Photovoltaic output power prediction method and device based on sandstorm duration

Abstract

The present invention discloses a method and device for predicting photovoltaic output power based on the duration of sandstorms. The method comprises: collecting historical data sequences, calculating the rate of change of dust concentration and wind speed, and training a prediction model using training data. The prediction model predicts the current duration of the sandstorm, as well as future data sequences of dust concentration and wind speed; calculating the dust deposition rate on the photovoltaic panel at future moments using the inclination angle and sand particle size of the photovoltaic panel, combined with the dust concentration and wind speed at future moments; using the dust deposition rate, combined with the dust accumulation attenuation coefficient, calculating the amount of dust accumulated on the photovoltaic panel at multiple moments during the duration of the sandstorm; and calculating the corresponding photovoltaic panel photoelectric conversion efficiency based on the dust accumulation amount, thereby calculating the photovoltaic output power. Using the above technical solution, the photovoltaic output power during the duration of the sandstorm can be accurately calculated.

Description

Technical Field

[0001] The present invention relates to the field of new energy power grid monitoring and control technology, and in particular to a photovoltaic output power prediction method and device based on the duration of sandstorm weather. Background Art

[0002] Current distribution network research aims to improve grid stability. With the large-scale integration of renewable energy generation, its fluctuating and intermittent output will exacerbate uncertainty in distribution network operations. Meteorological factors are a key variable in photovoltaic output forecasting. However, photovoltaic and other new energy equipment are often located in remote areas, which are often subject to unusually severe weather conditions, such as sandstorms.

[0003] Dust storms directly impact the energy conversion process of photovoltaic modules, and therefore their output power. Since most renewable energy generation units are now connected to the grid, significant fluctuations in their output power can adversely impact the grid's stability. Therefore, grid control systems predict photovoltaic output power to adapt to fluctuations.

[0004] However, among the photovoltaic output power prediction schemes used in existing technologies, there are few schemes for forecasting severe weather. They mainly use historical meteorological data as training data for prediction models, and do not distinguish between severe weather conditions. The output results obtained are difficult to accurately describe the photovoltaic power output when severe weather (such as sandstorms) occurs. Deviations in the output power prediction results will affect the stable operation of the power grid and the normal power consumption of surrounding electrical equipment. Summary of the Invention

[0005] Purpose of the invention: The present invention provides a method and device for predicting photovoltaic output power based on the duration of sandstorm weather, aiming to solve the technical problem of low accuracy in predicting photovoltaic output power under sandstorm weather in the prior art.

[0006] Technical solution: The present invention provides a photovoltaic output power prediction method based on the duration of sandstorm weather, comprising: collecting historical data sequences of dust concentration and wind speed, and calculating the dust concentration change rate and wind speed change rate at historical moments; using the dust concentration change rate and wind speed change rate at historical moments as training data to train a prediction model, inputting the collected current dust concentration change rate and wind speed change rate into the prediction model for prediction, and predicting the current sandstorm weather duration, as well as future data sequences of dust concentration and wind speed; calculating the dust deposition rate on the photovoltaic panel at future moments through the inclination angle and sand particle size of the photovoltaic panel, in combination with the dust concentration and wind speed at future moments; using the dust deposition rate, in combination with the dust accumulation attenuation coefficient, calculating the dust accumulation amount on the photovoltaic panel at multiple moments during the duration of the sandstorm weather one by one; the dust accumulation attenuation coefficient is positively correlated with the existing dust accumulation amount on the photovoltaic panel; calculating the corresponding photoelectric conversion rate of the photovoltaic panel through the dust accumulation amount, and obtaining the photovoltaic output power at multiple moments during the duration of the sandstorm weather through the light intensity and the light collection area of ​​the photovoltaic panel.

[0007] Specifically, the rate of change of sand and dust concentration at the selected moment is calculated by the first average sand and dust concentration, the second average sand and dust concentration and the corresponding time interval; the first average sand and dust concentration is the average value of the sand and dust concentration change rates of at least three adjacent moments before the selected moment, the second average sand and dust concentration is the average value of the sand and dust concentration change rates of at least two adjacent moments after the selected moment, the time interval between the first average sand and dust concentration and the second average sand and dust concentration is the interval between the middle moment of the first average sand and dust concentration and the middle moment of the second average sand and dust concentration; the wind speed change rate at the selected moment is obtained by the deviation between the wind speed at the selected moment and the historical average wind speed.

