includes the following steps: Step S1: Obtain photovoltaic power station information; conduct meteorological resource assessment based on photovoltaic power station information to obtain meteorological resource data; predict power generation based on meteorological resource data to obtain power generation data; Step S2: Analyze the energy loss of the components based on the power generation data to obtain energy loss data; detect the degradation of power generation performance based on the energy loss data to obtain power generation performance degradation data; evaluate the aging of the components based on the power generation performance degradation data to obtain component aging data. Step S3: Based on the component aging data, perform power generation capacity prediction and correction on the power generation data to obtain corrected power generation data; perform encapsulation material degradation analysis based on the component aging data to obtain encapsulation material degradation data; perform light transmittance anomaly analysis based on the encapsulation material degradation data to obtain light transmittance anomaly data. Step S4: Detect the light absorption capacity of the module based on the abnormal light transmittance data to obtain the light absorption capacity data of the module; determine the photoelectric conversion efficiency of the photovoltaic power station model based on the light absorption capacity data of the module and the corrected power generation data to obtain the photoelectric conversion efficiency data.

[0005] This invention addresses the issue that traditional photovoltaic (PV) power plant modeling methods typically rely on static model parameters and idealized weather conditions for power generation prediction, neglecting the impact of performance degradation and material aging of PV modules on model accuracy over long-term operation. Establishing a fixed model based on initial design parameters lacks dynamic perception and feedback adjustment for factors such as energy loss, inverter efficiency changes, encapsulation material degradation, and light transmittance decline during actual operation. This leads to significant discrepancies between the modeled results and actual power generation performance, hindering refined prediction and performance evaluation. Furthermore, the invention fails to incorporate the microscopic effects of changes in the optical response of encapsulation materials and thermal stress aging paths, lacking multi-scale modeling capabilities. This makes it difficult to meet the emerging demands of intelligent operation and maintenance, performance prediction, and lifespan management of PV systems, limiting its application in the full-lifecycle digital management of large-scale PV power plants.

[0006] Preferably, this specification also provides a photovoltaic power plant model determination system for executing the production parameter optimization method for lightweight glass bottles as described above. The photovoltaic power plant model determination system includes: The power generation prediction module is used to acquire photovoltaic power plant information; conduct meteorological resource assessment based on the photovoltaic power plant information to obtain meteorological resource data; and predict power generation based on the meteorological resource data to obtain power generation data. The module aging assessment module is used to analyze the energy loss of the module based on the power generation data to obtain energy loss data; to detect the degradation of power generation performance based on the energy loss data to obtain power generation performance degradation data; and to assess the aging of the module based on the power generation performance degradation data to obtain module aging data. The transmittance anomaly analysis module is used to predict and correct the power generation capacity of the power generation data based on the component aging data to obtain corrected power generation data; to perform encapsulation material degradation analysis based on the component aging data to obtain encapsulation material degradation data; and to perform transmittance anomaly analysis based on the encapsulation material degradation data to obtain transmittance anomaly data. The photoelectric conversion efficiency determination module is used to detect the light absorption capacity of the module based on the abnormal light transmittance data, and obtain the light absorption capacity data of the module; based on the light absorption capacity data of the module and the corrected power generation data, the photoelectric conversion efficiency of the photovoltaic power station model is determined, and the photoelectric conversion efficiency data is obtained.

[0007] The photovoltaic power plant model determination system of the present invention can implement any of the photovoltaic power plant model determination methods of the present invention. It is used to combine the operation and signal transmission medium between various modules to complete the photovoltaic power plant model determination method. The internal modules of the system cooperate with each other, dynamically correct the power generation capacity and accurately evaluate the photoelectric conversion efficiency, thereby improving the accuracy and reliability of photovoltaic power plant model determination.

[0008] Optionally, this specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements any one of the photovoltaic power plant model determination methods. Attached Figure Description

[0009] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the steps in the method for determining a photovoltaic power station model according to the present invention; Figure 2 This is a detailed flowchart of step S1 in the present invention; Figure 3 This is a schematic diagram of the photovoltaic power station equipment in this invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0010] The technical method of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0011] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0012] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0013] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a method for determining a photovoltaic power plant model, the method comprising the following steps: Step S1: Obtain photovoltaic power station information; conduct meteorological resource assessment based on photovoltaic power station information to obtain meteorological resource data; predict power generation based on meteorological resource data to obtain power generation data; In this embodiment, basic information about the target photovoltaic power station is obtained through database access or physical surveying. This information includes geographical coordinates (latitude and longitude), altitude (meters), installation tilt angle (degrees), module array orientation angle (degrees), module model, cell type (e.g., crystalline silicon or thin film), and installed capacity (kW). This data is then used as the basic input. Based on these location coordinates, data extraction is performed using a public API interface or a third-party platform (such as Meteonorm) to obtain hourly meteorological data for the past ten years. Data fields include annual sunshine hours and total solar radiation (W / m²). 2 The data includes temperature, humidity, wind speed, and rainfall. Then, based on the total solar radiation data and array tilt angle, irradiance is converted. The ISO 52010-1 standard is used to convert the irradiance of the tilted surface, obtaining the annual average effective radiation value per unit area. Using PVWatts or HelioScope tools, hourly power generation simulations are performed based on radiation data, battery module specifications, temperature coefficient, reflection loss rate, shading loss rate, and other factors, outputting the total annual power generation (unit: kWh) and monthly power generation curves as power generation data.

[0014] Step S2: Analyze the energy loss of the components based on the power generation data to obtain energy loss data; detect the degradation of power generation performance based on the energy loss data to obtain power generation performance degradation data; evaluate the aging of the components based on the power generation performance degradation data to obtain component aging data. In this embodiment, the actual monitored power generation data is compared with the theoretical power generation predicted in step S1, and the deviation ratio between the two is calculated on a monthly or quarterly basis. The formula is: Loss rate = (Theoretical power generation - Actual power generation) / Theoretical power generation × 100%. When the loss rate exceeds 10%, it is recorded as significant energy loss. The loss data is used to build a time-series dataset in chronological order. Subsequently, according to the IEC 61724-1 standard, the power degradation rate is fitted using the least squares method through the historical trends of the module open-circuit voltage (Voc), short-circuit current (Isc), and maximum power point (Pm), to obtain the annual average power degradation rate (unit: % / year). If the degradation rate exceeds 2%, it is considered as performance degradation according to EN50530. Then, combined with the module operating temperature (Tmod) and cumulative ultraviolet radiation (UV value), the Arrhenius aging model is used to calculate the active layer degradation. The failure rate of the PN junction was determined by scanning capacitance degradation test (SCDT), and the crack growth rate was extracted by scanning electron microscopy (SEM) images of the material fracture surface. Thus, an aging degree classification curve was established, which was divided into three levels: Level 1 (slight), Level 2 (moderate), and Level 3 (severe), corresponding to different component aging datasets.

[0015] Step S3: Based on the component aging data, perform power generation capacity prediction and correction on the power generation data to obtain corrected power generation data; perform encapsulation material degradation analysis based on the component aging data to obtain encapsulation material degradation data; perform light transmittance anomaly analysis based on the encapsulation material degradation data to obtain light transmittance anomaly data. In this embodiment, a power generation capacity correction curve is first established based on the component aging level and annual average power decay rate obtained in step S2. Setting the initial rated power of the component as P0, the annual decay coefficient as k, and the number of years as n, the predicted power is then... The power generation records are retrospectively revised quarterly. To further analyze the material encapsulation performance, EVA (ethylene-vinyl acetate) encapsulation layer samples were extracted from the component surface, and Fourier transform infrared spectroscopy (FTIR) was used to detect the peak positions of EVA molecular bond breakage. Differential scanning calorimetry (DSC) was used to measure the glass transition temperature change; if the temperature change exceeded ±5℃, it was determined to be due to thermo-oxidative aging. Integrating sphere ultraviolet-visible spectroscopy was used to measure the transmittance curve of the EVA layer in the 280-800nm ​​band, especially detecting its yellowing effect in the 400-600nm range; if the transmittance decreased by more than 15%, an anomaly indicator was output. Finally, the transmittance anomaly data were used to establish multivariate time series data indexed by timestamps, including dimensional features such as band, transmittance, and difference rate.

