A photovoltaic output prediction method based on micro-topography influence

Abstract

The photovoltaic output prediction method based on the influence of micro-topography provided by the present invention includes step S1, obtaining the latitude and longitude, weather, altitude and other data of the photovoltaic power station; step S2, considering the influence of micro-topography on photovoltaic output through direct and indirect channels, multi-level clustering of the data obtained in step S1, and constructing a photovoltaic output prediction model; step S3, monitoring the meteorological data of the photovoltaic power station, and calculating the current photovoltaic output prediction value according to the photovoltaic output prediction model; step S4, collecting historical photovoltaic outputs of different categories according to the multi-level clustering method of step S2, comparing the photovoltaic output prediction value obtained in step S3 with the historical photovoltaic output of the same type, and optimizing the photovoltaic output prediction model according to the comparison result; step S5, using the optimized photovoltaic output prediction model to predict the photovoltaic output of the photovoltaic power station. The present invention fully considers the influence of micro-topography on photovoltaic output and improves the accuracy of photovoltaic output prediction.

Description

Technical Field

[0001] The present invention relates to the field of new energy technology, and in particular to a photovoltaic output prediction method based on micro-topography influence. Background Art

[0002] As the global demand for renewable energy continues to grow, photovoltaic power generation has been widely used as a clean and sustainable form of energy. The power generation capacity of photovoltaic power stations is directly affected by environmental factors, especially meteorological conditions. Factors such as radiation intensity, temperature, and wind speed play an important role in the working efficiency and power generation performance of photovoltaic modules. However, in addition to traditional meteorological factors, micro-topography (such as terrain undulations, slope, and slope direction within a small area) also has a significant impact on the output of photovoltaic power stations.

[0003] Micro-topography influence refers to the regulatory effect of terrain on micro-meteorological conditions at a small scale (such as local ups and downs, wind ducts, shadows, etc. inside the power station). This regulatory effect can change the radiation distribution, temperature distribution, local wind speed and other meteorological factors of the photovoltaic power station, thereby affecting the performance of photovoltaic modules. Although the micro-topography effect has an important influence on the design and operation of photovoltaic power stations, most of the existing photovoltaic output prediction methods ignore it, usually only considering the overall meteorological conditions or large-scale terrain features (such as overall slope, altitude, etc.). Although this method can provide a reference for rough output prediction, it lacks sufficient consideration of the micro-topography effect in high-precision photovoltaic output prediction.

[0004] In photovoltaic power stations, the impact of micro-topography on photovoltaic power stations is mainly manifested in three aspects. First, the shielding effect, such as the shadows of surrounding obstacles or highlands, may cause insufficient radiation received by photovoltaic modules; second, the uneven distribution of radiation and the undulating terrain will cause different light intensities in different areas, thus affecting the power generation; finally, local meteorological changes, such as temperature and wind speed, may vary in different terrain areas, which will affect the working efficiency and heat dissipation performance of photovoltaic modules. Therefore, the impact of micro-topography on photovoltaic output is a complex local effect, and traditional photovoltaic output prediction methods often find it difficult to accurately capture such details, resulting in limited prediction accuracy. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a photovoltaic output prediction method based on micro-topography influence, which fully considers the influence of micro-topography on photovoltaic output and improves the accuracy of photovoltaic output prediction.

[0006] In order to solve the above technical problems, the present invention adopts the following technical methods: a photovoltaic output prediction method based on micro-topography influence, comprising:

[0007] Step S1, obtaining the terrain and meteorological data of the photovoltaic power station, including latitude and longitude, weather, and altitude, and normalizing the data;

[0008] Step S2, considering the direct and indirect effects of micro-topography on photovoltaic output, multi-level clustering is performed on the data obtained in step S1 to construct a photovoltaic output prediction model;

[0009] The multi-level clustering method is as follows:

[0010] The first layer is spatial division based on terrain characteristics: the region is divided into several sub-regions according to the overall terrain type and terrain parameters of the region. The micro-topography characteristics in each sub-region are similar, but the micro-topography of each sub-region is different; the overall terrain types include hills, plains, mountains, and plateaus, and the terrain parameters include altitude, slope, and relative undulation;

[0011] The second layer is the local pattern division based on meteorological variables: on the basis of the first layer of terrain feature division, combined with the distribution characteristics of local meteorological variables, each sub-area is divided into several micro-meteorological pattern categories, and the meteorological characteristics and terrain characteristics in each pattern are similar; the meteorological variables include precipitation, temperature, wind speed, and solar radiation;