[0008] Specifically, the dust concentration change rate is calculated using the following formula:

[0009] E t =((C t +C t+1 +C t+2 )-(C t-1 +C t-2 +C t-3 )) / (t t+1 -t t-2 ),

[0010] Among them, E t represents the dust concentration change rate at time t, C t-3 to C t+2 They represent the dust concentration from time t-3 to time t+2, t t+1 and t t-2Represent the time at time t+1 and time t-2 respectively;

[0011] The wind speed change rate is calculated using the following formula:

[0012] S t =v t -v',

[0013] Among them, S t represents the wind speed change rate at time t, v t represents the wind speed at time t, and v' represents the historical average wind speed.

[0014] Specifically, the duration of sandstorm weather, the change rate of sandstorm concentration and the change rate of wind speed are used as training data to train the weather duration prediction model; the sandstorm concentration data sequence and the sandstorm concentration change rate are used as training data to train the concentration prediction model, and the wind speed data sequence and the wind speed change rate are used as training data to train the wind speed prediction model.

[0015] Specifically, the dust deposition rate on the photovoltaic panels is calculated based on the following parameter relationships: wind speed and dust concentration are positively correlated with the dust deposition rate, the tilt angle of the photovoltaic panels is negatively correlated with the dust deposition rate, when the average sand particle size is within the standard value, it is positively correlated with the dust deposition rate, and when the average sand particle size is greater than the standard value, it is negatively correlated with the dust deposition rate.

[0016] Specifically, the dust deposition rate on the photovoltaic panel is calculated using the following formula:

[0017] Y t =C t v t (k b / e tanθ ),

[0018] Among them, Y t represents the dust deposition rate at time t, C t represents the dust concentration at time t, v t represents the wind speed at time t, e represents the natural logarithm, θ represents the tilt angle of the photovoltaic panel, k represents the particle size sedimentation coefficient, k>1, b represents the corresponding value parameter of the average sand particle size d, when d is less than or equal to the standard value, b=(nd+1) 2 , when d is greater than the standard value, b=1 / (nd+1) 2 , n represents the particle size influence coefficient.

[0019] Specifically, the dust accumulation amount at the first moment is obtained based on the dust settling rate and the corresponding accumulation time, and the dust accumulation amount at the second moment is calculated through the negative influence of the dust accumulation amount at the first moment and the dust accumulation attenuation coefficient on the dust accumulation amount, as well as the accumulation time between the second moment and the first moment, until the dust accumulation amount on the photovoltaic panel at multiple moments during the duration of the sandstorm weather is calculated; the dust accumulation attenuation coefficient determines the degree of influence of the dust settling rate, the accumulated dust accumulation amount and the inclination angle of the photovoltaic panel on the dust accumulation amount.

[0020] Specifically, the amount of dust accumulation on the photovoltaic panels is calculated using the following formula:

[0021] M1=Y1t1,

[0022] M t =Y t M t - (t t -t t-1 )(fY t-1 M t-1 sinθ), t>1,

[0023] Among them, M1 and M t Y1 and Y2 represent the dust accumulation at time 1 and time t, respectively. t-1 represent the dust deposition rates at time 1 and time t-1, respectively. t-1 and t t They represent the time at time 1, time t-1 and time t respectively, and f represents the dust accumulation attenuation coefficient.

[0024] Specifically, historical data on dust accumulation under unit light intensity and unit light collection area and the corresponding photovoltaic output power of photovoltaic panels are obtained, and a photoelectric conversion rate function between dust accumulation and photovoltaic output power is obtained by fitting;

[0025] The photovoltaic output power at the future time is calculated using the following formula:

[0026] P t =O(M t )qh t ,

[0027] Among them, P t represents the photovoltaic output power at time t, O(M t ) represents the photoelectric conversion rate function, M t represents the amount of dust accumulation at time t, q represents the light collection area of ​​the photovoltaic panel, and h t represents the light intensity at time t.

