Early warning method for influence of extreme high temperature on photovoltaic power generation efficiency based on meteorological numerical model

Abstract

The invention relates to an early warning method for the influence of extreme high temperature on photovoltaic power generation efficiency based on a meteorological numerical model, and the method comprises the steps: building a high-temperature feature database based on the background of a photovoltaic equipment deployment region and the hierarchical sampling of historical meteorological data, and determining a high-temperature evolution path, the method comprises the following steps: dividing regions based on an evolution path, collecting temperature parameters, constructing a dynamic distribution model, generating high-temperature event intensity characteristics and an influence range, marking a risk level, obtaining equipment temperature and performance data for a high-risk region, and constructing an influence evaluation matrix in combination with historical records; based on the evaluation matrix, an early warning report is generated through multi-dimensional index analysis and visualized, and effective early warning on the influence of the extreme high temperature on the photovoltaic power generation efficiency is achieved.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic power generation thermal environment effect monitoring technology, and in particular relates to an early warning method for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models. Background Technology

[0002] With the rapid development of new energy technologies, photovoltaic (PV) power generation has become an important component of the energy sector due to its clean and renewable characteristics. As PV installed capacity continues to increase, the impact of extreme high temperatures on the power generation efficiency of PV equipment is becoming increasingly significant. Therefore, early warning technologies targeting the impact of high temperatures are gradually gaining attention. This technology aims to predict the degree of impact of high-temperature weather on PV power generation efficiency through monitoring and assessment. Traditional technologies often rely on simple statistical analysis of historical temperature data and power generation efficiency, or determine the risk level using a single temperature threshold. This method typically ignores the spatiotemporal dynamics of the temperature field and fails to combine the actual operating parameters and physical characteristics of PV equipment for refined evaluation. Current early warning methods have significant shortcomings: firstly, the lack of dynamic simulation of the evolution path of high-temperature weather makes it impossible to accurately capture the spatial distribution and intensity variation patterns of the temperature field, resulting in low spatiotemporal accuracy of the early warning; secondly, the absence of a quantitative correlation model between high-temperature events and equipment performance degradation makes it difficult to scientifically assess the specific impact of high temperatures of different intensities on power generation efficiency, thus preventing the early warning results from providing accurate support for the operation and maintenance decisions of PV systems. Summary of the Invention

[0003] Based on this, a method for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models is provided, which can solve the above problems.

[0004] Firstly, this application provides a method for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models, including:

[0005] Based on the regional background characteristics and historical meteorological data of the photovoltaic equipment deployment area, the temperature field data is sampled in layers to construct a high temperature characteristic database, and the evolution path of high temperature weather is determined by combining time series changes.

[0006] Based on the evolution path, the photovoltaic equipment deployment area is divided into multiple sub-regions, temperature parameters of the sub-regions are collected, and a dynamic distribution model of high temperature weather is constructed by combining the temperature parameters with the evolution path.

[0007] Based on the dynamic distribution model of high temperature weather, the event intensity characteristics and impact range of high temperature weather are generated, and the spatiotemporal changes of intensity characteristics are compared with preset temperature thresholds. Sub-regions are marked as low-risk, medium-risk and high-risk levels respectively.

[0008] For high-risk sub-regions, temperature monitoring data and performance output data of photovoltaic equipment are obtained. Combined with the historical operation records of the equipment, the specific impact of high temperature weather on equipment performance is determined, and an impact assessment matrix is ​​constructed.

[0009] Based on the impact assessment matrix, a multi-dimensional index analysis method is used to generate an early warning report on the impact of power generation efficiency according to the correlation characteristics between event intensity and impact range, and the report is visualized using spatial visualization technology.

[0010] In one embodiment, a high-temperature characteristic database is constructed, and the evolution path of high-temperature weather is determined by combining time series changes, including:

[0011] Based on regional background characteristics and geographical features, climate types and historical temperature data in historical meteorological data, the temperature field data is preprocessed in time and space to obtain standardized temperature data.

[0012] Based on geographical features, standardized temperature field data is sampled in layers, and spatial distribution maps of temperature are generated by interpolation.

[0013] Frequency domain analysis was performed on the time series components of historical temperature data, wavelet transform was used to extract the periodic characteristics of high-temperature weather, and a high-temperature characteristic database was constructed by combining the spatial distribution map of temperature.

[0014] Based on the spatiotemporal trajectories and high-temperature characteristic databases of historical high-temperature events, statistical analysis is used to determine the movement path, intensity changes, and duration evolution patterns of high-temperature weather, thereby generating the evolution path of high-temperature weather.

[0015] In one embodiment, based on a dynamic distribution model of high-temperature weather, the event intensity characteristics and impact range of high-temperature weather are generated, including:

[0016] Based on the dynamic distribution model of high temperature weather, dynamic temperature parameters of each sub-region are extracted, the proportion of spatial coverage area exceeding the preset temperature threshold is calculated, and the scope of influence is generated.

[0017] Based on the time change rate of dynamic temperature parameters, calculate the temperature fluctuation amplitude and the cumulative duration of continuous temperature exceeding the preset threshold, and generate an intensity cumulative gradient.

[0018] By using a preset weighting function, correlation analysis is performed on the spatial coverage area and intensity cumulative gradient to generate event intensity features.

[0019] In one embodiment, for high-risk sub-regions, temperature monitoring data and performance output data of photovoltaic equipment are acquired. Combined with historical operating records of the equipment, the specific impact of high temperatures on equipment performance is determined, and an impact assessment matrix is ​​constructed, including:

[0020] For high-risk sub-regions, acquire temperature monitoring data and performance output data of photovoltaic equipment deployed in the region during historical high-temperature events. The temperature monitoring data and performance output data include module surface temperature, backsheet temperature, ambient temperature, DC output power, AC output power, and conversion efficiency.

[0021] Based on the physical and installation parameters of photovoltaic equipment, combined with temperature monitoring data and performance output data, a temperature response model of photovoltaic modules is established to calculate the theoretical temperature response coefficient of photovoltaic equipment under current and predicted high temperature conditions.

[0022] By associating with historical operation records, the actual power generation efficiency degradation rate of photovoltaic equipment during the corresponding historical high-temperature period can be extracted.

[0023] The theoretical temperature response coefficient is compared and analyzed with the actual power generation efficiency decay rate to calculate the performance loss deviation caused by high temperature.

[0024] An impact assessment matrix is ​​constructed with theoretical temperature response coefficient, actual power generation efficiency attenuation rate, and performance loss deviation as key dimensions. The elements of the impact assessment matrix are associated with specific equipment models, installation locations, high-temperature event intensity characteristics, and impact ranges.

[0025] In one embodiment, based on the impact assessment matrix, a multi-dimensional index analysis method is used to generate a power generation efficiency impact early warning report according to the correlation characteristics between event intensity and impact range, including:

[0026] Principal component analysis was performed on the theoretical temperature response coefficient, actual power generation efficiency attenuation rate, and performance loss deviation in the impact assessment matrix to extract the core principal component indicators characterizing the sensitivity to high temperature impact.

