A method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning

Abstract

The present invention discloses a method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning. The method comprises the following steps: collecting long-term meteorological data series, inputting the data into a meteorological data preprocessing model, and generating preprocessed meteorological data; inputting the preprocessed meteorological data into an extreme weather event recognition model, extracting the meteorological characteristics, changing trends, and hysteresis characteristics of extreme weather events, and identifying and marking the extreme weather event period; establishing a hybrid scenario generation model based on the meteorological characteristics, changing trends, and hysteresis characteristics of the extreme weather event period to generate extreme weather scenarios; performing dimensionality reduction, visualization, and quality assessment comparison between the generated extreme weather scenarios and real extreme weather scenarios using a multi-level dimensionality reduction visualization model, and calculating correction factors based on the quality assessment comparison results; and establishing a photovoltaic output scenario generation model based on the correction factors to simulate photovoltaic output conditions under extreme weather scenarios. The present invention can improve the diversity and realism of extreme weather scenario generation.

Description

Technical Field

[0001] The present invention belongs to the field of meteorological data processing and artificial intelligence technology, and specifically relates to a method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning. Background Art

[0002] As climate change intensifies, the frequency and intensity of extreme weather events are increasing, with increasingly significant impacts on society and the economy. Accurately simulating and generating extreme weather scenarios is crucial for disaster prevention, emergency management, climate research, and other fields. Weather variations, such as light intensity, temperature, and humidity, directly impact the power generation efficiency of photovoltaic power plants. Extreme weather conditions, in particular, exacerbate the volatility of photovoltaic output. To accurately assess photovoltaic output in extreme weather conditions, an efficient and accurate method for generating photovoltaic output scenarios is required, taking weather factors into account. This method can provide a scientific basis for power grid planning, design, and operation.

[0003] In existing technology systems, the generation of photovoltaic output scenarios typically relies on numerical weather forecasts and physical models. While traditional physical modeling methods can accurately model the physical characteristics and performance parameters of solar panels, their generalization capabilities are limited when dealing with complex and changing weather conditions. In extreme weather conditions, the simulation accuracy of physical models can significantly decrease. Furthermore, traditional statistical modeling methods often struggle to effectively capture the dynamic characteristics of time series data. Furthermore, traditional methods struggle to generate diverse extreme weather scenarios, failing to meet the needs of diverse application scenarios. Summary of the Invention

[0004] In response to the above-mentioned deficiencies in the prior art, the present invention provides a method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning, which can improve the diversity and realism of extreme weather scenario generation, and solve the problem that traditional methods are difficult to generate diverse extreme weather scenarios.

[0005] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning, comprising the following steps:

[0006] S1. Collect long-term meteorological data and input it into a meteorological data preprocessing model to generate preprocessed meteorological data;

[0007] S2. Input the pre-processed meteorological data into the extreme weather event recognition model to extract the meteorological characteristics, change trends and hysteresis characteristics when extreme weather occurs, and identify and mark the extreme weather event period;

[0008] S3. Establish a hybrid scenario generation model based on the meteorological characteristics, change trends, and hysteresis characteristics of the extreme weather event period to generate extreme weather scenarios;

[0009] S4. Perform dimensionality reduction visualization and quality assessment comparison on the generated extreme weather scenario and the real extreme weather scenario through a multi-level dimensionality reduction visualization model, and calculate the correction factor based on the quality assessment comparison results;

[0010] S5. Establish a photovoltaic output scenario generation model based on the correction factor to simulate the photovoltaic output under extreme weather scenarios.

[0011] Furthermore: in S1, the long-term meteorological data include: temperature, humidity, wind speed, solar radiation intensity, precipitation, snow depth and air pressure;

[0012] The method for generating preprocessed meteorological data is as follows:

[0013] Use linear interpolation or historical data mean to fill missing data in long-term series meteorological data;

[0014] Identify and classify outliers in long-term meteorological data using the Z-score method, including deleting and interpolating obvious erroneous values, and retaining and labeling true values.