[0028] The present invention also provides a photovoltaic output power prediction device based on the duration of sandstorm weather, comprising: a historical data acquisition and processing unit, a future data prediction unit, a sand and dust deposition rate calculation unit, a sand and dust accumulation amount calculation unit and a power prediction unit, wherein: the historical data acquisition and processing unit is used to collect the historical data sequence of sand and dust concentration and the historical data sequence of wind speed, and calculate the sand and dust concentration change rate and wind speed change rate at the historical moment; the future data prediction unit is used to extract the sand and dust concentration change rate feature and the wind speed change rate feature from the sand and dust concentration change rate and wind speed change rate at the historical moment, and use the extracted features to make predictions to predict the duration of sandstorm weather and the future dust concentration. data sequence and future data sequence of wind speed; the dust deposition rate calculation unit is used to calculate the dust deposition rate on the photovoltaic panel at a future moment by using the inclination angle of the photovoltaic panel and the sand particle size, in combination with the dust concentration and wind speed at a future moment; the dust accumulation amount calculation unit is used to use the dust deposition rate, in combination with the dust accumulation attenuation coefficient, to calculate the dust accumulation amount on the photovoltaic panel at multiple moments during the duration of the sandstorm weather; the dust accumulation amount attenuation coefficient is positively correlated with the existing dust accumulation amount on the photovoltaic panel; the power prediction unit is used to obtain the corresponding photoelectric conversion rate of the photovoltaic panel through the dust accumulation calculation, and obtain the photovoltaic output power at a future moment by combining the light intensity and the light collection area of ​​the photovoltaic panel.

[0029] Beneficial effects: Compared with the existing technology, the present invention has the following significant advantages: using historical data on sandstorms to predict the duration, dust concentration and wind speed of sandstorms, combining factors such as the inclination angle of the photovoltaic panels and the particle size of the sand particles to calculate the sedimentation rate of the sand and dust on the photovoltaic panels, considering the influence of the amount of sand and dust accumulation on the photovoltaic panels on the conversion of light into output power, thereby accurately calculating the photovoltaic output power during the duration of the sandstorm. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of the photovoltaic output power prediction method provided by the present invention. DETAILED DESCRIPTION

[0031] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0032] See Figure 1 , which is a flow chart of the photovoltaic output power prediction method provided by the present invention.

[0033] In the embodiment of the present invention, a historical data series of dust concentration and a historical data series of wind speed are collected, and the dust concentration change rate and the wind speed change rate at historical moments are calculated.

[0034] In specific implementation, the dust concentration change rate and the wind speed change rate represent the change amplitude of the dust concentration and wind speed at the corresponding moment. The dust concentration and wind speed are the most intuitive factors that reflect the severity of sandstorms. Therefore, the present invention uses dust concentration and wind speed as the indicators to judge the duration of sandstorms. More importantly, the dust concentration and wind speed are also related to the dust deposition rate on the photovoltaic panels. The dust deposition rate has an important influence on the energy conversion rate of the photovoltaic panels. Therefore, the dust concentration and wind speed are important basic parameters in the present invention.

[0035] In a specific implementation, the data sequence represents relevant data (sand and dust concentration, wind speed) at multiple moments arranged in chronological order.

[0036] In an embodiment of the present invention, the rate of change of sand and dust concentration at a selected moment is calculated by using the first average sand and dust concentration, the second average sand and dust concentration, and the corresponding time interval; the first average sand and dust concentration is the average value of the rates of change of sand and dust concentration at at least three adjacent moments before the selected moment, the second average sand and dust concentration is the average value of the rates of change of sand and dust concentration at and after the selected moment, and the time interval between the first average sand and dust concentration and the second average sand and dust concentration is the interval between the middle moment of the first average sand and dust concentration and the middle moment of the second average sand and dust concentration.

[0037] In specific implementation, when calculating the sand and dust concentration change rate at a certain moment (selected moment), considering that the sand and dust concentration collection point is at a fixed position, if the wind speed at this position changes instantaneously at the corresponding moment, it may cause the sand and dust concentration to also change significantly. Therefore, in order to more accurately calculate the sand and dust concentration change rate at the selected moment, the sand and dust concentration at multiple moments can be expanded forward and backward. The sand and dust concentration change rate calculated in this way can more accurately represent the actual sand and dust concentration change rate at the selected moment.

[0038] In specific implementations, considering that the time intervals between adjacent moments are often the same, the middle moment of the first (second) average dust concentration refers to the time of the middle moment among the three moments. It can also be further calculated according to the average time corresponding to the average dust concentration, for example, the average time of three (or more) moments corresponding to the first (second) average dust concentration.