[0016] Step S4: Detect the light absorption capacity of the module based on the abnormal light transmittance data to obtain the light absorption capacity data of the module; determine the photoelectric conversion efficiency of the photovoltaic power station model based on the light absorption capacity data of the module and the corrected power generation data to obtain the photoelectric conversion efficiency data.

[0017] In this embodiment, based on the transmittance anomaly values ​​extracted in step S3 for each band, and using the AM1.5 standard spectrum as a reference input, the radiant energy per unit area for each band is integrated to obtain the effective incident light energy (unit: W / m²). 2 Further, the reflectance of the component's absorption layer was collected, and the actual light absorptivity was calculated using the integrating sphere reflectance test method combined with a spectral reflectance model. The formula is: Absorbance in For reflectivity, Transmittance. This yields data on the component's light absorption capacity. Then, based on the corrected power generation data and the calculated effective absorbed energy ratio, the photoelectric conversion efficiency η is calculated using the following formula: ,in The corrected output power (unit: W). In order to absorb energy, Component area (unit: m²) 2 Finally, the conversion efficiency will be output quarterly and a deviation analysis will be performed on it compared with industry standard reference values ​​(such as the typical conversion efficiency of crystalline silicon modules being 18%-22%), which will be used as the output parameters of the photovoltaic power plant model.

[0018] Of particular importance is that step S41 includes the following steps: Step S41: Perform transmission energy distribution integration processing based on the transmittance anomaly data to obtain spectral transmission energy data; In this embodiment, after detecting transmittance anomalies, the transmission characteristics of the abnormal regions need to be quantified. The instrument used for detection is an ultraviolet-visible-near-infrared spectrometer with a measurement wavelength range of 300–1100 nm and a resolution of 1 nm. A spectral transmittance testing system is formed by combining the component packaging material sample with a standard solar spectral light source (AM1.5G spectral density). Transmitted light intensity values ​​are collected using an integrating sphere, and the instrument's built-in integration function is used to multiply the transmittance at each wavelength by the standard spectral energy density. Finally, integration is performed across the entire band to extract the band-level "transmitted energy" value. The integration method uses discrete gradient integration, where the measurement data is mixed wavelength by wavelength, and the total transmitted energy is calculated by summing the results. The unit is W / m². 2The measurement environment must be maintained at a temperature of 22±1°C and a relative humidity of 45–55%. Instrument calibration should adopt a secondary light source standard, with calibration intervals not exceeding three months, ensuring data accuracy within ±2%. The integrated output transmittance energy data includes fields such as sample number, measurement time, Wavelength_range, transmittance curve, and integral value (W / m²). 2 ).

[0019] Step S42: Calculate the light energy loss rate based on the spectral transmittance energy data; In this embodiment, after obtaining the transmitted energy value of the component, the light energy loss rate is calculated. First, the standard irradiation incident energy is obtained from the standard installation angle. This value is the total energy value of AM1.5G, approximately 1000W / m. 2 Using the transmitted energy as the measured value, the measured value is compared with 1000 W / m. 2 The ratio calculation is performed, and the light energy loss rate is defined as (1000 − transmitted energy) / 1000 × 100%, representing the proportion of energy lost during the light transmission process of the material encapsulation. The transmitted energy of each component sample is compared with 1000 W / m². 2 The ratio is used to calculate the optical energy loss rate. Due to sampling error, the allowable range is ±0.5%; measurements outside this range must be repeated. The final output table includes the following fields: sample number, measurement time, and transmitted energy (W / m²). 2 Light energy loss rate (%).

[0020] Step S43: Evaluate the light absorption capacity of the module based on the light energy loss rate to obtain the light absorption capacity data of the module; In this embodiment, the measured light energy loss rate is used as a reverse indicator of the module's light absorption capacity, converted into an absorption capacity value. The module's light absorption capacity is defined as 1 − light energy loss rate (e.g., a loss rate of 20% corresponds to an absorption capacity of 80%). To further correct for the actual power absorption of the module, a reflectivity compensation factor needs to be added. The reflectivity value is measured using an integrating sphere reflectance test, with the measurement band set to match the transmission range. Combining the transmittance (T) and reflectance (R) data, the absorptivity (A = 1 - TR) can be calculated. The absorptivity is output as the module's light absorption capacity value and recorded in percentage form. The environmental test conditions are kept consistent with step S41, ensuring a temperature of 22±1°C and humidity of 45–55%. The final module light absorption capacity data table includes the following fields: module number, average transmittance, average reflectance, light absorption capacity (%), and measurement time.

[0021] Step S44: Determine the photoelectric conversion efficiency of the photovoltaic power station model based on the component light absorption capacity data and the corrected power generation data, and obtain the photoelectric conversion efficiency data.

[0022] In this embodiment, this step combines light absorption capacity data with corrected power generation data to calculate the photoelectric conversion efficiency (η). First, the corrected power generation output P_out (in kWh) of the same component or component group is collected, converted into average power (W), and the output power per unit time is calculated. Then, the energy absorbed per unit area of ​​the component (obtained from the spectral transmittance in step S41) is acquired, with units of W / m². 2 And multiply by the actual light-receiving area of ​​the component (m²) 2 The total absorbed energy is obtained. The output power is divided by the absorbed energy to obtain the efficiency value η = P_out / (absorbed energy × area), expressed as a percentage. To reduce the impact of weather fluctuations, efficiency data is averaged monthly or quarterly, and data with daily transmission and radiation less than 200 W / m² is excluded. 2 The data is recorded for low-light conditions. A final summary table of photoelectric conversion efficiency data is generated, with each row containing fields such as component number, measurement period, average output power (W), absorbed energy (W), and photoelectric conversion efficiency (%). Database records include calibration certificate number and test condition information, ensuring an accuracy of ±3%. For further long-term analysis, the data can be accumulated and summarized into a power plant-wide photoelectric conversion efficiency report.

[0023] Preferably, step S1 includes the following steps: Step S11: Obtain photovoltaic power station information and determine the location of the photovoltaic power station to obtain location data; In this embodiment, basic information about the photovoltaic power station is collected through engineering construction documents, site archives, and project approval materials. This information includes, but is not limited to, the power station name, component model, inverter type, rated power, installed capacity, bracket type, and layout. A high-precision handheld GNSS locator is used on-site to obtain the longitude, latitude, and altitude information of the power station's center point. The WGS84 coordinate system is used for recording, and the data is retained to five decimal places. The GNSS device samples five times per second, continuously sampling for ten minutes and then averaging the results to avoid positioning offsets caused by instantaneous errors. If location data cannot be obtained on-site, satellite imagery is overlaid with the power station planning map. The spatial vector overlay function is used in the GIS platform to extract the center position of the map patch, ensuring that the extracted data is consistent with the actual layout. The final output is standard location data, including the fields: power station number, longitude, latitude, altitude, orientation angle, and tilt angle.

[0024] Step S12: Match meteorological observation stations based on location data to obtain observation station data; In this embodiment, the obtained latitude and longitude of the power station are used as the center point. The scope is expanded outwards in increments of 30 kilometers from the meteorological information service platform or an internationally available meteorological database, with a maximum range not exceeding 100 kilometers. Meteorological stations with at least 10 years of hourly recording capabilities are selected, prioritizing those equipped with Class I or Class II benchmark stations, such as solar total radiation observers, sunshine duration meters, and wind speed and humidity monitoring devices. Point layers of each meteorological station are loaded into the GIS platform. After filtering stations that meet distance constraints using a spatial buffer algorithm, spatial connectivity is used to extract data such as the meteorological station number, altitude, type of observation equipment, establishment year, and meteorological record completeness rate. Structured observation station data is output as the basis for subsequent meteorological data acquisition.

[0025] Step S13: Extract historical meteorological records based on observation station data; calculate the total solar radiation based on historical meteorological records; In this embodiment, based on the meteorological station numbers matched in the previous stage, hourly historical meteorological data for the past ten years is retrieved from the meteorological data center via a data interface. The format requirement is a CSV structure with timestamps, and each record includes indicators such as radiation, sunshine duration, wind speed, relative humidity, and temperature. The collected historical data undergoes field standardization processing, converting the time field to the local time zone, and using a sliding window validation method to remove record segments with more than three consecutive hours of missing data. The radiation field is statistically summarized, with hourly radiation values ​​accumulated over a year. The daily maximum, minimum, and annual average values ​​are detected, and the number of valid records each day is recorded to ensure the continuity of the time series. Finally, total solar radiation data at daily and monthly scales are output.