[0015] Key features related to photovoltaic output are extracted and normalized to adapt to the extreme weather event identification model. The key features related to photovoltaic output include solar radiation intensity, temperature change rate, and the correlation between humidity and solar radiation.

[0016] Further: In S2, the meteorological characteristics when extreme weather occurs are extracted The specific expression is:

[0017]

[0018] Where, The temperature when extreme weather occurs. is the dew point temperature when extreme weather occurs. is the amount of precipitation when extreme weather occurs. is the air pressure when extreme weather occurs. The depth of snow when extreme weather occurs. is the solar radiation when extreme weather occurs, For visibility during extreme weather events, is the wind speed when extreme weather occurs, is the relative humidity when extreme weather occurs, is the absolute humidity when extreme weather occurs, is the cloud cover when extreme weather occurs. is the cloud height when extreme weather occurs. is the water vapor pressure when extreme weather occurs, is the precipitation intensity when extreme weather occurs, is the evaporation characteristic when extreme weather occurs;

[0019] Extracting change trends The specific expression is:

[0020]

[0021] Where, is the temperature change before or after extreme weather occurs. is the change in dew point temperature before or after extreme weather occurs. is the change in precipitation before or after extreme weather occurs. It is the change in air pressure before or after extreme weather occurs. is the change in snow depth before or after extreme weather occurs, is the change in solar radiation before or after extreme weather occurs. is the visibility change before or after extreme weather occurs, is the change in wind speed before or after extreme weather occurs, is the change in relative humidity before or after extreme weather occurs. is the absolute humidity change before or after extreme weather occurs, is the change in cloud cover before or after extreme weather occurs. is the cloud height change before or after extreme weather occurs, is the change in water vapor pressure before or after extreme weather occurs. is the change in precipitation intensity before or after extreme weather occurs, It is the change in evaporation before or after the occurrence of extreme weather;

[0022] Extracting hysteresis features The specific expression is:

[0023]

[0024] Where, for Photovoltaic output scene at all times, When extreme weather occurs, This is the photovoltaic output scenario within a few hours after the extreme weather occurs. is the lag time window.

[0025] Furthermore: in S3, the hybrid scenario generation model includes a generator and a discriminator, and the method for generating extreme weather scenarios is specifically: simulating and generating extreme weather scenarios by the generator and the discriminator together;

[0026] Among them, the loss function of the generator is and the loss function of the discriminator The specific expression is:

[0027]

[0028] Where, Represents the distribution A random noise vector sampled from Perform expected value calculations, and Respectively represent the distribution Sampling and the distribution of Sampling Perform expected value calculations, is a random noise vector, 、 、 are the distributions of random noise, real data, and generated data, respectively, These are extreme weather labels, including high temperature, cold wave, heavy rain, blizzard, strong wind, drought, thunderstorm, hail, heavy fog, extreme precipitation and sandstorm. 、 are the discriminator and the generator respectively, 、 is the weight of the mean square error and gradient penalty term, It is a set of real extreme weather scene characteristics, including: air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation meteorological characteristics; The generated extreme weather scene feature set includes: air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation meteorological characteristics; is the mean square error between the generated extreme weather scene feature set and the real extreme weather scene feature set, represents the gradient penalty term, and are the probabilities of real extreme weather scenarios and generated extreme weather scenarios, respectively. is a smoothing parameter less than 1, which is used to control the confidence of the discriminator in the generated scene.

[0029] Furthermore, in S4, the expression for dimensionality reduction visualization of the generated extreme weather scenario and the real extreme weather scenario is specifically:

[0030]

[0031] in, For the scene after dimensionality reduction, Principal component analysis, is a t-distributed random neighborhood embedding, is the uniform manifold approximation and projection, and They are the feature sets of real and generated extreme weather scenarios, respectively, including meteorological characteristics of air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation.

[0032] Furthermore, in S4, the method for performing quality assessment and comparison between the generated extreme weather scenario and the real extreme weather scenario is specifically as follows:

[0033] Calculate evaluation indicators to obtain comparison results between the generated extreme weather scenarios and real extreme weather scenarios, and calculate correction factors based on the comparison results. The correction factors are dynamically adjusted according to the type and intensity of extreme weather to ensure that the model can accurately simulate the impact of extreme weather on PV output;

[0034] Among them, the correction factors include extreme weather direct radiation, solar azimuth, diffuse radiation, ground reflected radiation and photovoltaic power correction factors.