[0039] In the embodiment of the present invention, the dust concentration change rate is calculated using the following formula:

[0040] E t =((C t +C t+1 +C t+2 )-(C t-1 +C t-2 +C t-3 )) / (t t+1 -tt-2 ),

[0041] Among them, E t represents the dust concentration change rate at time t, C t-3 to C t+2 They represent the dust concentration from time t-3 to time t+2, t t+1 and t t-2 They represent the time at time t+1 and time t-2 respectively.

[0042] In a specific implementation, time t+1 is the middle time between time t and time t+2, and time t-2 is the middle time between time t-1 and time t-3.

[0043] In the embodiment of the present invention, the wind speed change rate at the selected moment is obtained by the deviation between the wind speed at the selected moment and the historical average wind speed.

[0044] In specific implementation, in remote non-urban areas where photovoltaic systems are installed, wind exists all year round, and there are very few cases of no wind or low wind. Therefore, it can be regarded as that there is a basic wind speed in the area. In such a case, if only the amplitude of the wind speed change before and after the moment is calculated, if the wind speed does not change much before and after the moment (it may be faster or slower), the result of a smaller wind speed change amplitude is obtained. However, this cannot reflect the difference between the wind speed in sandstorm weather and the basic wind speed in normal weather, and thus the wind speed characteristics in sandstorm weather cannot be effectively obtained. Therefore, in order to reflect the actual wind speed change rate at the selected moment, the present invention subtracts the wind speed at the selected moment from the historical average wind speed in sandstorm weather to obtain the accurate change of wind speed.

[0045] In the embodiment of the present invention, the wind speed change rate is calculated using the following formula:

[0046] S t =v t -v',

[0047] Among them, S t represents the wind speed change rate at time t, v t represents the wind speed at time t, and v' represents the historical average wind speed.

[0048] In specific implementation, the historical average wind speed refers to the historical average wind speed under sandstorm weather.

[0049] In an embodiment of the present invention, dust concentration change rate features and wind speed change rate features are extracted from the dust concentration change rates and wind speed change rates at historical moments, the extracted features are used to train a prediction model, and the collected current dust concentration change rate and wind speed change rate are input into the prediction model for prediction, so as to predict the duration of the current sandstorm weather, as well as the future data sequence of dust concentration and the future data sequence of wind speed.

[0050] In specific implementation, the sand and dust concentration change rate and wind speed change rate can reflect the changes in sandstorm weather. When the sand and dust concentration and wind speed are high, the sandstorm weather is in a more severe state. When the sand and dust concentration and wind speed are low, the sandstorm weather is in a milder state. Therefore, the model trained by the sand and dust concentration change rate characteristics and the wind speed change rate characteristics can predict the current sandstorm weather duration. At the same time, the model trained by the sand and dust concentration change rate characteristics can obtain the future data sequence of sand and dust concentration, and the model trained by the wind speed change rate characteristics can obtain the future data sequence of wind speed.

[0051] In practice, the current dust concentration and wind speed change rates input into the prediction model are based on data collected during the current dust storm, calculated using the formulas described in this disclosure. Generally speaking, the more sufficient the input data for dust concentration and wind speed change rates, the more accurate the model's prediction results.

[0052] In an embodiment of the present invention, the duration of sandstorm weather, the rate of change of sandstorm concentration, and the rate of change of wind speed are used as training data to train a weather duration prediction model; the sandstorm concentration data sequence and the rate of change of sandstorm concentration are used as training data to train a concentration prediction model; and the wind speed data sequence and the rate of change of wind speed are used as training data to train a wind speed prediction model.

[0053] In practice, the present invention involves three prediction models: a weather duration prediction model, a concentration prediction model, and a wind speed prediction model. The weather duration prediction model uses the dust concentration change rate and wind speed change rate as input data, with the dust weather duration (historical data) as labels. The concentration prediction model uses the dust concentration change rate as input data, with dust concentration data sequences as labels. The wind speed prediction model uses the wind speed change rate as input data, with wind speed data sequences as labels.

[0054] In the embodiment of the present invention, the dust settling rate on the photovoltaic panel at a future time is calculated by using the tilt angle of the photovoltaic panel and the sand particle size, in combination with the dust concentration and wind speed at a future time.

[0055] In specific implementations, the sedimentation rate of dust on photovoltaic panels is not only affected by dust concentration and wind speed, but also by the inclination angle of the photovoltaic panels and the particle size of the sand particles. Therefore, the present invention also incorporates these two factors into the calculation to improve the accuracy of the final result.