[0026] Step S14: Perform seasonal illumination fluctuation analysis based on total solar radiation to obtain seasonal fluctuation data; In this embodiment, daily total solar radiation data are grouped by month, and the average and standard deviation for each month are calculated to determine the seasonal fluctuation trend of annual solar radiation. Typical months with high and low radiation are identified by arranging the annual average values ​​month by month. Differences in monthly average radiation between different seasons are statistically analyzed, and the fluctuation range is assessed by combining the monthly standard deviation. Based on this, monthly seasonal fluctuation data is generated. To ensure the stability of the analysis, monthly data for at least five consecutive years are analyzed, and the same month is compared year by year to confirm its annual consistency and extreme value fluctuations. Output fields include month, monthly average radiation, maximum value, minimum value, and fluctuation range, forming a standardized seasonal fluctuation data table.

[0027] Step S15: Calculate the effective sunshine duration based on seasonal fluctuation data to obtain effective sunshine duration data; conduct an irradiance suitability assessment based on the effective sunshine duration data to obtain meteorological resource data; In this embodiment, hourly radiation data is read annually, and a radiation threshold of 120 watts per square meter is set. A Boolean judgment is performed to determine whether the threshold is reached every hour, and the effective sunshine period is recorded. The effective sunshine duration is accumulated daily, and the effective sunshine duration for each day of the year is recorded in hours. Then, the total effective sunshine hours for the whole year are calculated and compared with the annual total of 8760 hours to calculate the annual sunshine utilization rate. Based on this, the balance of sunshine resource distribution at the power station's location in different seasons is determined according to the average effective sunshine duration for different months. Using an annual average effective sunshine hours exceeding 1400 hours as the minimum threshold, and combined with the seasonal fluctuation index in step S14, a resource level assessment is performed for each observation period. The output meteorological resource data includes: annual effective hours, monthly average effective hours, fluctuation coefficient, resource level (e.g., excellent, good, moderate), data coverage, etc.

[0028] Step S16: Predict power generation based on meteorological resource data to obtain power generation data.

[0029] In this embodiment, the component parameters include the area of ​​a single component, the component conversion efficiency, the installation angle, the ventilation conditions on the back of the component, and the temperature coefficient. The temperature coefficient value is extracted based on the experimental report provided by the component manufacturer and is uniformly accurate to 0.01% per degree Celsius. Using daily total radiation data and effective sunshine duration data as inputs, the theoretical power generation is calculated based on the effective received radiation after correction according to the component tilt angle and orientation. Simultaneously, a performance ratio factor is introduced to reflect various systemic losses in actual power plant operation, such as cable loss, inverter loss, and temperature decay. The performance ratio factor is set between 0.75 and 0.82, selected from the historical operating experience of the local power plant. Daily power generation is predicted and accumulated to obtain the theoretical total annual power generation. Finally, structured power generation data is output, including fields such as: date, daily radiation, component efficiency, theoretical power generation, performance ratio, and annual cumulative power generation.

[0030] Preferably, step S16 includes the following steps: Step S161: Obtain the photovoltaic power station model and identify the radiation receiving structure; In this embodiment, based on the completion of the on-site survey and engineering drawing data analysis, the structural information of the photovoltaic power station needs to be obtained, and the component structures used to receive solar radiation need to be identified. This step first involves calling 3D modeling drawings or BIM system data to clarify the installation method of the components, extracting the positioning coordinates, tilt angle, orientation, spacing, number of rows and columns, installation height, and array arrangement of each component. During the identification of the radiation-receiving structure, all physical structural units capable of receiving solar radiation need to be focused on, with only the front of the component panel retained as the effective receiving surface. The component name needs to be further specified as either a crystalline silicon component or a thin-film component, and the component model needs to clearly indicate the panel size (e.g., 1722mm × 1134mm), single-unit rated power (e.g., 550W), open-circuit voltage (e.g., 49.5V), and short-circuit current (e.g., 13.8A). On-site, the installation angle and arrangement height are measured using a laser rangefinder and a laser tilt meter, with a measurement error not exceeding 1.0%. The final structured data table of the receiving radiation structure of the component is formed, which includes component number, component center coordinates (latitude and longitude), installation tilt angle (degrees), azimuth angle (angle relative to due south), effective area of ​​receiving surface (square meters), array obstruction distance, component type and number, etc.

[0031] Step S162: Perform radiation irradiation simulation on the receiving radiation structure based on meteorological resource data, where the direct radiation is set to 300-950 W / m. 2 To obtain radiation exposure data; In this embodiment, solar radiation simulation is performed based on the received radiation structure data output from the previous stage, combined with meteorological resource data of the photovoltaic power station's location. The radiation simulation process is based on hourly solar radiation data and comprehensively considers the angle correction of the incident radiation caused by the module tilt angle and azimuth angle. The direct radiation range is set to 300 W / m² during the simulation. 2 Up to 950W / m 2 According to 50W / m 2 The simulation was performed in segments with varying step sizes. Diffuse reflection and scattered radiation components were superimposed within each simulation segment. Diffuse reflection was processed using a fixed ground reflectance value of 0.2, while scattered radiation was corrected using measured atmospheric visibility and cloud cover data from reference observation stations. All simulation calculations were based on the SOLPOS solar position algorithm to determine the solar altitude angle and azimuth angle for each day and hour. An incident angle correction coefficient was calculated by combining the component orientation and tilt angle. This coefficient was used for radiation reception only when the angle between the incident angle and the normal was less than 85°. Hourly cumulative statistics of incident radiation were performed on a unit area of ​​each receiving surface, generating a time-series data table indexed by the component number. Fields included time (e.g., 2024-06-01 12:00) and direct radiation (unit: W / m²). 2 ), scattered radiation (W / m 2 Corrected total radiation (W / m) 2), angle of incidence on the receiving surface (degrees), actual received radiation (W / m²) 2 The final output is 8760 hours of radiation exposure data for the whole year.

[0032] Step S163: Perform power generation conversion calculation based on radiation irradiation data to obtain power generation data.

[0033] In this embodiment, the hourly actual received radiation data obtained in step S162 is read. Based on the nominal conversion efficiency of the module, the conversion efficiency is corrected by combining the temperature coefficient of the module type and environmental meteorological data, and then the power generation conversion calculation is performed. For example, a monocrystalline PERC module with an efficiency of 23.1%. The temperature correction coefficient is based on the module model datasheet, such as −0.35% / ℃, and the real-time operating temperature of the panel is calculated by combining the ambient air temperature data and the NOCT (normal operating temperature) model. The NOCT value is based on the value shown in the module datasheet, such as 45℃, and the final module temperature rise value is obtained after correction with reference to the ambient wind speed and installation height. In the power generation conversion process, the power generation per unit hour is equal to the radiation per unit area multiplied by the module conversion efficiency multiplied by the module area, and then multiplied by the performance ratio factor. It is recommended that the performance ratio factor be fixed at 0.80 (representing systemic losses such as inverter, connecting wires, dust, heat attenuation, etc.), and the hourly power generation data is accurate to 0.01kWh. By summing up all the hourly data, the theoretical power generation data for each module is obtained for daily, monthly, and annual periods. The final power generation data file is then generated, with fields including module number, time, hourly power generation (kWh), cumulative daily power generation, cumulative annual power generation, conversion efficiency, temperature correction factor, performance ratio, etc.