[0035] Furthermore, in S4, the evaluation indicators for comparing the quality of the generated extreme weather scenarios with the real extreme weather scenarios include:

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042] Where, 、 、 and They are the comprehensive evaluation index, Wasserstein distance, mean square error and mutual information between the generated extreme weather scene feature set and the real extreme weather scene feature set; is a Lipschitz continuous function The constraints, is the expected value of the real extreme weather scene feature set, is the expected value of the generated extreme weather scene feature set, is the total number of scenes, and Respectively A real extreme weather scene feature set and a generated extreme weather scene feature set, is the joint probability distribution, and are the marginal probability distributions of the real extreme weather scene feature set and the generated extreme weather scene feature set, 、 、 is the weight coefficient, which is dynamically adjusted according to the extreme weather type to reflect the focus under different conditions.

[0043] Furthermore: In S5, the method for simulating photovoltaic output under extreme weather scenarios is specifically as follows:

[0044] S51, the sun elevation angle It is divided into several intervals, each interval is 15 degrees. The radiation value of each 15-degree solar elevation angle is calculated by the extreme weather direct radiation, solar azimuth, diffuse radiation and ground reflected radiation correction factor, and then the total solar radiation received is calculated. ;

[0045]

[0046]

[0047]

[0048]

[0049] in, For the The solar elevation angle corresponding to each interval divides 180° into 12 intervals. 、 、 and They are The radiation attenuation of direct radiation, diffuse radiation, ground reflected radiation and extreme weather received in each interval is related to the type and intensity of extreme weather. 、 and They are The direct radiation, diffuse radiation and total radiation received on the ground in each interval, is the tilt angle of the photovoltaic panel, 、 、 and They are the correction factors for extreme weather direct radiation, solar azimuth, diffuse radiation and ground reflected radiation, and their values ​​are determined according to the type and intensity of extreme weather. is the ground albedo, It is the serial number of the day. is the latitude of the photovoltaic power station, is the solar declination;

[0050] S52, based on the total solar radiation received , the photovoltaic power correction factor is used to consider the impact of extreme weather on photovoltaic power, and the photovoltaic output under extreme weather scenarios is calculated. ;

[0051]

[0052] Where, is the conversion efficiency of the photovoltaic system, is the area of ​​the photovoltaic panel, is the total solar radiation received by the photovoltaic system, is the extreme weather photovoltaic power correction factor, and its value is determined according to the type and intensity of extreme weather.

[0053] The beneficial effects of the present invention are:

[0054] (1) This invention provides a method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning. This method significantly improves the accuracy of generating photovoltaic output scenarios under extreme weather conditions, based on a hybrid scenario generation model architecture using long short-term memory networks, generative adversarial networks, transformer architecture, and attention mechanisms. This method not only effectively addresses the scarcity of extreme weather scenarios, but also improves the generalization capability of the model through data augmentation technology. Furthermore, it comprehensively considers the direct impact and lag effect of extreme weather on photovoltaic output, making the generated results more realistic and resolving the difficulty of traditional methods in generating diverse extreme weather scenarios.

[0055] (2) The present invention provides a more reliable basis for grid planning, operation optimization and risk management, helps to optimize power dispatching, reduce grid operation costs, and enhance the stability and resilience of the grid under extreme weather conditions, thereby promoting the healthy development of the photovoltaic industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1This is a flow chart of a method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning according to the present invention. DETAILED DESCRIPTION

[0057] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0058] like Figure 1 As shown, in one embodiment of the present invention, a method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning includes the following steps:

[0059] S1. Collect long-term meteorological data and input it into a meteorological data preprocessing model to generate preprocessed meteorological data;

[0060] S2. Input the pre-processed meteorological data into the extreme weather event recognition model to extract the meteorological characteristics, change trends and hysteresis characteristics when extreme weather occurs, and identify and mark the extreme weather event period;

[0061] S3. Establish a hybrid scenario generation model based on the meteorological characteristics, change trends, and hysteresis characteristics of the extreme weather event period to generate extreme weather scenarios;

[0062] S4. Perform dimensionality reduction visualization and quality assessment comparison on the generated extreme weather scenario and the real extreme weather scenario through a multi-level dimensionality reduction visualization model, and calculate the correction factor based on the quality assessment comparison results;

[0063] S5. Establish a photovoltaic output scenario generation model based on the correction factor to simulate the photovoltaic output under extreme weather scenarios.