[0027] Establish a correlation coefficient matrix between core principal component indicators and event intensity characteristics and scope of impact;

[0028] Based on the correlation coefficient matrix, a comprehensive impact index on power generation efficiency for each high-risk sub-region is generated through weighted fusion calculation.

[0029] Based on the comprehensive impact index of power generation efficiency, and according to the preset threshold range of the comprehensive impact index, different levels of early warning for power generation efficiency decline are divided.

[0030] Based on the predicted evolution path, impact range, and duration of high-temperature weather, a power generation efficiency impact warning report is generated. The warning report includes the warning level, the expected impact time window, the affected geographical area, the expected maximum reduction in power generation efficiency, and recommended measures.

[0031] In one embodiment, the comprehensive impact index of power generation efficiency is calculated using the following formula:

[0032]

[0033] Where δ=|η actual -R theory ·T excess | represents the performance loss deviation, η actual R represents the actual power generation efficiency degradation rate. theory T is the theoretical temperature response coefficient. excess S represents the cumulative duration of exceeding the preset temperature threshold. event As an event intensity characteristic, it is determined by the sum of the temperature fluctuation amplitude and the cumulative duration T of exceeding a preset temperature threshold. excess Gradient integral generation, A impact For the scope of influence, α i β i and γ i Principal component loading factor, through (δ,S event A impact Principal component analysis was performed to determine whether the following conditions were met. w i As a dynamic weighting factor, based on the event intensity feature S event Scope of influence A impact The spatiotemporal correlation coefficient allocation, and satisfying k is the number of principal components to be retained, determined by a cumulative variance contribution rate of ≥85%.

[0034] In one embodiment, after generating a power generation efficiency impact warning report, the process further includes visualizing the warning according to the following steps:

[0035] By overlaying the warning level, the affected geographical range, and the regional map on the geographic information system platform, a dynamic heat map is generated; among them, the heat map color scale is associated with the comprehensive impact index of power generation efficiency, and high-risk areas are rendered with preset warning color marks.

[0036] Secondly, this application also provides an early warning device for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models, comprising:

[0037] The temperature path modeling module is used to perform layered sampling of temperature field data based on the regional background characteristics and historical meteorological data of the photovoltaic equipment deployment area, construct a high temperature characteristic database, and determine the evolution path of high temperature weather by combining time series changes.

[0038] The dynamic distribution modeling module is used to divide the photovoltaic equipment deployment area into multiple sub-regions according to the evolution path, collect the temperature parameters of the sub-regions, and combine the temperature parameters with the evolution path to build a dynamic distribution model of high temperature weather.

[0039] The risk level classification module is used to generate the event intensity characteristics and impact range of high temperature weather based on the dynamic distribution model of high temperature weather, and compare the spatiotemporal changes of intensity characteristics with preset temperature thresholds to mark sub-regions as low-risk, medium-risk and high-risk levels respectively.

[0040] The impact assessment matrix module is used to acquire temperature monitoring data and performance output data of photovoltaic equipment for high-risk sub-regions, and combine them with the equipment's historical operating records to determine the specific impact of high-temperature weather on equipment performance and construct an impact assessment matrix.

[0041] The early warning report generation module is used to generate early warning reports on the impact on power generation efficiency based on the impact assessment matrix and using multi-dimensional index analysis methods, according to the correlation characteristics between event intensity and impact range.

[0042] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-mentioned method for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models.

[0043] Fourthly, this application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described method for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models.

[0044] The aforementioned method, device, computer equipment, and storage medium for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models, perform stratified sampling of temperature field data based on regional background characteristics and historical meteorological data of the photovoltaic equipment deployment area to construct a high-temperature characteristic database. It determines the evolution path of high-temperature weather by combining time series changes, divides the region into sub-regions, and constructs a dynamic distribution model of high-temperature weather by combining temperature parameters and evolution paths. After generating event intensity characteristics and impact range, it compares them with preset temperature thresholds to mark the risk level. For high-risk areas, it acquires temperature monitoring data and performance output data of photovoltaic equipment and constructs an impact assessment matrix by combining historical operation records. It uses a multi-dimensional index analysis method to generate and visualize power generation efficiency impact early warning reports based on the correlation characteristics between event intensity and impact range. Through refined analysis of the spatiotemporal dynamic characteristics of the temperature field and quantitative correlation modeling between high-temperature events and equipment performance degradation, it solves the problems of low spatiotemporal accuracy of early warnings and difficulty in scientifically assessing the degree of impact caused by insufficient dynamic simulation of high-temperature evolution paths in traditional technologies. This provides accurate power generation efficiency impact early warning support for photovoltaic system operation and maintenance decisions. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart of the early warning method for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models according to the present invention.

[0047] Figure 2 This is a structural diagram of the early warning device for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models, according to the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] In one embodiment, such as Figure 1 As shown, a method for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models is provided. This embodiment takes the application of this method to a system including terminals and servers as an example. The hardware architecture of the system includes sensors deployed in the photovoltaic equipment area (collecting air temperature parameters, equipment temperature and performance data in real time), a server (storing historical meteorological data and equipment operation records, performing high temperature characteristic modeling, risk level classification and impact assessment matrix construction), and terminal devices (integrating a geographic information system, receiving early warning reports and performing spatial visualization). In the application scenario, when the photovoltaic area faces extreme high temperatures, the sensors transmit data to the server for model calculation. After the server generates an early warning report, it sends it to the terminal for dynamic thermal map display, realizing the collaborative interaction of data collection, analysis, and visualization. It can be understood that this method can also be applied to a system where the server independently performs data processing and early warning generation, or it can be applied to a system where the terminal and server collaborate, completing the entire process from high temperature characteristic analysis to early warning visualization through data interaction between the two. In this embodiment, the method includes the following steps:

[0050] S01. Based on the regional background characteristics and historical meteorological data of the photovoltaic equipment deployment area, the temperature field data is sampled in layers to construct a high temperature characteristic database, and the evolution path of high temperature weather is determined by combining time series changes.

[0051] Based on the regional background characteristics of the photovoltaic equipment deployment area (including geographical features such as topography, altitude distribution, land cover type, and climate type such as arid or humid climate characteristics) and historical meteorological data (such as long-term recorded temperature data, humidity, and solar radiation parameters), spatiotemporal preprocessing of temperature field data is performed, including data cleaning, outlier removal, and normalization operations, to generate standardized temperature data. The standardized temperature field data is then stratified and sampled according to geographical features, allowing the region to be divided into different levels based on topographic gradients or climate zones. Spatial interpolation methods (such as Kriging interpolation or inverse distance weighting) are used to generate a spatial distribution map of temperature, reflecting the spatial distribution characteristics of the temperature field. The time series components of historical temperature data are further processed... Frequency domain analysis is used to extract the periodic characteristics of high-temperature weather (such as seasonal fluctuations or interannual variations) using wavelet transform algorithms. These periodic characteristics are then fused with spatial temperature distribution maps to construct a high-temperature feature database (containing spatial distribution, time series features, and key indicators of high-temperature events). Combining the spatiotemporal trajectories of historical high-temperature events (such as the movement path of high-temperature centers), statistical analysis methods (such as regression analysis or clustering algorithms) are used to determine the evolution path of high-temperature weather, including the direction of movement, intensity change gradient, and duration evolution patterns, thereby achieving a quantitative description of the dynamic characteristics of high-temperature weather. Through hierarchical sampling and feature fusion, the accuracy and predictability of the high-temperature evolution path are ensured, solving the problem of neglecting spatiotemporal dynamic changes in traditional methods.