[0034] Preferably, step S2 includes the following steps: Step S21: Perform component loss analysis based on power generation data to obtain component loss data; In this embodiment, hourly measured power generation data for the most recent 12 consecutive months needs to be obtained. This data should be uploaded in real-time by the power plant's data acquisition unit and recorded via a remote terminal. The data format should be uniformly timestamp + power (unit: kWh). The theoretical predicted power generation is output in step S16, calculated based on Standard Radiation Conditions (STC) and component performance parameters. For comparison, the two types of data need to be aligned at the same time granularity, and linear interpolation should be used to fill in missing data periods (filling should not exceed 3 consecutive hours of missing data; otherwise, the period is discarded). A comparison threshold is set; when the measured power generation is less than 80% of the theoretical power generation, that period is recorded as an abnormal loss period. Daily and monthly cumulative power loss is calculated, and combined with the corresponding component array number, component groups are partitioned for analysis. Groups with losses exceeding 5% of the total annual power generation are marked as key focus areas. Component loss data is output, including fields such as each component group number, theoretical power generation, measured power generation, power loss, loss percentage, and number of anomalies.

[0035] Of particular importance, step S21 includes the following steps: Step S211: Identify the periods of sharp drop in power generation based on the power generation data, and obtain the data on the periods of sharp drop in power generation; In this embodiment, power generation data of the target photovoltaic power station is collected with a time resolution of no less than 5 minutes and a data duration covering no less than 60 consecutive solar cycles. The data is normalized according to solar cycle time, and the minimum start time and maximum sunset end time during the day are set. Subsequently, the power curve within each solar cycle is processed using first-order difference to extract the rate of power decrease within a continuous time period. A threshold for identifying a sharp drop in power generation is set, defined as the ratio of the power difference between two consecutive time periods to the power value of the previous time period. If this ratio is greater than 0.35 and the duration of two consecutive drops exceeds 10 minutes, it is identified as a period of sharp drop in power generation. To further exclude the influence of shading and sudden weather changes, the power generation trend comparison of adjacent days or adjacent module groups needs to be introduced to determine whether it is a systemic anomaly. The final output data of the period of sharp drop in power generation should include fields such as timestamp, module number, start and end time of drop, and percentage of drop rate, which serve as the basis for the target time period for subsequent thermal imaging detection.

[0036] Step S212: Perform thermal imaging detection of the module cells based on the data during the period of sharp drop in power generation to obtain thermal imaging data; In this embodiment, on-site infrared thermal imaging detection is arranged during the identified period of sharp drop in power generation. A non-contact infrared thermal imager is used, with a wavelength range covering 8 to 14 micrometers, a thermal sensitivity of no more than 0.05°C, and a spatial resolution of no less than 640×480 pixels. The detection time must be controlled during peak power generation, preferably between 12:00 PM and 2:00 PM on a sunny day. The thermal imaging device must be equipped with GPS-synchronized timestamps and perform a full thermal image scan of each target component. The shooting distance is controlled between 1.2 meters and 1.5 meters, and the shooting angle should maintain a vertical deviation of no more than 10 degrees from the component surface. Each image frame should have a temperature calibration function and automatically mark the isotherm distribution of the temperature range in the shooting area. The obtained thermal images are archived, and data parameters such as the maximum temperature, minimum temperature, average temperature, and hotspot location of each area in the image are extracted to form thermal imaging data for subsequent cold spot identification and analysis.

[0037] Step S213: Identify edge cold spots based on thermal imaging data to obtain cold spot data; In this embodiment, a gray-scale distribution-based image segmentation algorithm is used to extract temperature gradient features from the acquired thermal imaging image data. Edge recognition boundaries are set for the surrounding edge regions of the component, defined as the range from 0 to 50 mm from the component boundary. Pixels in this region with significantly lower temperature values ​​than the regional average are clustered. A cold spot identification temperature difference threshold of 8°C is set; if the temperature of a connected region is lower than the average of the surrounding regions by this threshold and the continuous area exceeds 10 cm², then the cold spot identification is considered successful. 2 If a region is found to be abnormal, it is marked as a cold spot. Gaussian filtering is used to preprocess the thermal image to eliminate the influence of local abnormal noise. Combined with the component layout logic, the consistency of time periods of multiple frames is compared to eliminate the influence of transient shadows. Finally, cold spot data is generated, including cold spot location coordinates (X,Y), area, edge distance, minimum temperature value and relative temperature difference, cold spot quantity, average spacing and other indicators, which are uniformly organized into component edge cold spot data.

[0038] Step S214: Determine the cell breakage of the module based on the cold spot data, and obtain the cell breakage data; In this embodiment, the identified cold spot locations are overlaid and matched with the module cell layout. If the overlap rate between the cold spot distribution and the solder joints between cells exceeds 70%, and its length exceeds one-third of the length of a single module cell, then a cell fracture is identified at that location. Fracture determination requires further integration with EL images or historical EL data. If the cold spot area shows black bands or crack propagation marks in the EL image, the reliability of the fracture confirmation is enhanced. Furthermore, if the cold spot is located between multiple series-connected cells and forms a continuous cold area, it is considered a series path fracture. The degree of fracture is indicated by fracture levels: Level 1 is a slight surface crack with no output impact; Level 2 is a local fracture with power reduction; and Level 3 is a severe fracture causing series failure. The final output cell fracture data includes parameters such as fracture area number, number of cold spots, fracture level, thermal map coordinates, and module number.

[0039] Step S215: Based on the cell breakage data, the module loss level is calibrated to obtain the module loss data.

[0040] In this embodiment, the loss level is classified based on the fracture level distribution in the cell fracture data and the module power test data. First, a standard power output test of the module is performed using standard illumination (AM1.5, 1000W / m²) at 1000V. 2The actual output power of the modules is tested under the specified conditions. The tested value is compared with the factory nominal power. If the output decrease is less than 5%, it is classified as Level I loss; 5% to 15% is Level II loss; and more than 15% is Level III loss. The power attenuation level and the failure level are fused in a matrix to construct a loss level assessment rule base. For example, if the failure level is Level II and the power decrease is 8%, the module is judged as Level II loss. The codes for each loss level are D1, D2, and D3, respectively. Finally, the loss results are output in a structured manner to generate a module loss dataset, which includes fields such as module ID, failure level, power attenuation percentage, loss level, and detection time, and serves as the module performance input parameters in the photovoltaic power plant model.

[0041] Step S22: Analyze inverter losses based on power generation data to obtain inverter loss data; In this embodiment, power data from the DC input and AC output terminals of the inverter are collected, requiring a data sampling frequency of no less than once every 15 minutes, with a collection period of 8760 hours per year. Each inverter is equipped with an independent power data acquisition device, and an index mapping is established through the correspondence between the inverter number and the component group. According to IEC standard 61724, the inverter efficiency is calculated as the ratio of AC output power to DC input power, and the daily efficiency fluctuation of each inverter is recorded. A reference efficiency is set as the inverter nameplate efficiency minus 2% (to account for heat loss and aging). If the daily efficiency of an inverter is more than 5% lower than the reference efficiency (i.e., the efficiency is too low and exceeds the threshold), it is determined that the inverter has significant losses, and parameters such as the efficiency anomaly value, duration, and lowest efficiency value are recorded. At the same time, the ambient temperature and inverter casing temperature are collected and the heat dissipation status is analyzed to ensure that the temperature rise does not exceed the rated value of 15°C. Finally, inverter loss data is generated, with fields including inverter number, input power, output power, actual efficiency, reference efficiency, anomaly duration, and cumulative power loss.

[0042] Step S23: Detect power generation performance degradation based on energy loss data to obtain power generation performance degradation data; In this embodiment, energy loss data is generated by merging the aforementioned component loss and inverter loss data, and the actual energy loss ratio is calculated using an energy balance method. The difference between the theoretical total power generation and the grid-connected power generation is defined as the total energy loss. After removing the inverter loss, the component-side loss value is obtained. Based on this, power generation performance degradation detection is performed, with a detection period set at three years. Representative component groups are selected as detection objects, and their monthly power generation data are fitted with linear regression. The slope value in the regression result reflects the component performance degradation rate. If the regression slope is less than 0.5 percentage points per year, it is determined that the component has a significant degradation trend. Further, the cumulative energy loss of each component group is normalized, and the power generation change rate corresponding to each thousand hours of operation is calculated as a degradation intensity indicator. At the same time, parameters such as cumulative operating time, historical maximum monthly power generation, and average power generation of the most recent six months are recorded. The final output of power generation performance degradation data includes component number, annual degradation rate, regression slope, cumulative operating hours, standardized power generation decline rate, and degradation level. The degradation level is divided into mild, moderate, and severe, with the judgment criteria being an annual degradation rate of no more than 0.5 percentage points, more than 0.5 percentage points but no more than 1 percentage point, and more than 1 percentage point, respectively.