[0064] In S1, long-term meteorological data include: temperature, humidity, wind speed, solar radiation intensity, precipitation, snow depth and air pressure;

[0065] In this embodiment, the present invention extracts long-term series meteorological data in real time to continuously capture the impact of diurnal and seasonal changes on photovoltaic output 24 hours a day.

[0066] Data is collected from multiple channels, including meteorological stations, satellite remote sensing data, on-site monitoring equipment at photovoltaic power plants, and historical meteorological archives. These data cover hourly meteorological records for the past 10 years, ensuring comprehensiveness and diversity. The collected meteorological data include but are not limited to temperature, humidity, wind speed, solar radiation intensity, precipitation, snow depth, air pressure, etc. Special attention is paid to extreme weather events such as high temperatures (over 35°C), low temperatures (below -10°C), heavy rain (daily precipitation exceeding 50mm), blizzards (wind speed exceeding 20m / s accompanied by snowfall), etc., and the time series integrity of the data is ensured. The data is recorded 24 hours a day, 365 days a year, to capture the impact of diurnal variations and seasonal patterns on photovoltaic output.

[0067] The method for generating preprocessed meteorological data is as follows:

[0068] Linear interpolation or the mean of historical data is used to fill missing data in long-term meteorological data series; for example, if the solar radiation data for a certain hour is missing, linear interpolation will be performed using the data from the two hours before and after.

[0069] Outliers in long-term meteorological data are identified and classified using the Z-score method. This involves removing and interpolating obviously erroneous values, while retaining and annotating true values. For example, if the maximum temperature on a given day exceeds three standard deviations of the historical high, it is considered an outlier. Outliers are handled according to the following circumstances: If the outlier is clearly erroneous (e.g., an unusually low temperature of -50°C), the data point is removed and interpolated. If the outlier is extreme but potentially true (e.g., an unusually high temperature or heavy rainfall), the data point is retained and specifically annotated in subsequent analysis.

[0070] Key features related to photovoltaic output are extracted and normalized to adapt to the extreme weather event identification model. The key features related to photovoltaic output include solar radiation intensity, temperature change rate, and the correlation between humidity and solar radiation.

[0071] Through this process, the preprocessed data loss rate is reduced, the proportion of outliers is reduced, and the data integrity and accuracy are significantly improved. The normalized data is better suited to the input requirements of deep learning models, improving model training efficiency. The model's generalization ability under extreme weather conditions is significantly enhanced, enabling more accurate generation of PV output scenarios in different regions and under different types of extreme weather conditions.

[0072] In S2, the meteorological characteristics of extreme weather are extracted The specific expression is:

[0073]

[0074] Where, The temperature when extreme weather occurs. is the dew point temperature when extreme weather occurs. is the amount of precipitation when extreme weather occurs. is the air pressure when extreme weather occurs. The depth of snow when extreme weather occurs. is the solar radiation when extreme weather occurs, For visibility during extreme weather events, is the wind speed when extreme weather occurs, is the relative humidity when extreme weather occurs, is the absolute humidity when extreme weather occurs, is the cloud cover when extreme weather occurs. is the cloud height when extreme weather occurs. is the water vapor pressure when extreme weather occurs, is the precipitation intensity when extreme weather occurs, is the evaporation characteristic when extreme weather occurs;

[0075] Extracting change trends The specific expression is:

[0076]