[0052] S02. Based on the evolution path, the photovoltaic equipment deployment area is divided into multiple sub-regions. Temperature parameters of the sub-regions are collected, and a dynamic distribution model of high-temperature weather is constructed by combining the temperature parameters with the evolution path.

[0053] Based on the established evolution path of high-temperature weather, the photovoltaic equipment deployment area is divided into multiple spatially continuous sub-regions according to the homogeneity of meteorological characteristics and risk distribution characteristics. Temperature parameters (including dynamic indicators such as temperature values ​​and time-varying rates) of each sub-region can be collected in real time through meteorological monitoring station networks, satellite remote sensing data, and numerical model outputs. These temperature parameters are then coupled with the spatiotemporal characteristics of the high-temperature evolution path for analysis. A spatiotemporal sequence modeling technique, combined with Gaussian process regression or dynamic data assimilation algorithms, is used to construct a dynamic distribution model of high-temperature weather: integrating the spatiotemporal covariance structure of temperature parameters in the sub-regions to generate a continuous spatial distribution of the temperature field; introducing the movement vector of the evolution path as a constraint to establish a transition matrix for the temperature field's evolution over time; and using a Kalman filter algorithm to assimilate new observation data in real time and dynamically update the model parameters. This addresses the technical deficiency of traditional static models in capturing the dynamic spatiotemporal evolution of the temperature field, providing high-precision input for subsequent risk classification.

[0054] S03, based on the dynamic distribution model of high temperature weather, generates the event intensity characteristics and impact range of high temperature weather, and compares the spatiotemporal changes of intensity characteristics with preset temperature thresholds, and marks the sub-regions as low-risk, medium-risk and high-risk levels respectively.

[0055] Based on a dynamic distribution model of high-temperature weather, dynamic temperature parameters (including temperature values, time-varying rates of change, and other continuous changes) of each sub-region are extracted. Spatial analysis algorithms are used to calculate the proportion of spatial coverage exceeding a preset temperature threshold (set according to the thermal tolerance critical value of photovoltaic equipment materials), generating the influence range. Based on the time-varying rate of change of the dynamic temperature parameters, the amplitude of temperature fluctuations and the cumulative duration of continuous exceedances are calculated. An intensity cumulative gradient is generated using a gradient integral algorithm, and a preset weighting function is used to perform correlation analysis on the above parameters to generate event intensity characteristics. The spatiotemporal variation matrix of the intensity characteristics is compared grid-by-grid with the preset temperature threshold. A three-level risk labeling mechanism is adopted: when the event intensity characteristic ≤ 0.3 and the influence range ≤ 15%, it is labeled as low risk (power generation efficiency reduction < 5%); when the event intensity characteristic ≤ 0.7 and the influence range ≤ 40%, it is labeled as medium risk (power generation efficiency reduction 5-15%); and when the event intensity characteristic > 0.7 and the influence range > 40%, it is labeled as high risk (power generation efficiency reduction > 15%). Spatial topology analysis is used to delineate the boundaries of sub-regions.

[0056] S04. For high-risk sub-regions, acquire temperature monitoring data and performance output data of photovoltaic equipment, combine with historical operating records of the equipment, determine the specific impact of high temperature weather on equipment performance, and construct an impact assessment matrix.

[0057] Specifically, for sub-regions marked as high-risk, real-time monitoring data of photovoltaic (PV) equipment deployed in these areas (including temperature monitoring data such as module surface temperature, backsheet temperature, and ambient temperature, as well as performance output data such as DC output power, AC output power, and conversion efficiency) are acquired and combined with similar parameters stored in the equipment's historical operation records to form a spatiotemporal correlation dataset. Based on the physical parameters of PV modules (including material thermal expansion coefficient and specific heat capacity) and installation parameters (including tilt angle and azimuth angle), a temperature response model of PV modules is established through the heat conduction equation to calculate the theoretical temperature response coefficient (representing the efficiency degradation rate caused by a unit temperature rise). Simultaneously, by associating with historical operation records, the actual power generation efficiency degradation rate under the same high-temperature intensity characteristics and influence range is extracted. The performance loss deviation is calculated through a deviation analysis algorithm to reflect the specific impact of high-temperature weather on equipment performance. A three-dimensional impact assessment matrix is ​​constructed, with row vectors corresponding to specific equipment units and column vectors mapping the theoretical temperature response coefficient, actual power generation efficiency degradation rate, and performance loss deviation, respectively. Extended fields are used to associate with equipment model, installation location coordinates, and characteristic parameters of the current high-temperature event. By constructing this matrix, a dual verification mechanism of physical model and measured data is realized; the deviation serves as a quantitative indicator of the degree of equipment performance degradation; the spatiotemporal correlation accuracy reaches the equipment level, solving the technical deficiency of traditional methods in quantifying individual differences of equipment.

[0058] S05, based on the impact assessment matrix, uses a multi-dimensional index analysis method to generate a power generation efficiency impact early warning report based on the correlation characteristics between event intensity and impact range, and uses spatial visualization technology for visualization output.

[0059] Based on the impact assessment matrix, a multidimensional index analysis method can be used to perform principal component analysis on the matrix data to extract core principal component indicators characterizing the sensitivity of high temperature impacts. A correlation coefficient matrix between these core indicators and the event intensity characteristics and impact range can be established. A comprehensive power generation efficiency impact index can be generated through weighted fusion calculation (considering a linear combination of performance loss deviation, event intensity characteristics, and impact range, with weight factors dynamically allocated based on spatiotemporal correlation). According to preset threshold ranges (determined based on the actual power generation efficiency decline, corresponding to mild <5%, moderate 5-15%, and severe >15%), the comprehensive impact index is divided into different warning levels (e.g., green, yellow, and red corresponding to mild, moderate, and severe power generation efficiency declines, respectively). A structured warning report is generated by combining the predicted evolution path, impact range, and duration of high-temperature weather. This report includes the warning level, expected impact time window, affected geographical area, expected maximum power generation efficiency decline, and targeted measures (e.g., cooling system startup suggestions or power dispatch schemes), and is visualized. Through multidimensional index fusion and dynamic rendering technology, the problem of traditional warning results failing to intuitively display the spatial risk gradient distribution is solved.