[0056] In an embodiment of the present invention, the dust deposition rate on the photovoltaic panel is calculated based on the following parameter relationships: wind speed and dust concentration are positively correlated with the dust deposition rate, the inclination angle of the photovoltaic panel is negatively correlated with the dust deposition rate, when the average sand particle size is within the standard value, it is positively correlated with the dust deposition rate, and when the average sand particle size is greater than the standard value, it is negatively correlated with the dust deposition rate.

[0057] In practice, higher wind speeds and dust concentrations increase the dust settling rate, and vice versa. The photovoltaic panel's tilt angle refers to the angle between the panel and the horizontal. The greater the tilt angle, the easier it is for dust to slide off the panel. Therefore, the greater the tilt angle, the lower the dust settling rate. The average sand particle size is a special case. When the average sand particle size is within (and including) the standard value (which can be set based on actual application scenarios), the larger the particle size, the greater the friction between the particle and the photovoltaic panel. This also increases the friction between dust and other particles, making them more susceptible to wind-induced deposition on the panel. Therefore, the larger the particle size, the higher the dust settling rate. However, when the average sand particle size is greater than the standard value, the larger the particle size, the heavier the dust, making it more likely to slide off the photovoltaic panel. Friction is insufficient to hold it on the panel, and it is also more difficult for it to aggregate with other dust particles. Therefore, the larger the particle size, the lower the dust settling rate.

[0058] In the embodiment of the present invention, the dust deposition rate on the photovoltaic panel is calculated using the following formula:

[0059] Y t =C t v t (k b / e tanθ ),

[0060] Among them, Y t represents the dust deposition rate at time t, C t represents the dust concentration at time t, v t represents the wind speed at time t, e represents the natural logarithm, θ represents the tilt angle of the photovoltaic panel, tanθ represents the tangent value of the tilt angle, k represents the particle size sedimentation coefficient (which can be set according to the actual application scenario), k>1, b represents the corresponding value parameter of the average sand particle size d, when d is less than or equal to the standard value, b=(nd+1) 2 , when d is greater than the standard value, b=1 / (nd+1) 2 , n represents the particle size influence coefficient (which can be set according to the actual application scenario).

[0061] In the embodiment of the present invention, the dust settling rate is used in combination with the dust accumulation attenuation coefficient to calculate the amount of dust accumulation on the photovoltaic panel at multiple moments during the duration of the dust weather.

[0062] In the embodiment of the present invention, the dust accumulation attenuation coefficient is positively correlated with the amount of dust accumulation already on the photovoltaic panel.

[0063] In specific implementations, photovoltaic panels are usually set at an angle, so the dust accumulated on the photovoltaic panels has a tendency to slide down. Therefore, when calculating the amount of dust accumulation on the photovoltaic panels at each moment, the dust accumulation attenuation coefficient needs to be taken into account. The dust accumulation attenuation coefficient represents the degree of dust sliding.

[0064] In an embodiment of the present invention, the amount of dust accumulation at the first moment is obtained based on the dust settling rate and the corresponding cumulative time. The amount of dust accumulation at the first moment and the negative influence of the dust accumulation attenuation coefficient on the dust accumulation, as well as the cumulative time between the second moment and the first moment, are used to calculate the amount of dust accumulation at the second moment, until the calculation of the amount of dust accumulation on the photovoltaic panel at multiple moments during the duration of the sandstorm weather is completed.

[0065] In specific implementation, the amount of dust accumulation can be calculated by multiplying the dust settling rate and the corresponding cumulative time. When considering the amount of dust accumulation at subsequent adjacent moments, the dust sliding off the photovoltaic panels needs to be included in the calculation. Therefore, the amount of dust accumulation at the second moment needs to consider the amount of dust accumulation and the dust accumulation attenuation coefficient at the first moment. The amount of dust accumulation at the third moment needs to consider the amount of dust accumulation and the dust accumulation attenuation coefficient at the second moment, and so on.

[0066] In the embodiment of the present invention, the dust accumulation attenuation coefficient determines the degree of influence of the dust settling rate, the accumulated dust accumulation amount, and the tilt angle of the photovoltaic panel on the dust accumulation amount.

[0067] In specific implementation, the dust accumulation attenuation coefficient, as a parameter in the calculation, is actually an adjustment to the degree of influence of factors affecting the amount of dust accumulation. The influencing factors mainly involve the dust settling rate, the accumulated amount of dust accumulation and the tilt angle of the photovoltaic panel, and of course the time interval between adjacent moments.