[0077] Where, is the temperature change before or after extreme weather occurs. is the change in dew point temperature before or after extreme weather occurs. is the change in precipitation before or after extreme weather occurs. is the change in air pressure before or after extreme weather occurs. is the change in snow depth before or after extreme weather occurs, is the change in solar radiation before or after extreme weather occurs. is the visibility change before or after extreme weather occurs, is the change in wind speed before or after extreme weather occurs, is the change in relative humidity before or after extreme weather occurs. is the absolute humidity change before or after extreme weather occurs, is the change in cloud cover before or after extreme weather occurs. is the cloud height change before or after extreme weather occurs, is the change in water vapor pressure before or after extreme weather occurs. is the change in precipitation intensity before or after extreme weather occurs, It is the change in evaporation before or after the occurrence of extreme weather;

[0078] Extracting hysteresis features The specific expression is:

[0079]

[0080] Where, for Photovoltaic output scene at all times, When extreme weather occurs, This is the photovoltaic output scenario within a few hours after the extreme weather occurs. is the lag time window.

[0081] In S3, the hybrid scenario generation model includes a generator and a discriminator. The method for generating extreme weather scenarios is as follows: the generator and the discriminator jointly simulate and generate extreme weather scenarios;

[0082] Among them, the loss function of the generator is and the loss function of the discriminator The specific expression is:

[0083]

[0084]

[0085] Where, Represents the distribution A random noise vector sampled from Perform expected value calculations, and Respectively represent the distribution Sampling and the distribution of Sampling Perform expected value calculations, is a random noise vector, 、 、 are the distributions of random noise, real data, and generated data, respectively, These are extreme weather labels, including high temperature, cold wave, heavy rain, blizzard, strong wind, drought, thunderstorm, hail, heavy fog, extreme precipitation and sandstorm. 、 are the discriminator and the generator respectively, 、 is the weight of the mean square error and gradient penalty term, It is a set of real extreme weather scene characteristics, including: air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation meteorological characteristics; The generated extreme weather scene feature set includes: air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation meteorological characteristics; is the mean square error between the generated extreme weather scene feature set and the real extreme weather scene feature set, represents the gradient penalty term, and are the probabilities of real extreme weather scenarios and generated extreme weather scenarios, respectively. is a smoothing parameter less than 1, which is used to control the confidence of the discriminator in the generated scene.

[0086] In this embodiment, the hybrid scenario generation model (LGTA) is a model architecture based on the long short-term memory network (LSTM), the generative adversarial network (GAN), the Transformer architecture, and the attention mechanism. By alternately training the generator and discriminator of the generative adversarial network (GAN), the discriminator is trained to distinguish between real and false data, and the generator is trained to generate realistic data. The model is continuously optimized to generate more realistic extreme weather scenarios.

[0087] The model parameters are set as follows: the dimension of the random noise vector z is Lz, and the dimension of the extreme weather label y is Ly. The LSTM hidden layer dimension is H, and the output dimension is consistent with the real scene, including hourly meteorological features. Both the generator and discriminator use an L-layer LSTM structure, the number of Transformer layers is T, and the number of multi-head attention heads is A. The Adam optimizer is used, and the learning rate is decayed by Dr every 100 iterations.

[0088] In S4, the expression for dimensionality reduction visualization of the generated extreme weather scenarios and the real extreme weather scenarios is as follows:

[0089]

[0090] in, For the scene after dimensionality reduction, Principal component analysis, is a t-distributed random neighborhood embedding, is the uniform manifold approximation and projection, and They are the feature sets of real and generated extreme weather scenarios, respectively, including meteorological characteristics of air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation.

[0091] In this embodiment, through dimensionality reduction visualization, the high-dimensional extreme weather scene feature set can be mapped to a low-dimensional space, and the similarities and differences between the generated extreme weather scenes and the real extreme weather scenes can be intuitively observed for subsequent quality assessment and comparison.

[0092] In S4, the method for comparing the quality of the generated extreme weather scenarios with the real extreme weather scenarios is as follows:

[0093] Calculate evaluation indicators to obtain comparison results between the generated extreme weather scenarios and real extreme weather scenarios, and calculate correction factors based on the comparison results. The correction factors are dynamically adjusted according to the type and intensity of extreme weather to ensure that the model can accurately simulate the impact of extreme weather on PV output;

[0094] Among them, the correction factors include extreme weather direct radiation, solar azimuth, diffuse radiation, ground reflected radiation and photovoltaic power correction factors.