[0060] The aforementioned early warning method for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models constructs a high-temperature feature database by hierarchically sampling the temperature field based on the background of the photovoltaic equipment deployment area and historical meteorological data. It determines the high-temperature evolution path by combining time series data, divides regions to collect temperature parameters to construct a dynamic distribution model, generates event intensity characteristics and impact ranges, and compares them with preset thresholds to mark risk levels. For high-risk areas, it acquires equipment temperature and performance data and constructs an impact assessment matrix based on historical records. Based on the assessment matrix, it generates and visualizes early warning reports through multi-dimensional index analysis. By analyzing the spatiotemporal dynamic characteristics of the temperature field, it extracts the spatiotemporal evolution law of high temperatures using hierarchical sampling and wavelet transform, solving the problem of low spatiotemporal accuracy in early warning caused by insufficient dynamic simulation of high-temperature evolution paths in traditional technologies. By establishing a photovoltaic module temperature response model and impact assessment matrix, and combining principal component analysis and weighted fusion to calculate the comprehensive impact index of power generation efficiency, it achieves quantitative correlation modeling between high-temperature events and equipment performance degradation, scientifically assesses the specific impact of high temperatures of different intensities on power generation efficiency, and provides accurate early warning support for photovoltaic system operation and maintenance decisions. This effectively solves the shortcomings of traditional methods in scientifically assessing the impact of extreme high temperatures on photovoltaic power generation efficiency.

[0061] In one embodiment, a high-temperature characteristic database is constructed, and the evolution path of high-temperature weather is determined by combining time series changes, including:

[0062] S11. Based on regional background characteristics and geographical features, climate types and historical temperature data in historical meteorological data, the temperature field data is preprocessed in time and space to obtain standardized temperature data.

[0063] S12, based on geographical features, performs stratified sampling of standardized temperature field data and generates a spatial distribution map of temperature through interpolation;

[0064] S13. Frequency domain analysis is performed on the time series components of historical temperature data. Wavelet transform is used to extract the periodic characteristics of high-temperature weather, and a high-temperature characteristic database is constructed by combining the spatial distribution map of temperature.

[0065] S14, based on the spatiotemporal trajectory and high temperature characteristic database of historical high temperature events, determines the movement path, intensity change and duration evolution of high temperature weather through statistical analysis, and generates the evolution path of high temperature weather.

[0066] Specifically, regional background characteristics can include geographical features (such as topography, altitude, and land cover type), climate type (arid / humid climate, monsoon characteristics, etc.), and historical meteorological data can include hourly / daily temperature data for more than 3 years, humidity, solar radiation, and other auxiliary parameters. Geographic Information Systems (GIS) can be used to match discrete meteorological station data to a unified spatial coordinate system (such as UTM projection) to eliminate geographical location bias. By removing outliers (such as temperature data exceeding physically reasonable ranges), filling missing values ​​(using interpolation from nearby stations or time series prediction), and normalizing the data (such as Z-score standardization), standardized temperature data is obtained, making temperature data from different regions and times comparable. A stratified sampling strategy can be based on geographical stratification, dividing sampling layers according to altitude (such as 0-100m, 100-500m, etc.), topography type (plains, mountains, valleys), or climate zone (such as temperate, subtropical). The sampling method is to collect standardized temperature data within each layer in a grid (such as 1km × 1km) to ensure that the sample covers typical geographical units. Spatial interpolation algorithms (Kriging or Inverse Distance Weighted (IDW)) can be used, considering factors such as altitude and terrain slope as weighting factors. A rasterized spatial distribution map of temperature can be generated, visually demonstrating temperature differences in mountainous and plain areas (e.g., temperature decreases with increasing altitude in mountainous regions). Historical temperature data (e.g., daily maximum temperature) can be decomposed into trend, periodic, and random components. Meyer wavelets or Daubechies wavelets can be used to perform multi-scale analysis of the time series, extracting periodic features at different time scales (e.g., annual, seasonal, and monthly cycles). For example, the seasonal pattern of frequent high temperatures in July and August can be identified, and a high-temperature feature database can be constructed. Its data dimensions include: spatial dimension (mean, extreme values, and gradient distribution of temperature in each stratified region); temporal dimension (frequency, duration, and periodic parameters of high-temperature events); and correlation dimension (correlation indicators between high temperatures and geographical features (e.g., topographic barriers) and climate phenomena (e.g., subtropical high pressure). The storage structure uses a spatiotemporal database (e.g., PostGIS), supporting rapid querying of high-temperature features by region and time. Historical high-temperature events are defined as weather processes where the daily maximum temperature exceeds the local climatic mean + 2σ (σ being the standard deviation) for one or more consecutive days. Temperature field grid data output from meteorological numerical models can be used to track the daily movement of the high-temperature center, forming a spatiotemporal trajectory chain. The movement speed and directional preference of the high-temperature center (e.g., moving from inland to coastal areas) can be calculated, and the blocking effect (e.g., the obstruction of high-temperature movement by mountains) can be analyzed in conjunction with topographic features. Regression equations can be established between high-temperature intensity (e.g., peak temperature) and movement distance and duration; for example, for every 10 km of movement, the peak temperature decreases by 1.5℃.Survival analysis was used to determine the average duration of high-temperature events in different geographical regions (e.g., 7 days in plains and 3 days in mountains). Based on this data, high-temperature paths were plotted in the form of vector lines, and intensity change gradients (e.g., arrow thickness represents temperature peaks) and duration contour lines were marked to obtain the evolution path of high-temperature weather.

[0067] In one embodiment, based on a dynamic distribution model of high-temperature weather, the event intensity characteristics and impact range of high-temperature weather are generated, including:

[0068] S21, based on the dynamic distribution model of high temperature weather, extracts the dynamic temperature parameters of each sub-region, calculates the proportion of spatial coverage area exceeding the preset temperature threshold, and generates the scope of influence.

[0069] S22, based on the time change rate of dynamic temperature parameters, calculate the temperature fluctuation amplitude and the cumulative duration of continuous temperature exceeding the preset temperature threshold, and generate an intensity cumulative gradient;

[0070] S23 uses a preset weighting function to perform correlation analysis on the spatial coverage area and intensity cumulative gradient to generate event intensity features.

[0071] For example, dynamic temperature parameters (including real-time temperature values ​​and their temporal rate of change) of each sub-region are extracted. A spatial analysis algorithm is used to calculate the proportion of spatial coverage area exceeding a preset temperature threshold (set based on the thermal tolerance critical value of photovoltaic equipment materials), generating geographical parameters characterizing the spatial extent of high-temperature impact. Based on the temporal rate of change of the dynamic temperature parameters, the temperature fluctuation amplitude (the range of temperature changes per unit time) and the cumulative duration of continuous temperature exceeding the preset temperature threshold are calculated. A gradient integral algorithm is used to generate an intensity cumulative gradient (reflecting the cumulative effect of high temperature over time). A preset spatiotemporal correlation weighting function (linear weighting or nonlinear integral function) is used to perform multidimensional correlation analysis on the spatial coverage area proportion and the intensity cumulative gradient, fusing them to generate a comprehensive event intensity feature characterizing the severity of the high-temperature event. A dynamic distribution model is used to achieve the quantitative conversion of meteorological parameters into risk indicators. The spatial coverage area proportion reflects the breadth of high-temperature impact, the intensity cumulative gradient quantifies the persistence and volatility of high temperatures, and the weighting function (linear weighting or gradient integral, with weights determined using the entropy method) uses preset coefficients to achieve the coupled analysis of spatiotemporal characteristics, forming a composite event intensity feature indicator with clear physical meaning.