[0068] In the embodiment of the present invention, the following formula is used to calculate the amount of dust accumulation on the photovoltaic panel:

[0069] M1=Y1t1,

[0070] M t =Y t M t - (t t -t t-1 )(fY t-1 M t-1 sinθ), t>1,

[0071] Among them, M1 and M tY1 and Y2 represent the dust accumulation at time 1 and time t, respectively. t-1 represent the dust deposition rates at time 1 and time t-1, respectively. t-1 and t t They represent the time at moment 1, moment t-1 and moment t respectively, f represents the dust accumulation attenuation coefficient, and the moment 1 and moment t in the formula mainly represent future moments, that is, the predicted data.

[0072] In an embodiment of the present invention, the photoelectric conversion rate of the corresponding photovoltaic panel is calculated by calculating the amount of sand and dust accumulation, and the photovoltaic output power at multiple moments during the duration of the sandstorm weather is obtained by combining the light intensity and the light collection area of ​​the photovoltaic panel.

[0073] In practice, dust accumulation on photovoltaic panels affects their power generation, effectively affecting the amount of sunlight they can collect. The photoelectric conversion rate represents the relationship between dust accumulation and photovoltaic output power. This relationship can be calculated using historical data, allowing the corresponding photovoltaic panel output power to be calculated based on dust accumulation.

[0074] In an embodiment of the present invention, historical data of dust accumulation under unit light intensity and unit light collection area and the corresponding photovoltaic output power of the photovoltaic panel are obtained, and a photoelectric conversion rate function between the dust accumulation and the photovoltaic output power is fitted.

[0075] In specific implementation, multiple dust accumulation amounts and corresponding multiple photovoltaic output powers (unit intensity of light and unit area of ​​light collection) can be collected in historical data to obtain the conversion value between each pair of dust accumulation amount and photovoltaic output power. By fitting multiple conversion values, the photoelectric conversion rate function can be obtained, which is used to calculate the conversion relationship between dust accumulation amount and output power.

[0076] In the embodiment of the present invention, the photovoltaic output power at a future time is calculated using the following formula:

[0077] P t =O(M t )qh t ,

[0078] Among them, P t represents the photovoltaic output power at time t, O(M t ) represents the photoelectric conversion rate function, M t represents the amount of dust accumulation at time t, q represents the light collection area of ​​the photovoltaic panel, and h t represents the light intensity at time t.

[0079] The present invention also provides a photovoltaic output power prediction device based on the duration of sandstorm weather, comprising: a historical data acquisition and processing unit, a future data prediction unit, a sand and dust deposition rate calculation unit, a sand and dust accumulation amount calculation unit and a power prediction unit, wherein: the historical data acquisition and processing unit is used to collect the historical data sequence of sand and dust concentration and the historical data sequence of wind speed, and calculate the sand and dust concentration change rate and wind speed change rate at the historical moment; the future data prediction unit is used to extract the sand and dust concentration change rate feature and the wind speed change rate feature from the sand and dust concentration change rate and wind speed change rate at the historical moment, and use the extracted features to make predictions to predict the duration of sandstorm weather and the future dust concentration. data sequence and future data sequence of wind speed; the dust deposition rate calculation unit is used to calculate the dust deposition rate on the photovoltaic panel at a future moment by using the inclination angle of the photovoltaic panel and the sand particle size, in combination with the dust concentration and wind speed at a future moment; the dust accumulation amount calculation unit is used to use the dust deposition rate, in combination with the dust accumulation attenuation coefficient, to calculate the dust accumulation amount on the photovoltaic panel at multiple moments during the duration of the sandstorm weather; the dust accumulation amount attenuation coefficient is positively correlated with the existing dust accumulation amount on the photovoltaic panel; the power prediction unit is used to obtain the corresponding photoelectric conversion rate of the photovoltaic panel through the dust accumulation calculation, and obtain the photovoltaic output power at a future moment by combining the light intensity and the light collection area of ​​the photovoltaic panel.

[0080] In a specific implementation, the photovoltaic output power prediction device based on the duration of sandstorm weather provided by the present invention, wherein the execution unit for executing functions, steps or methods, the functions, steps or methods executed by it can refer to the functions, steps or methods in the photovoltaic output power prediction method based on the duration of sandstorm weather provided by the present invention.