[0095] In this embodiment, the value of the correction factor is dynamically adjusted according to the type and intensity of extreme weather (including high temperature, cold wave, heavy rain, heavy snow, strong wind, drought, thunderstorm, hail, heavy fog, extreme precipitation and sandstorm) to ensure that the model can accurately simulate the impact of extreme weather on photovoltaic output.

[0096] In S4, the evaluation indicators for quality evaluation and comparison of generated extreme weather scenarios and real extreme weather scenarios include:

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103] Where, 、 、 and They are the comprehensive evaluation index, Wasserstein distance, mean square error and mutual information between the generated extreme weather scene feature set and the real extreme weather scene feature set; is a Lipschitz continuous function The constraints, is the expected value of the real extreme weather scene feature set, is the expected value of the generated extreme weather scene feature set, is the total number of scenes, and Respectively A real extreme weather scene feature set and a generated extreme weather scene feature set, is the joint probability distribution, and are the marginal probability distributions of the real extreme weather scene feature set and the generated extreme weather scene feature set, 、 、 is the weight coefficient, which is dynamically adjusted according to the extreme weather type to reflect the focus under different conditions.

[0104] In S5, the method for simulating photovoltaic output under extreme weather scenarios is as follows:

[0105] S51, the sun elevation angle It is divided into several intervals, each interval is 15 degrees. The radiation value of each 15-degree solar elevation angle is calculated by the extreme weather direct radiation, solar azimuth, diffuse radiation and ground reflected radiation correction factor, and then the total solar radiation received is calculated. ;

[0106]

[0107]

[0108]

[0109]

[0110] in, For the The solar elevation angle corresponding to each interval divides 180° into 12 intervals. 、 、 and They are The radiation attenuation of direct radiation, diffuse radiation, ground reflected radiation and extreme weather received in each interval is related to the type and intensity of extreme weather. 、 and They are The direct radiation, diffuse radiation and total radiation received on the ground in each interval, is the tilt angle of the photovoltaic panel, 、 、 and They are the correction factors for extreme weather direct radiation, solar azimuth, diffuse radiation and ground reflected radiation, and their values ​​are determined according to the type and intensity of extreme weather. is the ground albedo, It is the serial number of the day. is the latitude of the photovoltaic power station, is the solar declination;

[0111] In this embodiment, the total solar radiation received by the photovoltaic system consists of direct solar radiation, diffuse solar radiation and ground reflected radiation. At the same time, considering the shielding and attenuation effects of extreme weather on solar radiation, a correction factor is introduced to make the result more accurate. Divide it into several intervals, each interval is 15 degrees, and calculate the radiation value for every 15 degrees of solar elevation angle.

[0112] S52, based on the total solar radiation received , the photovoltaic power correction factor is used to consider the impact of extreme weather on photovoltaic power, and the photovoltaic output under extreme weather scenarios is calculated. ;

[0113]

[0114] Where, is the conversion efficiency of the photovoltaic system, is the area of ​​the photovoltaic panel, is the total solar radiation received by the photovoltaic system, is the extreme weather photovoltaic power correction factor, and its value is determined according to the type and intensity of extreme weather.

[0115] In this embodiment, the photovoltaic power output under extreme weather conditions is affected by the increase in photovoltaic panel temperature (high temperature) and shading (heavy rain, snowstorm). The impact of extreme weather on photovoltaic power is taken into account by introducing a photovoltaic power correction factor.

[0116] In the description of the present invention, it should be understood that the terms "center", "thickness", "upper", "lower", "horizontal", "top", "bottom", "inner", "outer", "radial", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and cannot be understood as indicating or implying the relative importance or the number of technical features implicitly specified. Therefore, the features defined by "first", "second", and "third" may explicitly or implicitly include one or more of such features.