[0072] In one embodiment, for high-risk sub-regions, temperature monitoring data and performance output data of photovoltaic equipment are acquired. Combined with historical operating records of the equipment, the specific impact of high temperatures on equipment performance is determined, and an impact assessment matrix is ​​constructed, including:

[0073] S31, for high-risk sub-regions, acquire temperature monitoring data and performance output data of photovoltaic equipment deployed in the region during historical high-temperature events. The temperature monitoring data and performance output data include module surface temperature, backsheet temperature, ambient temperature, DC output power, AC output power, and conversion efficiency.

[0074] S32, based on the physical parameters and installation parameters of the photovoltaic equipment, combined with temperature monitoring data and performance output data, establishes a temperature response model for the photovoltaic module and calculates the theoretical temperature response coefficient of the photovoltaic equipment under current and predicted high temperature conditions;

[0075] S33, associate with historical operation records, and extract the actual power generation efficiency degradation rate of photovoltaic equipment during the corresponding historical high temperature period;

[0076] S34 compares and analyzes the theoretical temperature response coefficient with the actual power generation efficiency decay rate to calculate the performance loss deviation caused by high temperature.

[0077] S35 constructs an impact assessment matrix with theoretical temperature response coefficient, actual power generation efficiency attenuation rate, and performance loss deviation as key dimensions. The elements of the impact assessment matrix are associated with specific equipment models, installation locations, high-temperature event intensity characteristics, and impact ranges.

[0078] For example, for high-risk sub-regions, multi-dimensional monitoring data of photovoltaic equipment deployed in these regions during historical high-temperature events are obtained (including temperature parameters such as module surface temperature, backsheet temperature, and ambient temperature, which can be monitored using infrared thermal imagers; and performance parameters such as DC output power, AC output power, and conversion efficiency). This data forms the original observation basis for the equipment's thermal response and power generation efficiency. Based on the physical parameters of the photovoltaic equipment (including material thermal expansion coefficient and specific heat capacity) and installation parameters (including tilt angle and azimuth angle), a photovoltaic module temperature response model is established using real-time monitoring data (quantifying the mapping relationship between temperature rise and efficiency through the heat conduction equation). The theoretical temperature response coefficient (characterizing the expected efficiency degradation rate caused by a unit temperature rise) is calculated. The historical operating records of the equipment are simultaneously correlated to extract the actual power generation efficiency degradation rate of the equipment during the same historical period under the same high-temperature intensity characteristics. A deviation analysis algorithm is used to compare the theoretical temperature response coefficient with the actual power generation efficiency degradation rate to calculate the performance loss deviation caused by high temperature (calculated using the formula δ=|η actual -R theory ·T excess | where δ is the performance loss deviation, η actual R represents the actual power generation efficiency degradation rate. theory T is the theoretical temperature response coefficient. excessThe cumulative duration exceeding the preset temperature threshold is used to measure the deviation, which reflects the difference between the actual performance degradation of the equipment and the theoretical model. A three-dimensional impact assessment matrix is ​​constructed, with row vectors mapping specific equipment units and column vectors corresponding to the three core dimensions: theoretical temperature response coefficient, actual power generation efficiency degradation rate, and performance loss deviation. By extending the fields to associate equipment model, installation location coordinates, current high-temperature event intensity characteristics, and impact range, a spatiotemporally correlated assessment system is formed. An innovative mechanism integrates physical models and measured data verification. The theoretical temperature response coefficient is derived through the heat conduction equation, and the performance loss deviation is obtained by dynamically comparing model predictions with historical measured values. The impact assessment matrix uses equipment model and installation location as index units to achieve equipment-level quantitative assessment.

[0079] In one embodiment, based on the impact assessment matrix, a multi-dimensional index analysis method is used to generate a power generation efficiency impact early warning report according to the correlation characteristics between event intensity and impact range, including:

[0080] S41, principal component analysis is performed on the theoretical temperature response coefficient, actual power generation efficiency attenuation rate and performance loss deviation in the impact assessment matrix to extract the core principal component indicators characterizing the sensitivity of high temperature impact.

[0081] S42, Establish the correlation coefficient matrix between the core principal component indicators and the event intensity characteristics and scope of influence;

[0082] S43, based on the correlation coefficient matrix, generates a comprehensive impact index on power generation efficiency for each high-risk sub-region through weighted fusion calculation;

[0083] S44, based on the comprehensive impact index of power generation efficiency, divides different levels of power generation efficiency decline warning according to the preset threshold range of the comprehensive impact index;

[0084] S45, combining the predicted evolution path, impact range and duration of high temperature weather, generates a power generation efficiency impact warning report; the warning report includes the warning level, the expected impact time window, the affected geographical range, the expected maximum power generation efficiency reduction and the recommended measures.

[0085] Specifically, principal component analysis is performed on the multi-dimensional indicators (theoretical temperature response coefficient, actual power generation efficiency degradation rate, and performance loss deviation) in the impact assessment matrix. Eigenvectors are extracted through covariance matrix decomposition, and principal components with a cumulative variance contribution rate ≥85% are retained to generate core principal component indicators characterizing the sensitivity to high-temperature impacts. These indicators integrate the common characteristics of equipment physical characteristics and historical performance deviations. A Pearson correlation coefficient matrix is ​​established between the core indicators, the intensity characteristics of high-temperature events (generated by combining temperature fluctuation amplitude and duration exceeding the threshold), and the scope of impact. Based on the correlation coefficient matrix, a comprehensive power generation efficiency impact index is generated through weighted fusion calculation, where weighting factors are dynamically allocated according to spatial distribution correlation and temporal continuity. A preset comprehensive impact index threshold range is established (which can be based on historical degradation rate statistical analysis; for example, mild ≤0.3 corresponds to an efficiency reduction of <5%, moderate ≤0.3). -0.7 corresponds to a 5%-15% reduction, and >0.7 corresponds to a >15% reduction. The system classifies warning levels into green, yellow, and red. It integrates the direction of movement of the high-temperature evolution path, the spatial topological boundaries of the affected area, and the predicted duration to generate a structured warning report. This report may include: warning level identifiers (three-level color coding based on a comprehensive index threshold), time window predictions (calculated based on the path's movement speed to determine the start and end times of the impact), geographical coverage (GIS boundary coordinates and the thermally affected area), maximum efficiency reduction (calculated through historical attenuation and comprehensive index mapping), and proactive protection measures (such as automatically activating the cooling system when the warning level is red, the corresponding activation threshold for the cooling system, and the power grid dispatching scheme). Principal component dimensionality reduction eliminates indicator redundancy, and dynamic weight allocation strengthens the spatiotemporal correlation, addressing the insufficient warning accuracy caused by static thresholds and isolated parameters in traditional methods.