[0043] Step S24: Collect ultraviolet irradiation data based on power generation performance degradation data; perform photo-oxidation reaction analysis on the material based on the ultraviolet irradiation data to obtain photo-oxidation reaction data; perform chain breakage detection on the material based on the photo-oxidation reaction data to obtain chain breakage data. In this embodiment, according to the aforementioned degradation component area numbering, light and ultraviolet radiation acquisition devices are set up in the target area. The acquisition cycle is once every 5 minutes, and the continuous observation time is not less than 30 days. The measuring device must meet the ISO9060 Class 1 spectral sensor standard, be able to cover the 280-400nm ultraviolet band, and have an irradiance measurement error of no more than ±2%. The cumulative ultraviolet irradiation is counted, and the distribution of UV-A (315-400nm) and UV-B (280-315nm) bands is analyzed. Combined with the irradiation time and temperature field data, the daily average total UV radiation energy per unit area is calculated, with the unit being kJ / m². 2 ·day. Based on the known photo-oxidation reaction threshold of polymer materials (usually set at 200kJ / m²), 2 ), marking areas of excessive UV exposure. The material's reactivity constant K value (obtained through laboratory testing, unit: min) was used. -1 The reaction rate was estimated by comparing the cumulative ultraviolet irradiation. Next, molecular bond identification was performed on the field component samples using Fourier transform infrared spectroscopy (FTIR), detecting changes in the characteristic peak positions of C–C, C–O, and C=O bond breaking, with a scanning range of 400–4000 cm⁻¹. -1 Resolution not less than 4cm -1If the carbon-oxygen bond peak intensity decreases by more than 15% while the carbonyl group formation rate increases by more than 10%, it is determined that chain scission has occurred in the material. The final output is material chain scission data, with fields including component number, sampling location, cumulative UV irradiation, chain scission frequency, functional group change, and chain scission confirmation result.

[0044] Step S25: Perform component aging assessment based on material chain breakage data to obtain component aging data.

[0045] In this embodiment, a multi-parameter coupled evaluation method is adopted, using chain breakage frequency, degradation rate, color change index, and water vapor transmission rate as inputs. The color change index is obtained by acquiring the yellowing value ΔYI using a digital image colorimeter, and the yellowing threshold is set to ΔYI>12. The water vapor transmission rate is tested using the calcium chloride desiccator method, with the test temperature fixed at 38℃ and relative humidity at 90%, and the test period at 240 hours. If the transmission rate is higher than 3.0 g / m 2 The date · is recorded as a day of deterioration in sealing performance. A weighted comprehensive score is applied to all degradation parameters, with chain breakage weighted at 0.4, degradation rate at 0.3, color change index at 0.2, and water vapor transmission rate at 0.1, forming an aging score index for each component. The final component aging data structure includes fields such as component number, chain breakage parameter, optical degradation index, encapsulation aging index, aging score, and recommended maintenance level (scores below 70 indicate medium aging, and below 50 indicate severe aging).

[0046] Please see Figure 3 This is a schematic diagram of the photovoltaic power station equipment in this invention, including a heat collection point and the main body of the photovoltaic inverter; Preferably, step S22 includes the following steps: Step S221: Collect inverter heat data based on power generation data to obtain inverter heat data; In this embodiment, when collecting heat from the inverter, it is necessary to first determine the inverter's serial number, rated power, rated efficiency, packaging level, cooling method, and its installation location in the photovoltaic system. Each inverter must be equipped with at least one thermocouple temperature sensor and one power acquisition module. The thermocouple temperature sensor uses K-type or T-type thermocouples with a response time of less than 5 seconds and a temperature measurement accuracy of no less than ±1℃. The temperature measurement position is fixed at the center area of ​​the inverter's heat sink surface. The temperature acquisition frequency is set to once every 1 minute, and the data recording period is 30 consecutive days. The collected data includes the surface temperature of the inverter's heat sink, the internal air duct temperature, and the ambient temperature. The power acquisition module measures the input DC power and output AC power, with a sampling frequency of once every 5 minutes and an accuracy better than 0.5%. The heat calculation method is as follows: first, using the difference between the inverter's internal DC input and AC output power, combined with the sampling time, the heat power generation is calculated in W; then, the heat power is combined with the inverter's structural heat capacity parameters to estimate the inverter's cumulative heat energy in J. To ensure data stability, data collection must be completed under constant ambient temperature conditions, with fluctuations not exceeding ±2℃. The final output inverter thermal data fields include: inverter number, timestamp, DC power, AC power, thermal power (W), surface temperature (℃), ambient temperature (℃), and cumulative thermal energy (J).

[0047] Step S222: Perform conduction loss analysis based on inverter heat data to obtain conduction loss data; In this embodiment, the conduction loss analysis relies on the thermal power data obtained in step S221 and the conduction parameters of the power devices (IGBTs or MOSFETs) inside the inverter. First, the transistor on-resistance (Rds_on) or saturation voltage drop (Vce_sat) from the inverter's technical manual is obtained, and the overall conduction path resistance is determined based on the device's current carrying capacity and the number of devices. For example, for a typical three-phase inverter, if each phase uses two parallel IGBTs with a Vce_sat of 1.6V and a current carrying capacity of 30A, the conduction power consumption is calculated as: P = 1.6V × 30A × 3 phases = 144W. The actual conduction loss value needs to be adjusted based on the on-site measured input current and voltage sampling data. The input current sampling error should not exceed ±1%, and a Hall current sensor or shunt should be used for acquisition; the voltage sampling accuracy should not be less than 0.5%. The conduction loss is calculated using the point-by-point instantaneous power method, that is, the instantaneous voltage and current under the device's conduction state at each time point are multiplied and summed to obtain the total conduction power. The final output conduction loss data fields include: inverter number, time, input current, device on-state voltage drop, instantaneous conduction power, average conduction power, and conduction loss percentage.

[0048] Step S223: Perform temperature rise statistics based on conduction loss data, wherein the temperature rise response time is set to 2-20 min, and the temperature rise response data is obtained; In this embodiment, when performing temperature rise statistics, the temperature rise response time period must first be defined as the minimum duration from the start of inverter load change to temperature stabilization, with a range of 2 to 20 minutes. During sampling, a typical day with drastic inverter load changes (e.g., midday on a sunny day) is selected for monitoring. A time period with a temperature change rate greater than 0.5℃ / min is selected, and the initial temperature T0 and the final stable temperature T1 are recorded. The response time is defined as the shortest time required to reach the T1±1℃ range, in minutes. If the response time is greater than 15 minutes three times consecutively within a certain period, or the maximum temperature rise exceeds 25℃, the inverter is marked as being in a high-heat-sensitive state. The statistical process requires simultaneous acquisition of ambient temperature to correct for the heat rise (T1−T0−T_ambient temperature change). The final output temperature rise response data fields include: inverter number, load change time, T0, T1, response time, maximum temperature rise, ambient temperature change, and temperature rise correction value.

[0049] Step S224: Perform thermal cycling stress detection on the solder joint based on the temperature rise response data to obtain the thermal stress of the solder joint; In this embodiment, the thermal stress detection of the solder joint is based on the principle of thermal cycling fatigue. By analyzing the number of temperature rises and the temperature difference, the stress level borne by the solder joint is evaluated. A temperature rise cycle is defined as a complete process of the temperature rising from a reference temperature to a peak temperature and then returning. Each cycle with a temperature difference greater than 10°C is considered a valid thermal cycle. The statistical period is 30 consecutive days of operation, during which the start and end times of each temperature rise, the peak temperature difference, and the duration are recorded. Using the temperature difference ΔT and the number of thermal cycles N as inputs, the plastic strain of the solder joint material (e.g., SnAgCu) is estimated according to the Coffin-Manson failure criterion. Where C and n are empirical material constants, with C≈0.5 and n≈0.6 for SnAgCu alloy. The equivalent stress (in MPa) of the weld joint is calculated based on strain data, and the maximum stress value and frequency distribution are recorded. The final output weld joint thermal stress data includes: inverter number, number of thermal cycles, average temperature difference, maximum temperature difference, material type, equivalent stress, strain amplitude, and weld joint failure trend.