Claims

1. A method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning, characterized in that: The following steps are involved: S1. Collect long-term meteorological data and input it into a meteorological data preprocessing model to generate preprocessed meteorological data; S2. Input the pre-processed meteorological data into the extreme weather event recognition model to extract the meteorological characteristics, change trends and hysteresis characteristics when extreme weather occurs, and identify and mark the extreme weather event period; S3. Establish a hybrid scenario generation model based on the meteorological characteristics, change trends, and hysteresis characteristics of the extreme weather event period to generate extreme weather scenarios; In S3, the hybrid scenario generation model includes a generator and a discriminator. The method for generating extreme weather scenarios is as follows: the generator and the discriminator jointly simulate and generate extreme weather scenarios; Among them, the loss function of the generator is and the loss function of the discriminator The specific expression is: Where, Represents the distribution A random noise vector sampled from Perform expected value calculations, and Respectively represent the distribution Sampling and the distribution of Sampling Perform expected value calculations, is a random noise vector, 、 、 are the distributions of random noise, real data, and generated data, respectively, These are extreme weather labels, including high temperature, cold wave, heavy rain, blizzard, strong wind, drought, thunderstorm, hail, heavy fog, extreme precipitation and sandstorm. 、 are the discriminator and the generator respectively, 、 is the weight of the mean square error and gradient penalty term, It is a set of real extreme weather scene characteristics, including: air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation meteorological characteristics; The generated extreme weather scene feature set includes: air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation meteorological characteristics; is the mean square error between the generated extreme weather scene feature set and the real extreme weather scene feature set, represents the gradient penalty term, and are the probabilities of real extreme weather scenarios and generated extreme weather scenarios, respectively. is a smoothing parameter less than 1, used to control the confidence of the discriminator in the generated scene; S4. Perform dimensionality reduction visualization and quality assessment comparison on the generated extreme weather scenario and the real extreme weather scenario through a multi-level dimensionality reduction visualization model, and calculate the correction factor based on the quality assessment comparison results; S5. Establish a photovoltaic output scenario generation model based on the correction factor to simulate the photovoltaic output under extreme weather scenarios.

2. The method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning according to claim 1, characterized in that: In S1, long-term meteorological data include: temperature, humidity, wind speed, solar radiation intensity, precipitation, snow depth and air pressure; The method for generating preprocessed meteorological data is as follows: Use linear interpolation or historical data mean to fill missing data in long-term series meteorological data; Identify and classify outliers in long-term meteorological data using the Z-score method, including deleting and interpolating obvious erroneous values, and retaining and labeling true values. Key features related to photovoltaic output are extracted and normalized to adapt to the extreme weather event identification model. The key features related to photovoltaic output include solar radiation intensity, temperature change rate, and the correlation between humidity and solar radiation.

3. The method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning according to claim 1, characterized in that: In S2, the meteorological characteristics of extreme weather are extracted The specific expression is: Where, The temperature when extreme weather occurs. is the dew point temperature when extreme weather occurs. is the amount of precipitation when extreme weather occurs. is the air pressure when extreme weather occurs. The depth of snow when extreme weather occurs. is the solar radiation when extreme weather occurs, For visibility during extreme weather events, is the wind speed when extreme weather occurs, is the relative humidity when extreme weather occurs, is the absolute humidity when extreme weather occurs, is the cloud cover when extreme weather occurs. is the cloud height when extreme weather occurs. is the water vapor pressure when extreme weather occurs, is the precipitation intensity when extreme weather occurs, is the evaporation characteristic when extreme weather occurs; Extracting change trends The specific expression is: Where, is the temperature change before or after extreme weather occurs. is the change in dew point temperature before or after extreme weather occurs. is the change in precipitation before or after extreme weather occurs. It is the change in air pressure before or after extreme weather occurs. is the change in snow depth before or after extreme weather occurs, is the change in solar radiation before or after extreme weather occurs. is the visibility change before or after extreme weather occurs, is the change in wind speed before or after extreme weather occurs, is the change in relative humidity before or after extreme weather occurs. is the absolute humidity change before or after extreme weather occurs, is the change in cloud cover before or after extreme weather occurs. is the cloud height change before or after extreme weather occurs, is the change in water vapor pressure before or after extreme weather occurs. is the change in precipitation intensity before or after extreme weather occurs, It is the change in evaporation before or after the occurrence of extreme weather; Extracting hysteresis features The specific expression is: Where, for Photovoltaic output scene at all times, When extreme weather occurs, This is the photovoltaic output scenario within a few hours after the extreme weather occurs. is the lag time window.