Claims

1. A photovoltaic output power prediction method based on the duration of sandstorm weather, characterized in that: include: Collect historical data series of dust concentration and wind speed, and calculate the dust concentration change rate and wind speed change rate at historical moments; The prediction model is trained using the historical dust concentration change rate and wind speed change rate as training data. The current dust concentration change rate and wind speed change rate collected are input into the prediction model for prediction. The prediction results include the duration of the current dust weather, as well as the future data series of dust concentration and wind speed. The dust deposition rate on the photovoltaic panels at future times is calculated by combining the tilt angle of the photovoltaic panels and the sand particle size with the dust concentration and wind speed at future times. The dust deposition rate on the photovoltaic panels is calculated based on the following parameter relationships: wind speed and dust concentration are positively correlated with the dust deposition rate, the tilt angle of the photovoltaic panels is negatively correlated with the dust deposition rate, and when the average sand particle size is within the standard value, it is positively correlated with the dust deposition rate, and when the average sand particle size is greater than the standard value, it is negatively correlated with the dust deposition rate. The dust deposition rate on the photovoltaic panels is calculated using the following formula: Among them, Y t represents the dust deposition rate at time t, C t represents the dust concentration at time t, v t represents the wind speed at time t, e represents the natural logarithm, θ represents the tilt angle of the photovoltaic panel, k represents the particle size sedimentation coefficient, k>1, b represents the corresponding value parameter of the average sand particle size d, when d is less than or equal to the standard value, b=(nd+1) 2 , when d is greater than the standard value, b=1 / (nd+1) 2 , n represents the particle size influence coefficient; The dust accumulation amount at the first moment is obtained based on the dust settling rate and the corresponding accumulation time. The dust accumulation amount at the second moment is calculated based on the dust accumulation amount at the first moment and the negative influence of the dust accumulation attenuation coefficient on the dust accumulation amount, as well as the accumulation time between the second moment and the first moment, until the dust accumulation amount on the photovoltaic panel at multiple moments during the duration of the sandstorm is calculated. The dust accumulation attenuation coefficient determines the degree of influence of the dust settling rate, the accumulated dust accumulation amount, and the tilt angle of the photovoltaic panel on the dust accumulation amount. The dust accumulation amount on the photovoltaic panel is calculated using the following formula: Among them, M1 and M t Y1 and Y2 represent the dust accumulation at time 1 and time t, respectively. t-1 represent the dust deposition rates at time 1 and time t-1, respectively. t-1 and t t represent the time at time 1, time t-1 and time t respectively, and f represents the dust accumulation attenuation coefficient; The photoelectric conversion rate of the corresponding photovoltaic panel is calculated by calculating the amount of sand and dust accumulation. Combined with the light intensity and the light collection area of ​​the photovoltaic panel, the photovoltaic output power at multiple moments during the duration of the sandstorm weather is obtained.

2. The photovoltaic output power prediction method based on the duration of sandstorm weather according to claim 1 is characterized in that: The calculation of the dust concentration change rate and wind speed change rate at a historical moment includes: The rate of change of dust concentration at a selected moment is calculated using the first average dust concentration, the second average dust concentration, and the corresponding time intervals; the first average dust concentration is the average of the rates of change of dust concentration at at least three adjacent moments before the selected moment; the second average dust concentration is the average of the rates of change of dust concentration at and after the selected moment; and the time interval between the first average dust concentration and the second average dust concentration is the interval between the midpoint of the first average dust concentration and the midpoint of the second average dust concentration. The rate of change of wind speed at a selected moment is obtained by the deviation between the wind speed at the selected moment and the historical average wind speed.

3. The photovoltaic output power prediction method based on the duration of sandstorm weather according to claim 2 is characterized in that: The calculation of the dust concentration change rate and wind speed change rate at a historical moment includes: The dust concentration change rate is calculated using the following formula: Among them, E t represents the dust concentration change rate at time t, C t-3 to C t+2 They represent the dust concentration from time t-3 to time t+2, t t+1 and t t-2 Represent the time at time t+1 and time t-2 respectively; The wind speed change rate is calculated using the following formula: Among them, S t represents the wind speed change rate at time t, v t represents the wind speed at time t, and v' represents the historical average wind speed.