[0086] In one embodiment, S51, the comprehensive impact index of power generation efficiency is calculated using the following formula:

[0087]

[0088] Where δ=|η actual -R theory ·T excess | represents the performance loss deviation, η actual R represents the actual power generation efficiency degradation rate. theory T is the theoretical temperature response coefficient. excess S represents the cumulative duration of exceeding the preset temperature threshold. event As an event intensity characteristic, it is determined by the sum of the temperature fluctuation amplitude and the cumulative duration T of exceeding a preset temperature threshold. excess Gradient integral generation, A impact For the scope of influence, α i β i and γ i Principal component loading factor, through (δ,Sevent A impact Principal component analysis was performed to determine whether the following conditions were met. w i As a dynamic weighting factor, based on the event intensity feature S event Scope of influence A impact The spatiotemporal correlation coefficient allocation, and satisfying k is the number of principal components to be retained, determined by a cumulative variance contribution rate of ≥85%.

[0089] For example, the performance loss deviation δ is determined by |η actual -R theory ·T excess |Calculations reflect the difference between the theoretical model and the measured attenuation, and the event intensity characteristic S. event The temperature fluctuation amplitude and the cumulative duration T exceeding the preset temperature threshold are used to measure the temperature fluctuation amplitude. excess The gradient integral is generated to quantify the time-varying intensity of high temperature and the range of influence A. impact Based on the calculation of the spatial coverage ratio of superheated areas, the geographical impact breadth is characterized by (δ, S) event A impact Principal component analysis was performed to extract loading factors α. i β i and γ i Eliminate multicollinearity among indicators; based on a cumulative variance contribution rate ≥ 85%, truncate the number of principal components k, and divide (δ, S) event A impact Compress to k dimensions and utilize weighting factor w i Strengthen the correlation mapping between the spatiotemporal evolution characteristics of high-temperature events and equipment performance degradation, generate a comprehensive impact index with clear physical meaning, and provide a quantitative basis for early warning classification.

[0090] In one embodiment, after generating a power generation efficiency impact warning report, the process further includes visualizing the warning according to the following steps:

[0091] S61 uses a geographic information system platform to overlay warning levels, affected geographical areas, and regional maps to generate dynamic heat maps; among them, the heat map color scale is associated with the comprehensive impact index of power generation efficiency, and high-risk areas are rendered with preset warning color marks.

[0092] Specifically, the spatial data engine of a Geographic Information System (GIS) platform can be used to overlay the warning levels (including green / yellow / red color indicators) corresponding to the comprehensive impact index of power generation efficiency with the spatial topological data of the affected geographical area (including sub-region boundary coordinates and high-temperature impact coverage areas). Based on the overlaid multi-dimensional data layer, a dynamic heat map is generated using a thermal rendering algorithm. The thermal color levels are strictly correlated with the numerical range of the comprehensive impact index of power generation efficiency (e.g., red corresponds to a severe decline area with an index > 0.7, yellow corresponds to a moderate decline area with an index of 0.3-0.7, and green indicates a safe area). Spatial interpolation technology is used to visualize the gradient distribution of the index value in continuous geographic space. For high-risk areas (sub-regions with a comprehensive impact index > 0.7), preset warning color marks can be used to enhance the rendering (e.g., dark red fills and overlays flashing boundaries). At the same time, the predicted vector line of the high-temperature event evolution path is associated (the arrow direction indicates the movement trend, and the line width represents the intensity change) to achieve warning visualization. This process employs: a dynamic thermal mapping mechanism, which converts discrete sub-region index values ​​into continuous color-level surfaces using a bilinear interpolation algorithm; enhanced rendering of high-risk areas, overlaying preset warning markers in red warning zones and using channel transparency gradient technology to soften the transition of risk boundaries; and spatiotemporal dynamic coupling expression, with the thermal map dynamically updated every 15 minutes, synchronously overlaid with high-temperature path prediction vectors. The affected geographical area is covered by a semi-transparent masking layer (40% transparency) over the base map, and the boundary line width increases as the affected area expands. A GIS spatial analysis engine achieves precise coupling between meteorological risk and geographical elements. The generation of the thermal map can utilize open-source rendering libraries (such as GeoServer WMS service), and the warning color scales are dynamically adjusted for saturation and brightness using the HSV color model, forming a visual decision support map that intuitively reflects the spatiotemporal evolution of high-temperature impacts.

[0093] The aforementioned early warning method for the impact of extreme high temperatures on photovoltaic power generation efficiency, based on meteorological numerical models, utilizes regional background characteristics (such as geographical features and climate type) and historical meteorological data of the photovoltaic equipment deployment area. It performs layered sampling of temperature field data, extracts periodic features through wavelet transform, and statistically analyzes the movement paths, intensity changes, and durations of historical high-temperature events to generate accurate high-temperature weather evolution paths, addressing the problem of neglecting the spatiotemporal dynamic changes of the temperature field in traditional techniques. Based on this evolution path, the region is divided into multiple sub-regions, and temperature parameters are collected to construct a dynamic distribution model of high-temperature weather. This model generates event intensity characteristics (such as temperature fluctuation amplitude and cumulative duration gradient integral results exceeding preset temperature thresholds) and impact range (such as the proportion of space covered by the temperature exceedance). By comparing these with preset temperature thresholds, low, medium, and high risk levels are marked, improving the spatiotemporal accuracy of the early warning. For high-risk sub-regions, temperature monitoring data (such as module surface temperature and backsheet temperature) and performance output data (such as DC output power and conversion efficiency) of photovoltaic equipment are acquired. Combined with historical operating records, the deviation between the theoretical temperature response coefficient and the actual power generation efficiency degradation rate is calculated, constructing an impact assessment matrix to achieve quantitative correlation modeling between high-temperature events and equipment performance degradation. Based on this matrix, a multi-dimensional index analysis method is used to perform principal component analysis on the theoretical temperature response coefficient, actual attenuation rate, and performance loss deviation, extracting core principal component indicators. A weighted fusion calculation of the comprehensive impact index on power generation efficiency is then performed. Combined with a correlation coefficient matrix, an early warning report (including warning level, impact time window, and maximum efficiency reduction) is generated. A dynamic heat map is generated by overlaying the warning level with the affected area using a geographic information system platform. Through refined analysis of the dynamic characteristics of the temperature field and dual verification using physical models and measured data of equipment performance, this approach solves the problems of low spatiotemporal accuracy in early warning and difficulty in scientifically assessing the impact of high temperatures in traditional technologies. It achieves efficient early warning and intuitive visualization of photovoltaic power generation efficiency decline, providing precise support for operation and maintenance decisions (such as cooling system startup or grid dispatch), and significantly improving the operational reliability of photovoltaic systems under extreme high temperatures.

[0094] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0095] Based on the same inventive concept, this application also provides an early warning device for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models, used to implement the aforementioned early warning method for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the early warning device for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models provided below can be found in the limitations of the early warning method for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models described above, and will not be repeated here.

[0096] In one exemplary embodiment, such as Figure 2 As shown, an early warning device for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models is provided, comprising:

[0097] The temperature path modeling module 101 is used to perform layered sampling of temperature field data based on the regional background characteristics and historical meteorological data of the photovoltaic equipment deployment area, construct a high temperature characteristic database, and determine the evolution path of high temperature weather by combining time series changes.