[0050] Step S225: Perform insulation failure detection based on the thermal stress of the solder joint to obtain insulation failure data; determine the degree of inverter loss based on the insulation failure data to obtain inverter loss data.

[0051] In this embodiment, insulation failure detection is performed using a combination of dielectric breakdown voltage method and insulation resistance monitoring. High-voltage test points are placed in areas where the thermal stress at the solder joints exceeds a critical value (e.g., SnAgCu solder joint stress > 40MPa). The test voltage level is set to 1000VDC, and the testing cycle is set to once a month. An insulation resistance tester (e.g., a megohmmeter) is used to measure the insulation resistance between the inverter casing and the DC bus, and between the casing and the AC terminals. An insulation resistance below 1MΩ is considered a potential failure. A constant voltage is further applied to the inverter with potential failure, and the resistance value is observed to continue to decrease within 5 minutes. If the decrease exceeds 20%, insulation aging is considered. This determination is linked to the aforementioned solder joint thermal stress data to confirm the carbonization or cracking path of the insulation material induced by thermal stress, ultimately quantifying the degree of failure. The inverter loss level is calculated as follows: the ratio of the time due to insulation failure causing downtime or efficiency decline to the total time in the cumulative annual power generation time, multiplied by the annual power generation loss percentage. The output inverter loss data fields include: inverter number, insulation resistance value, breakdown voltage value, cumulative hours of failure in the year, annual power generation loss value (kWh), loss percentage (%), and inverter loss level.

[0052] Preferably, step S3, which involves correcting the power generation data based on component aging data to predict power generation capacity, includes: Identify and output performance degradation factors based on component aging data; In this embodiment, the module aging data includes the photovoltaic module's service life, cumulative irradiance, light-induced degradation rate (LID), hot spot distribution, cumulative ultraviolet radiation dose, backsheet yellowing level, number of cracks in the EL (electroluminescent) image, and damp heat aging level. The cumulative irradiance is calculated using the average daily irradiance per year on a horizontal plane (unit: kWh / m²). 2 The LID value is calculated by multiplying the number of operating days by the number of days, with the range set between 1% and 3% according to the IEC 61215 standard. The total number of cracks and hotspot density are extracted by manually scanning EL images, and the aging levels are divided into four categories: 0–10, 11–30, 31–50, and above 50. For each module number, its historical power generation data and rated factory parameters are matched, and the output degradation rate is calculated using the difference between the monthly average power generation and the theoretical power over a five-year period. For example, if the annual output degradation of a module is 2.1%, then 2.1% is used as its annual performance degradation factor (DF), and a unified degradation factor table is output for each module. The data fields include: module number, installation time, cumulative irradiance, number of EL cracks, hotspot level, and annual average degradation rate (%).

[0053] Output power generation attenuation curve based on performance degradation factor; In this embodiment, power generation performance degradation detection is performed based on energy loss data to obtain power generation performance degradation data. In this embodiment, energy loss data is generated by merging the aforementioned component loss and inverter loss data, and the actual loss ratio is calculated using the energy balance method. The difference between the theoretical total power generation and the grid-connected power generation is defined as the total energy loss, and subtracting the inverter loss gives the component-side loss. Power generation performance degradation detection is performed based on component losses. First, the detection period is determined to be three years, and representative component groups are selected. Monthly power generation data is used for linear regression fitting, and the regression slope reflects the component performance degradation rate. If the linear regression slope is less than 0.5 percentage points per year, a significant degradation trend is identified. Cumulative loss normalization is performed on each component group, and the power generation change rate corresponding to each thousand hours of operation is calculated as a degradation intensity indicator. Parameters such as operating time, historical highest monthly power generation, and average power generation over the past six months are recorded. Output power generation performance degradation data, including component number, degradation rate, regression slope, cumulative operating hours, standardized power generation decline rate, and degradation level. The degradation level is divided into mild, moderate, and severe, with the boundaries set as an annual degradation rate not exceeding 0.5 percentage points, between 0.5 and 1 percentage point, and above 1 percentage point, respectively.

[0054] Calculate the power generation attenuation factor based on the power generation attenuation curve; In this embodiment, after constructing the power generation attenuation curve, the percentage output value at the end of year N is extracted, and the ratio of this value to the rated output power is defined as the power generation attenuation factor. This factor is used as a weighting coefficient for subsequent performance fitting correction. For a module that has been operating for 5 years, if the output at the end of the curve is 91.5%, the attenuation factor is 0.915. If a local hot spot active area is detected during actual operation, and its peak temperature rise exceeds 25°C, the attenuation factor of the module is further reduced by 3%, corrected to 0.887. This factor is categorized and organized according to the average output of the strings under the jurisdiction of each inverter, serving as a representative attenuation value at the inverter level. The final output attenuation factor table fields include: inverter number, string number, operating years, output percentage, attenuation factor value, and correction flag (such as "hot spot adjustment" or "EL abnormal adjustment").

[0055] Performance ratio fitting correction is performed based on the power generation attenuation factor to obtain performance ratio correction data; In this embodiment, the Performance Ratio (PR) is an important parameter for measuring the efficiency of a photovoltaic system, defined as: PR = Actual Power Generation / Theoretical Power Generation. After obtaining the attenuation factor at the string or module level, it is used to correct the original performance ratio data. The system PR is calculated on a monthly basis. If the PR of a string is 85% and the attenuation factor is 0.915, then the theoretically corrected PR is: 85% / 0.915 ≈ 92.9%. This operation is implemented through a hierarchical matching method: PR databases are established at the module, string, and inverter levels respectively, and the corresponding correction coefficients are updated level by level. If the measured daily average PR of a string in a certain month is 0.80, and the theoretical solar irradiance is 5.2 kWh / m², then... 2 With a rated output of 5kW, its theoretical power generation is 5 × 5.2 × 30 = 780kWh, while the actual output is 624kWh. The corrected reference value should be: 624 / 0.915 ≈ 682kWh, and the corrected PR is 682 / 780 ≈ 87.4%. The final output performance ratio correction table fields include: time, component number, original PR value, attenuation factor, corrected PR value, and correction difference (%).

[0056] Based on the performance ratio correction data, the power generation data is corrected by predicting and correcting the power generation capacity to obtain the corrected power generation data.

[0057] In this embodiment, after obtaining the performance ratio correction data, it is used as a multiplier factor for power generation forecast correction in future power generation estimates. The correction method is as follows: the theoretical power generation obtained based on the meteorological resource prediction model is multiplied by the performance ratio correction factor to obtain the corrected power generation capacity. For example, if the theoretical power generation of a certain string in May 2026 is predicted to be 860 kWh, and its performance ratio correction value is 92.4%, then the corrected power generation is 860 × 0.924 ≈ 794.64 kWh. The correction operation needs to be combined with the seasonal light attenuation factor, temperature influence factor, and dust shading factor for the corresponding month (e.g., pollution level 3 corresponds to 5% shading). All correction coefficients need to be listed in tabular form and archived uniformly. Output fields include: month, string number, theoretical power generation (kWh), corrected PR value, environmental attenuation coefficient, temperature correction value, and corrected power generation (kWh).

[0058] Preferably, step S3, which involves analyzing the degradation of the encapsulation material based on the component aging data, includes: Identify aging areas of components based on component aging data; In this embodiment, the component aging data mainly includes electroluminescence (EL) imaging, infrared thermal imaging, cumulative ultraviolet light irradiation dose, operating time, and historical performance data. By collecting component EL imaging data, a high-sensitivity CCD camera with a resolution of at least 2048×2048 pixels is used to scan each component individually, identifying internal cracks, solder joint detachment, microcracks, and hotspot areas. By comparing the grayscale values ​​of each pixel, a grayscale threshold of 180 (0-255 grayscale scale) is set, and areas with a grayscale value significantly lower than the threshold and a continuous area greater than 50 square millimeters are identified as aging areas. An infrared thermal imager is used to scan the surface temperature of the components, and areas with a hotspot temperature difference exceeding 10 degrees Celsius are identified as key aging areas. A two-dimensional aging area distribution map is generated by correlating the collected aging data coordinates with the component's physical coordinates. The area data includes coordinate position (X, Y axes, unit mm) and area (mm²). 2 Temperature deviation (°C) and crack density were also measured. Data from the aging area were saved as a structured table for subsequent material performance testing.