4. The photovoltaic output power prediction method based on the duration of sandstorm weather according to claim 3 is characterized in that: The method of training the prediction model using the dust concentration change rate and wind speed change rate at historical moments as training data includes: The duration of sandstorm weather, the change rate of sandstorm concentration and the change rate of wind speed are used as training data to train the weather duration prediction model; the sandstorm concentration data sequence and the sandstorm concentration change rate are used as training data to train the concentration prediction model, and the wind speed data sequence and the wind speed change rate are used as training data to train the wind speed prediction model.

5. The photovoltaic output power prediction method based on sandstorm duration according to claim 1, characterized in that: The calculation of the photoelectric conversion rate of the corresponding photovoltaic panel based on the amount of dust accumulation includes: Obtain historical data on dust accumulation under unit light intensity and unit light collection area, and the corresponding photovoltaic output power of photovoltaic panels, and fit the photoelectric conversion rate function between dust accumulation and photovoltaic output power; The photovoltaic output power at multiple moments during the duration of the sandstorm weather is obtained by combining the light intensity and the light collection area of ​​the photovoltaic panel, including: The photovoltaic output power at the future time is calculated using the following formula: Among them, P t represents the photovoltaic output power at time t, O(M t ) represents the photoelectric conversion rate function, M t represents the amount of dust accumulation at time t, q represents the light collection area of ​​the photovoltaic panel, h t represents the light intensity at time t.

6. A photovoltaic output power prediction device based on the duration of sandstorm weather, characterized in that: include: Historical data collection and processing unit, future data prediction unit, sand and dust deposition rate calculation unit, sand and dust accumulation calculation unit and power prediction unit, including: The historical data collection and processing unit is used to collect historical data series of dust concentration and wind speed, and calculate the dust concentration change rate and wind speed change rate at historical moments; The future data prediction unit is used to extract dust concentration change rate characteristics and wind speed change rate characteristics from the dust concentration change rate and wind speed change rate at historical moments, and use the extracted characteristics to perform prediction to obtain the duration of the sandstorm weather, as well as the future data sequence of dust concentration and the future data sequence of wind speed; The dust deposition rate calculation unit is used to calculate the dust deposition rate on the photovoltaic panel at a future time by using the inclination angle of the photovoltaic panel and the sand particle size, in combination with the dust concentration and wind speed at the future time; the dust deposition rate on the photovoltaic panel is calculated based on the following parameter relationship: the wind speed and dust concentration are positively correlated with the dust deposition rate, the inclination angle of the photovoltaic panel is negatively correlated with the dust deposition rate, when the average sand particle size is within the standard value, it is positively correlated with the dust deposition rate, and when the average sand particle size is greater than the standard value, it is negatively correlated with the dust deposition rate; the dust deposition rate on the photovoltaic panel is calculated using the following formula: Among them, Y t represents the dust deposition rate at time t, C t represents the dust concentration at time t, v t represents the wind speed at time t, e represents the natural logarithm, θ represents the tilt angle of the photovoltaic panel, k represents the particle size sedimentation coefficient, k>1, b represents the corresponding value parameter of the average sand particle size d, when d is less than or equal to the standard value, b=(nd+1) 2 , when d is greater than the standard value, b=1 / (nd+1) 2 , n represents the particle size influence coefficient; The dust accumulation amount calculation unit is used to obtain the dust accumulation amount at a first moment based on the dust settling rate and the corresponding accumulation time, and calculate the dust accumulation amount at a second moment based on the dust accumulation amount at the first moment and the negative influence of the dust accumulation attenuation coefficient on the dust accumulation amount, as well as the accumulation time between the second moment and the first moment, until the dust accumulation amount on the photovoltaic panel at multiple moments during the duration of the sandstorm is calculated; the dust accumulation attenuation coefficient determines the degree of influence of the dust settling rate, the accumulated dust accumulation amount, and the inclination angle of the photovoltaic panel on the dust accumulation amount; the dust accumulation amount on the photovoltaic panel is calculated using the following formula: Among them, M1 and M t Y1 and Y2 represent the dust accumulation at time 1 and time t, respectively. t-1 represent the dust deposition rates at time 1 and time t-1, respectively. t-1 and t t represent the time at time 1, time t-1 and time t respectively, and f represents the dust accumulation attenuation coefficient; The power prediction unit is used to calculate the photoelectric conversion rate of the corresponding photovoltaic panel through the amount of sand and dust accumulation, and to obtain the photovoltaic output power at a future moment in time by combining the light intensity and the light collection area of ​​the photovoltaic panel.

Images