4. The method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning according to claim 1, characterized in that: In S4, the expression for dimensionality reduction visualization of the generated extreme weather scenarios and the real extreme weather scenarios is as follows: in, For the scene after dimensionality reduction, Principal component analysis, is a t-distributed random neighborhood embedding, is the uniform manifold approximation and projection, and They are the feature sets of real and generated extreme weather scenarios, respectively, including meteorological characteristics of air temperature, dew point temperature, precipitation, air pressure, snow depth, solar radiation, visibility, wind speed, relative humidity, absolute humidity, cloud cover, cloud height, water vapor pressure, precipitation intensity and evaporation.

5. The method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning according to claim 4, characterized in that: In S4, the method for comparing the quality of the generated extreme weather scenarios with the real extreme weather scenarios is as follows: Calculate evaluation indicators to obtain comparison results between the generated extreme weather scenarios and real extreme weather scenarios, and calculate correction factors based on the comparison results. The correction factors are dynamically adjusted according to the type and intensity of extreme weather to ensure that the model can accurately simulate the impact of extreme weather on PV output; Among them, the correction factors include extreme weather direct radiation, solar azimuth, diffuse radiation, ground reflected radiation and photovoltaic power correction factors.

6. The method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning according to claim 5, characterized in that: In S4, the evaluation indicators for quality evaluation and comparison of generated extreme weather scenarios and real extreme weather scenarios include: Where, 、 、 and They are the comprehensive evaluation index, Wasserstein distance, mean square error and mutual information between the generated extreme weather scene feature set and the real extreme weather scene feature set; is a Lipschitz continuous function The constraints, is the expected value of the real extreme weather scene feature set, is the expected value of the generated extreme weather scene feature set, is the total number of scenes, and Respectively A real extreme weather scene feature set and a generated extreme weather scene feature set, is the joint probability distribution, and are the marginal probability distributions of the real extreme weather scene feature set and the generated extreme weather scene feature set, 、 、 is the weight coefficient, which is dynamically adjusted according to the extreme weather type to reflect the focus under different conditions.

7. The method for generating photovoltaic output scenarios under extreme weather conditions based on deep learning according to claim 6, characterized in that: In S5, the method for simulating photovoltaic output under extreme weather scenarios is as follows: S51, the sun elevation angle It is divided into several intervals, each interval is 15 degrees. The radiation value of each 15-degree solar elevation angle is calculated by the extreme weather direct radiation, solar azimuth, diffuse radiation and ground reflected radiation correction factor, and then the total solar radiation received is calculated. ; in, For the The solar elevation angle corresponding to each interval divides 180° into 12 intervals. 、 、 and They are The radiation attenuation of direct radiation, diffuse radiation, ground reflected radiation and extreme weather received in each interval is related to the type and intensity of extreme weather. 、 and They are The direct radiation, diffuse radiation and total radiation received on the ground in each interval, is the tilt angle of the photovoltaic panel, 、 、 and They are the correction factors for extreme weather direct radiation, solar azimuth, diffuse radiation and ground reflected radiation, and their values ​​are determined according to the type and intensity of extreme weather. is the ground albedo, is the solar hour angle, is the latitude of the photovoltaic power station, is the solar declination; S52, based on the total solar radiation received , the photovoltaic power correction factor is used to consider the impact of extreme weather on photovoltaic power, and the photovoltaic output under extreme weather scenarios is calculated. ; Where, is the conversion efficiency of the photovoltaic system, is the area of ​​the photovoltaic panel, is the total solar radiation received by the photovoltaic system, is the extreme weather photovoltaic power correction factor, and its value is determined according to the type and intensity of extreme weather.

Images