[0059] Thermo-oxidative aging test of materials is performed based on the aging area of ​​the component to obtain thermo-oxidative aging data; In this embodiment, material samples, particularly those of the encapsulation layer and backplane, were collected from the aging area. The standard thickness for material samples was 0.5 mm to 1.0 mm. Differential scanning calorimetry (DSC) was used for thermal stability testing, with the test temperature range set from room temperature (25°C to 300°C) at a heating rate of 5°C / min. Thermogravimetric curves were recorded. Thermogravimetric analysis (TGA) was used to determine the mass change of the sample when heated to 300°C in air; a mass loss greater than 5% was defined as significant thermo-oxidative aging. Fourier transform infrared spectroscopy (FTIR) was used to test the chemical bond breaking of the samples, analyzing the carbonyl peak (1700 cm⁻¹). 1 (near) and hydroxyl peak (3400cm⁻) 1 The intensity change of the absorption peak (near the target area) is used as a quantitative indicator of the degree of oxidation. Thermo-oxidative aging data includes the percentage of mass loss, the intensity of characteristic absorption peaks and their relative change rates, which are normalized using a relative numerical scale of 0 to 1 to output a thermo-oxidative aging index table for the material.

[0060] Yellowing analysis was performed based on thermo-oxidative aging data to obtain yellowing data for the material; In this embodiment, samples were taken from material samples after thermo-oxidative aging treatment. The transmittance and color change of the material were measured using a spectrophotometer. The measurement wavelength range was set to 380 nm to 780 nm. The color of the sample was quantified using the CIE 1976Lab color space, and the color difference value ΔE was calculated. The color change of the thermo-oxidative aged sample was compared with that of the unaged material sample. When the color difference value exceeded 5, it was judged as obvious yellowing. The measurement content mainly included three colorimetric parameters: L (brightness), a (red-greenness), and b (yellow-blueness). The increase of the b value indicates the degree of yellowing. The measurement process was carried out under standard light source D65 illumination conditions, with the ambient temperature controlled at 23 degrees Celsius ± 2 degrees Celsius and the humidity controlled within the relative humidity range of 50% ± 5%. The measured color difference parameters and transmittance change data were summarized to form a material yellowing data table. The data table includes sample number, test time, color difference value, L value, a value, b value, and corresponding transmittance percentage to ensure that the measurement data is accurate and repeatable.

[0061] Spectral response shift detection was performed based on material yellowing data to obtain spectral response shift data; In this embodiment, the spectral transmittance curve of a material sample after yellowing was measured using a UV-Vis-NIR spectrometer, covering a wavelength range of 300 nm to 1100 nm with a step size of 1 nm. The collected data is a continuous curve of the material's transmittance as a function of wavelength. By comparing the transmittance curve with that of the unaged material, the response shift in the wavelength-sensitive region was calculated. Particular attention was paid to the spectral range of 400-700 nm, where the transmittance decreased by more than 5%, which was defined as the response shift band. By comparing the decrease in peak transmittance within the band, a response shift threshold of 10% was set. The weighted average change in transmittance over the entire spectral range was calculated using integral spectral analysis and quantified as a response shift index. A spectral response shift data table was output, including indicators such as wavelength range, percentage change in transmittance, and response shift index.

[0062] The degree of degradation of the packaging material is determined based on the spectral response shift data, and the degradation data of the packaging material is obtained.

[0063] In this embodiment, the degradation degree of the encapsulation material is assessed using a weighted scoring method, which integrates the spectral response shift index, material yellowing parameters, and thermo-oxidative aging index. The specific scoring rules are set as follows: spectral response shift index weight 0.5, yellowing ΔE value weight 0.3, and thermo-oxidative aging index weight 0.2. The scoring range is defined as 0 to 100 points, with higher scores indicating more severe degradation. Levels are categorized based on the score range: 0-20 for slight degradation, 21-50 for moderate degradation, and 51-100 for severe degradation. After standardizing each indicator for each sample, the data is input into the scoring system to calculate the overall degradation score. The final output is an encapsulation material degradation data table, including sample number, overall degradation score, specific indicator weight values, level classification, and corresponding recommended maintenance measure codes. Data storage formats support CSV and database entry for convenient subsequent system access.

[0064] Preferably, step S3, which involves analyzing the transmittance anomaly based on the degradation data of the encapsulation material, includes: Spectral transmittance was extracted based on data on the degradation of packaging materials. In this embodiment, the operation of extracting spectral transmittance based on encapsulation material degradation data first involves obtaining material fragments from samples within the identified degradation area of ​​the encapsulation material. A UV-Vis spectrophotometer is used to measure the transmittance of the samples, with the measurement wavelength range limited to 300 nm to 1100 nm. The measurement steps include placing the sample in the optical path, the spectrophotometer acquiring transmitted light intensity data corresponding to different wavelengths, and then correcting for background light and instrument errors using the instrument's built-in calibration function. Transmittance is calculated sequentially by wavelength, with a value range of 0 to 100%. A transmittance curve is plotted with wavelength on the x-axis and transmittance percentage on the y-axis. All measurement data are saved in a data file for subsequent processing. The measurement environment is controlled at a temperature of 23°C ± 1°C and a relative humidity maintained between 45% and 55% to ensure data stability.

[0065] Band attenuation difference analysis is performed based on spectral transmittance to obtain spectral degradation data; In this embodiment, the specific implementation steps for band attenuation difference analysis based on spectral transmittance include dividing the acquired transmittance data into predefined band regions, typically divided into the ultraviolet region (300-400 nm), the visible light region (400-700 nm), and the near-infrared region (700-1100 nm). The average transmittance is calculated for each band. Using the standard transmittance curve of the undegraded material as a benchmark, the average transmittance measured in each band is compared to calculate the transmittance attenuation percentage. The attenuation percentage is obtained using the formula: Attenuation Percentage = (Benchmark Transmittance - Actual Transmittance) / Benchmark Transmittance × 100%. A transmittance attenuation threshold of 5% is set for each band. Bands exceeding this threshold are marked as having significant attenuation. The band attenuation difference analysis results are presented in a data table organized by band, including the band range, benchmark transmittance, actual transmittance, and attenuation percentage.

[0066] Calculate the band energy loss based on spectral degradation data; In this embodiment, the implementation details of calculating band energy loss based on spectral degradation data include using standard solar spectral radiation data as a reference energy input. The solar spectral energy density data covers the wavelength range of 300 nm to 1100 nm. Combined with the band transmittance attenuation percentage, the corresponding band energy loss is calculated. The energy loss calculation formula is the band energy density multiplied by the transmittance attenuation percentage, with the unit being watts per square meter. The specific value is based on the daily average solar radiation energy, such as a standard solar radiation intensity of 1000 watts per square meter. The energy density is allocated according to the band division and allocation ratio. The energy loss data is summarized in the band energy loss table, which lists the band, energy density, transmittance attenuation percentage, and the calculated energy loss value. The environmental parameters during measurement are fixed at 23 degrees Celsius ± 2 degrees Celsius and relative humidity of 45% to 55%.

[0067] Transmittance anomaly analysis was performed based on band energy loss to obtain transmittance anomaly data.

[0068] In this embodiment, the specific steps for transmittance anomaly analysis based on band energy loss are as follows: The calculated band energy loss is compared with an industry standard threshold. The standard threshold is set at 20 watts per square meter, indicating an abnormal state. Combining the energy loss across multiple bands, the transmittance anomaly level of the encapsulation material is comprehensively assessed. During the analysis, anomaly values ​​for each band are cumulatively scored according to their weights. The weights are determined based on the influence of each band on the photoelectric conversion efficiency of the photovoltaic module; for example, the visible light region has the highest weight, followed by near-infrared, and the ultraviolet light region has the lowest. The final calculated comprehensive score is used as the transmittance anomaly data. All relevant data, including band energy loss, weight coefficients, and anomaly scores, are stored in the transmittance anomaly database to ensure data integrity and traceability. During the transmittance anomaly analysis, the ambient temperature is maintained at 23°C ± 2°C, and the humidity at 50% ± 5%, to avoid environmental factors affecting the measurement results.