[0098] The dynamic distribution modeling module 102 is used to divide the photovoltaic equipment deployment area into multiple sub-regions according to the evolution path, collect the temperature parameters of the sub-regions, and combine the temperature parameters with the evolution path to construct a dynamic distribution model of high temperature weather.

[0099] The risk level classification module 103 is used to generate the event intensity characteristics and impact range of high temperature weather based on the dynamic distribution model of high temperature weather, and compare the spatiotemporal changes of intensity characteristics with preset temperature thresholds to mark the sub-regions as low risk, medium risk and high risk levels respectively.

[0100] The impact assessment matrix module 104 is used to acquire temperature monitoring data and performance output data of photovoltaic equipment for high-risk sub-regions, and combine them with the historical operation records of the equipment to determine the specific impact of high temperature weather on equipment performance and construct an impact assessment matrix.

[0101] The early warning report generation module 105 is used to generate an early warning report on the impact on power generation efficiency based on the impact assessment matrix and using a multi-dimensional index analysis method, according to the correlation characteristics between the event intensity and the scope of impact.

[0102] In one embodiment, the temperature path modeling module 101 is further configured to:

[0103] Based on regional background characteristics and geographical features, climate types and historical temperature data in historical meteorological data, the temperature field data is preprocessed in time and space to obtain standardized temperature data.

[0104] Based on geographical features, standardized temperature field data is sampled in layers, and spatial distribution maps of temperature are generated by interpolation.

[0105] Frequency domain analysis was performed on the time series components of historical temperature data, wavelet transform was used to extract the periodic characteristics of high-temperature weather, and a high-temperature characteristic database was constructed by combining the spatial distribution map of temperature.

[0106] Based on the spatiotemporal trajectories and high-temperature characteristic databases of historical high-temperature events, statistical analysis is used to determine the movement path, intensity changes, and duration evolution patterns of high-temperature weather, thereby generating the evolution path of high-temperature weather.

[0107] In one embodiment, the risk level classification module 103 is further configured to:

[0108] Based on the dynamic distribution model of high temperature weather, dynamic temperature parameters of each sub-region are extracted, the proportion of spatial coverage area exceeding the preset temperature threshold is calculated, and the scope of influence is generated.

[0109] Based on the time change rate of dynamic temperature parameters, calculate the temperature fluctuation amplitude and the cumulative duration of continuous temperature exceeding the preset threshold, and generate an intensity cumulative gradient.

[0110] By using a preset weighting function, correlation analysis is performed on the spatial coverage area and intensity cumulative gradient to generate event intensity features.

[0111] In one embodiment, the influence assessment matrix module 104 is further configured to:

[0112] For high-risk sub-regions, acquire temperature monitoring data and performance output data of photovoltaic equipment deployed in the region during historical high-temperature events. The temperature monitoring data and performance output data include module surface temperature, backsheet temperature, ambient temperature, DC output power, AC output power, and conversion efficiency.

[0113] Based on the physical and installation parameters of photovoltaic equipment, combined with temperature monitoring data and performance output data, a temperature response model of photovoltaic modules is established to calculate the theoretical temperature response coefficient of photovoltaic equipment under current and predicted high temperature conditions.

[0114] By associating with historical operation records, the actual power generation efficiency degradation rate of photovoltaic equipment during the corresponding historical high-temperature period can be extracted.

[0115] The theoretical temperature response coefficient is compared and analyzed with the actual power generation efficiency decay rate to calculate the performance loss deviation caused by high temperature.

[0116] An impact assessment matrix is ​​constructed with theoretical temperature response coefficient, actual power generation efficiency attenuation rate, and performance loss deviation as key dimensions. The elements of the impact assessment matrix are associated with specific equipment models, installation locations, high-temperature event intensity characteristics, and impact ranges.

[0117] In one embodiment, the warning report generation module 105 is further configured to:

[0118] Principal component analysis was performed on the theoretical temperature response coefficient, actual power generation efficiency attenuation rate, and performance loss deviation in the impact assessment matrix to extract the core principal component indicators characterizing the sensitivity to high temperature impact.

[0119] Establish a correlation coefficient matrix between core principal component indicators and event intensity characteristics and scope of impact;

[0120] Based on the correlation coefficient matrix, a comprehensive impact index on power generation efficiency for each high-risk sub-region is generated through weighted fusion calculation.

[0121] Based on the comprehensive impact index of power generation efficiency, and according to the preset threshold range of the comprehensive impact index, different levels of early warning for power generation efficiency decline are divided.

[0122] Based on the predicted evolution path, impact range, and duration of high-temperature weather, a power generation efficiency impact warning report is generated. The warning report includes the warning level, the expected impact time window, the affected geographical area, the expected maximum reduction in power generation efficiency, and recommended measures.

[0123] In one embodiment, the early warning report generation module 105 is also used to calculate the comprehensive impact index of power generation efficiency using the following formula:

[0124]

[0125] Where δ=|η actual -R theory ·T excess | represents the performance loss deviation, η actual R represents the actual power generation efficiency degradation rate. theory T is the theoretical temperature response coefficient. excess S represents the cumulative duration of exceeding the preset temperature threshold. event As an event intensity characteristic, it is determined by the sum of the temperature fluctuation amplitude and the cumulative duration T of exceeding a preset temperature threshold. excess Gradient integral generation, A impact For the scope of influence, α i β i and γ i Principal component loading factor, through (δ,S event A impact Principal component analysis was performed to determine whether the following conditions were met. w i As a dynamic weighting factor, based on the event intensity feature S event Scope of influence A impact The spatiotemporal correlation coefficient allocation, and satisfying k is the number of principal components to be retained, determined by a cumulative variance contribution rate of ≥85%.

[0126] In one embodiment, the early warning report generation module 105 is further configured to perform early warning visualization processing according to the following steps:

[0127] By overlaying the warning level, the affected geographical range, and the regional map on the geographic information system platform, a dynamic heat map is generated; among them, the heat map color scale is associated with the comprehensive impact index of power generation efficiency, and high-risk areas are rendered with preset warning color marks.

[0128] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the aforementioned method for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models.

[0129] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0130] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0131] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A method for early warning of the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models, characterized in that, The method includes: Based on the regional background characteristics and historical meteorological data of the photovoltaic equipment deployment area, the temperature field data is sampled in layers to construct a high temperature characteristic database, and the evolution path of high temperature weather is determined by combining time series changes. Based on the evolution path, the photovoltaic equipment deployment area is divided into multiple sub-regions, temperature parameters of the sub-regions are collected, and a dynamic distribution model of high temperature weather is constructed by combining the temperature parameters with the evolution path. Based on the dynamic distribution model of high temperature weather, the event intensity characteristics and impact range of high temperature weather are generated, and the spatiotemporal changes of intensity characteristics are compared with preset temperature thresholds to mark the sub-regions as low-risk, medium-risk and high-risk levels respectively. For high-risk sub-regions, temperature monitoring data and performance output data of photovoltaic equipment are obtained. Combined with the historical operation records of the equipment, the specific impact of high temperature weather on equipment performance is determined, and an impact assessment matrix is ​​constructed. Based on the aforementioned impact assessment matrix, a multi-dimensional index analysis method is used to generate a power generation efficiency impact early warning report according to the correlation characteristics between event intensity and impact range, and spatial visualization technology is used for visualization output.