[0069] Preferably, this specification also provides a photovoltaic power plant model determination system for executing the production parameter optimization method for lightweight glass bottles as described above. The photovoltaic power plant model determination system includes: The power generation prediction module is used to acquire photovoltaic power plant information; conduct meteorological resource assessment based on the photovoltaic power plant information to obtain meteorological resource data; and predict power generation based on the meteorological resource data to obtain power generation data. The module aging assessment module is used to analyze the energy loss of the module based on the power generation data to obtain energy loss data; to detect the degradation of power generation performance based on the energy loss data to obtain power generation performance degradation data; and to assess the aging of the module based on the power generation performance degradation data to obtain module aging data. The transmittance anomaly analysis module is used to predict and correct the power generation capacity of the power generation data based on the component aging data to obtain corrected power generation data; to perform encapsulation material degradation analysis based on the component aging data to obtain encapsulation material degradation data; and to perform transmittance anomaly analysis based on the encapsulation material degradation data to obtain transmittance anomaly data. The photoelectric conversion efficiency determination module is used to detect the light absorption capacity of the module based on the abnormal light transmittance data, and obtain the light absorption capacity data of the module; based on the light absorption capacity data of the module and the corrected power generation data, the photoelectric conversion efficiency of the photovoltaic power station model is determined, and the photoelectric conversion efficiency data is obtained.

[0070] Optionally, this specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements any one of the photovoltaic power plant model determination methods.

[0071] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0072] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for determining a photovoltaic power plant model, characterized in that, Includes the following steps: Step S1: Obtain photovoltaic power station information; conduct meteorological resource assessment based on photovoltaic power station information to obtain meteorological resource data; predict power generation based on meteorological resource data to obtain power generation data; Step S2: Analyze component energy loss based on power generation data to obtain energy loss data; detect power generation performance degradation based on energy loss data to obtain power generation performance degradation data. Component aging assessment is performed based on power generation performance degradation data to obtain component aging data; Step S3: Based on the component aging data, perform power generation capacity prediction and correction on the power generation data to obtain corrected power generation data; perform encapsulation material degradation analysis based on the component aging data to obtain encapsulation material degradation data; perform light transmittance anomaly analysis based on the encapsulation material degradation data to obtain light transmittance anomaly data. Step S4: Detect the light absorption capacity of the module based on the abnormal light transmittance data to obtain the light absorption capacity data of the module; determine the photoelectric conversion efficiency of the photovoltaic power station model based on the light absorption capacity data of the module and the corrected power generation data to obtain the photoelectric conversion efficiency data.

2. The method for determining a photovoltaic power station model according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain photovoltaic power station information and determine the location of the photovoltaic power station to obtain location data; Step S12: Match meteorological observation stations based on location data to obtain observation station data; Step S13: Extract historical meteorological records based on observation station data; calculate the total solar radiation based on historical meteorological records; Step S14: Perform seasonal illumination fluctuation analysis based on total solar radiation to obtain seasonal fluctuation data; Step S15: Calculate the effective sunshine duration based on seasonal fluctuation data to obtain effective sunshine duration data; conduct an irradiance suitability assessment based on the effective sunshine duration data to obtain meteorological resource data; Step S16: Predict power generation based on meteorological resource data to obtain power generation data.

3. The method for determining a photovoltaic power station model according to claim 2, characterized in that, Step S16 includes the following steps: Step S161: Obtain the photovoltaic power station model and identify the radiation receiving structure; Step S162: Perform radiation irradiation simulation on the receiving radiation structure based on meteorological resource data, where the direct radiation is set to 300-950 W / m. 2 To obtain radiation exposure data; Step S163: Perform power generation conversion calculation based on radiation irradiation data to obtain power generation data.

4. The method for determining a photovoltaic power station model according to claim 2, characterized in that, Step S2 includes the following steps: Step S21: Perform component loss analysis based on power generation data to obtain component loss data; Step S22: Analyze inverter losses based on power generation data to obtain inverter loss data; Step S23: Detect power generation performance degradation based on energy loss data to obtain power generation performance degradation data; Step S24: Collect ultraviolet irradiation data based on power generation performance degradation data; perform photo-oxidation reaction analysis on the material based on the ultraviolet irradiation data to obtain photo-oxidation reaction data; perform chain breakage detection on the material based on the photo-oxidation reaction data to obtain chain breakage data. Step S25: Perform component aging assessment based on material chain breakage data to obtain component aging data.

5. The method for determining a photovoltaic power station model according to claim 4, characterized in that, Step S22 includes the following steps: Step S221: Collect inverter heat data based on power generation data to obtain inverter heat data; Step S222: Perform conduction loss analysis based on inverter heat data to obtain conduction loss data; Step S223: Perform temperature rise statistics based on conduction loss data, wherein the temperature rise response time is set to 2-20 min, and the temperature rise response data is obtained; Step S224: Perform thermal cycling stress detection on the solder joint based on the temperature rise response data to obtain the thermal stress of the solder joint; Step S225: Perform insulation failure detection based on the thermal stress of the solder joint to obtain insulation failure data; determine the degree of inverter loss based on the insulation failure data to obtain inverter loss data.

6. The method for determining a photovoltaic power plant model according to claim 1, characterized in that, Step S3, which involves revising the power generation data based on component aging data to predict power generation capacity, includes: Identify and output performance degradation factors based on component aging data; Output power generation attenuation curve based on performance degradation factor; Calculate the power generation attenuation factor based on the power generation attenuation curve; Performance ratio fitting correction is performed based on the power generation attenuation factor to obtain performance ratio correction data; Based on the performance ratio correction data, the power generation data is corrected by predicting and correcting the power generation capacity to obtain the corrected power generation data.

7. The method for determining a photovoltaic power station model according to claim 1, characterized in that, Step S3, which involves analyzing the degradation of encapsulation materials based on component aging data, includes: Identify aging areas of components based on component aging data; Thermo-oxidative aging test of materials is performed based on the aging area of ​​the component to obtain thermo-oxidative aging data; Yellowing analysis was performed based on thermo-oxidative aging data to obtain yellowing data for the material; Spectral response shift detection was performed based on material yellowing data to obtain spectral response shift data; The degree of degradation of the packaging material is determined based on the spectral response shift data, and the degradation data of the packaging material is obtained.

8. The method for determining a photovoltaic power station model according to claim 1, characterized in that, Step S3, which involves analyzing the transmittance anomalies based on the degradation data of the encapsulation material, includes: Spectral transmittance was extracted based on data on the degradation of packaging materials. Band attenuation difference analysis is performed based on spectral transmittance to obtain spectral degradation data; Calculate the band energy loss based on spectral degradation data; Transmittance anomaly analysis was performed based on band energy loss to obtain transmittance anomaly data.

9. A photovoltaic power plant model determination system, characterized in that, The photovoltaic power plant model determination system is used to perform the photovoltaic power plant model determination method as described in claim 1, and includes: The power generation prediction module is used to acquire photovoltaic power plant information; conduct meteorological resource assessment based on the photovoltaic power plant information to obtain meteorological resource data; and predict power generation based on the meteorological resource data to obtain power generation data. The module aging assessment module is used to analyze the energy loss of the module based on the power generation data to obtain energy loss data; to detect the degradation of power generation performance based on the energy loss data to obtain power generation performance degradation data; and to assess the aging of the module based on the power generation performance degradation data to obtain module aging data. The transmittance anomaly analysis module is used to predict and correct the power generation capacity of the power generation data based on the component aging data to obtain corrected power generation data; to perform encapsulation material degradation analysis based on the component aging data to obtain encapsulation material degradation data; and to perform transmittance anomaly analysis based on the encapsulation material degradation data to obtain transmittance anomaly data. The photoelectric conversion efficiency determination module is used to detect the light absorption capacity of the module based on the abnormal light transmittance data, and obtain the light absorption capacity data of the module; based on the light absorption capacity data of the module and the corrected power generation data, the photoelectric conversion efficiency of the photovoltaic power station model is determined, and the photoelectric conversion efficiency data is obtained.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the photovoltaic power plant model determination method as described in any one of claims 1 to 9.

Images