2. The method according to claim 1, characterized in that, The construction of a high-temperature characteristic database, and the determination of the evolution path of high-temperature weather by combining time series changes, includes: Based on the regional background characteristics and geographical features, climate types and historical temperature data in the historical meteorological data, the temperature field data is preprocessed in time and space to obtain standardized temperature data. Based on the geographical features, the standardized temperature field data is sampled in layers, and a spatial distribution map of temperature is generated by interpolation. Frequency domain analysis is performed on the time series components of the historical temperature data, wavelet transform is used to extract the periodic characteristics of high-temperature weather, and the high-temperature feature database is constructed by combining the spatial distribution map of the temperature. Based on the spatiotemporal trajectories of historical high-temperature events and the high-temperature characteristic database, statistical analysis is used to determine the movement path, intensity changes, and duration evolution patterns of high-temperature weather, thereby generating the evolution path of the high-temperature weather.

3. The method according to claim 1, characterized in that, The generation of event intensity characteristics and impact range of high-temperature weather based on the dynamic distribution model of high-temperature weather includes: Based on the dynamic distribution model of high temperature weather, dynamic temperature parameters of each sub-region are extracted, the proportion of spatial coverage area exceeding the preset temperature threshold is calculated, and the influence range is generated. Based on the time change rate of the dynamic temperature parameters, the temperature fluctuation amplitude and the cumulative duration of continuous temperature exceeding the preset threshold are calculated, and an intensity cumulative gradient is generated. The spatial coverage area and intensity cumulative gradient are correlated using a preset weighting function to generate the event intensity features.

4. The method according to claim 3, characterized in that, For the high-risk sub-regions, temperature monitoring data and performance output data of photovoltaic equipment are acquired. Combined with the equipment's historical operating records, the specific impact of high temperatures on equipment performance is determined, and an impact assessment matrix is ​​constructed, including: For the high-risk sub-regions, acquire temperature monitoring data and performance output data of photovoltaic equipment deployed in the region during historical high-temperature events. The temperature monitoring data and performance output data include module surface temperature, backsheet temperature, ambient temperature, DC output power, AC output power, and conversion efficiency. Based on the physical and installation parameters of the photovoltaic equipment, combined with the temperature monitoring data and performance output data, a photovoltaic module temperature response model is established, and the theoretical temperature response coefficient of the photovoltaic equipment under current and predicted high temperature conditions is calculated. By associating the historical operation records, the actual power generation efficiency degradation rate of the photovoltaic equipment during the corresponding historical high-temperature period is extracted; The theoretical temperature response coefficient is compared and analyzed with the actual power generation efficiency decay rate to calculate the performance loss deviation caused by high temperature. An impact assessment matrix is ​​constructed with the theoretical temperature response coefficient, actual power generation efficiency attenuation rate, and performance loss deviation as key dimensions. The elements of the impact assessment matrix are associated with specific equipment models, installation locations, high-temperature event intensity characteristics, and impact ranges.

5. The method according to claim 4, characterized in that, Based on the impact assessment matrix, a multi-dimensional index analysis method is used to generate a power generation efficiency impact early warning report according to the correlation characteristics between event intensity and impact range, including: Principal component analysis was performed on the theoretical temperature response coefficient, actual power generation efficiency attenuation rate, and performance loss deviation in the impact assessment matrix to extract the core principal component indicators characterizing the sensitivity to high temperature impact. Establish a correlation coefficient matrix between the core principal component indicators and the event intensity characteristics and scope of influence; Based on the correlation coefficient matrix, a comprehensive impact index on power generation efficiency for each high-risk sub-region is generated through weighted fusion calculation. Based on the comprehensive impact index of power generation efficiency, and according to the preset threshold range of the comprehensive impact index, different levels of power generation efficiency decline warning are divided. Based on the predicted evolution path, impact range, and duration of the high-temperature weather, a power generation efficiency impact warning report is generated; the warning report includes the warning level, the expected impact time window, the affected geographical range, the expected maximum power generation efficiency reduction, and recommended measures.

6. The method according to claim 5, characterized in that, The comprehensive impact index of power generation efficiency is calculated using the following formula: Where, δ=|η actual -R theory ·T excess | represents the performance loss deviation, η actual R represents the actual power generation efficiency degradation rate. theory T is the theoretical temperature response coefficient. excess S represents the cumulative duration of exceeding the preset temperature threshold. event As an event intensity characteristic, it is determined by the cumulative duration T of the temperature fluctuation amplitude exceeding the preset temperature threshold. excess Gradient integral generation, A impact For the scope of influence, α i β i and γ i Principal component loading factor, through (δ,S event A impact Principal component analysis was performed to determine whether the following conditions were met. w i As a dynamic weighting factor, based on the event intensity feature S event Scope of influence A impact The spatiotemporal correlation coefficient allocation, and satisfying k is the number of principal components to be retained, determined by a cumulative variance contribution rate of ≥85%.

7. The method according to claim 1, characterized in that, After generating the power generation efficiency impact early warning report, the process also includes visualizing the early warning according to the following steps: By overlaying the warning level, the affected geographical range, and the regional map on the geographic information system platform, a dynamic heat map is generated; among them, the heat map color scale is associated with the comprehensive impact index of power generation efficiency, and high-risk areas are rendered with preset warning color marks.

8. An early warning device for the impact of extreme high temperatures on photovoltaic power generation efficiency based on meteorological numerical models, characterized in that, The device includes: The temperature path modeling module is used to perform layered sampling of temperature field data based on the regional background characteristics and historical meteorological data of the photovoltaic equipment deployment area, construct a high temperature characteristic database, and determine the evolution path of high temperature weather by combining time series changes. The dynamic distribution modeling module is used to divide the photovoltaic equipment deployment area into multiple sub-regions according to the evolution path, collect the temperature parameters of the sub-regions, and combine the temperature parameters with the evolution path to construct a dynamic distribution model of high temperature weather. The risk level classification module is used to generate the event intensity characteristics and impact range of high temperature weather based on the dynamic distribution model of high temperature weather, and compare the spatiotemporal changes of the intensity characteristics with the preset temperature threshold to mark the sub-regions as low risk, medium risk and high risk levels respectively. The impact assessment matrix module is used to acquire temperature monitoring data and performance output data of photovoltaic equipment for high-risk sub-regions, and combine them with the equipment's historical operating records to determine the specific impact of high-temperature weather on equipment performance and construct an impact assessment matrix. The early warning report generation module is used to generate an early warning report on the impact on power generation efficiency based on the impact assessment matrix and using a multi-dimensional index analysis method, according to the correlation characteristics between event intensity and impact range.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images