[0012] The third layer combines the detailed features of micro-topography and the factors affecting the performance of PV power station components to refine the sub-array division range within the PV power station: Based on the second layer, the influence of the micro-topography undulations, local slope and orientation on light reception is evaluated, the influence of the shielding effect and the thermal effect between devices on the PV performance is considered, and the contribution of the surface albedo to radiation is considered. The PV power generation and temperature changes of each sub-area are analyzed, and each sub-area is further subdivided;

[0013] Step S3, monitoring the meteorological data of the photovoltaic power station, and calculating the current photovoltaic output prediction value according to the photovoltaic output prediction model;

[0014] Step S4, collecting historical photovoltaic outputs of different categories according to the multi-layer clustering method of step S2, comparing the photovoltaic output prediction value obtained in step S3 with the historical photovoltaic outputs of the same category, and optimizing the photovoltaic output prediction model according to the comparison result;

[0015] Step S5, using the optimized photovoltaic output prediction model to predict the photovoltaic output of the photovoltaic power station.

[0016] Furthermore, in step S2, a corresponding photovoltaic output prediction model is constructed according to multi-level clustering to capture the impact of different micro-topography on photovoltaic output. The expression of the photovoltaic output prediction model is as follows:

[0017] (1)

[0018] In the formula, Indicates the predicted value of photovoltaic output; Represents the terrain factors that affect the PV output prediction; Represents the micro-meteorological state factor affected by micro-topography; Indicates photovoltaic panels; Indicates the total number of photovoltaic panels; Indicates The effective area of ​​the photovoltaic panels; Indicates arrival The solar radiation intensity on the surface of a photovoltaic panel; Indicates the function affecting the photovoltaic module;

[0019] (2)

[0020] In the formula, They are respectively expressed as weight coefficients of slope, aspect, altitude and surface albedo; represents the slope factor; represents the orientation factor; represents the altitude factor; represents the surface albedo factor;

[0021] (3)

[0022] In the formula, Respectively expressed as The weight coefficient of Indicates temperature; Indicates humidity; Indicates the wind speed of the micro-topography; represents solar radiation; Indicates the regional base temperature; It represents the vertical temperature lapse rate; Indicates terrain elevation; Slope influence coefficient; Indicates the solar azimuth; Indicates the aspect angle of the ground slope; Average precipitation for the region; represents humidity impact factor; represents the adjustment factor used to control the rate at which humidity changes with altitude; Indicates height; Indicates the height of features associated with terrain; Indicates the reference wind speed when there is no micro-topography influence; Indicates the elevation growth rate coefficient; It indicates the gain or attenuation effect of micro-topography wind speed as the terrain height changes; Indicates the reference radiation intensity when there is no terrain influence;

[0023] (4)

[0024] In the formula, Represents the total area of ​​PV panels; represents the cloud occlusion factor; represents the inter-plate occlusion factor; Represents the coincidence factor. When cloud occlusion and inter-panel occlusion are completely independent, the coincidence factor ,When the cloud occlusion and the inter-board occlusion overlap, the smaller part of the two occlusion ratios is taken as the overlapping part; It represents the length of the cloud layer, which represents the horizontal scale of the cloud layer blocking the photovoltaic panel; Indicates the total time; Indicates the duration of the occlusion; Indicates the wind speed under the combined influence of micro-topography and overall topography; represents the inclination angle of the photovoltaic panel, Indicates the sun's altitude angle. When there is occlusion When , it returns a positive value; when there is no occlusion When , return 0;

[0025] (5)

[0026] In the formula, Indicates the amount of radiation intensity fluctuation; Indicates the maximum radiation intensity;

[0027] (6)

[0028] In the formula, Indicates The reflectivity of the surface of a photovoltaic panel reflects the part of the incident light that is reflected but not absorbed; Indicates The working temperature of photovoltaic modules Efficiency under Indicates Block PV panel system efficiency; Indicates reference temperature Efficiency under represents the temperature coefficient; Indicates the ambient temperature; represents the temperature rise caused by solar radiation; Indicates the temperature rise caused by thermal effects between devices; Indicates the maximum operating temperature of the PV panel.

[0029] Furthermore, in step S3, the prediction time scale of the photovoltaic output of the corresponding micro-topography is set according to the moving speed of the cloud layer in different sub-areas, the meteorological data of the photovoltaic power station is monitored at the set prediction time scale, and the current photovoltaic output prediction value is calculated according to the photovoltaic output prediction model.

[0030] Furthermore, in step S4, the photovoltaic output prediction value obtained in step S3 is compared with the historical photovoltaic output of the same type. If the prediction error is within the threshold range, step S5 is entered; if the prediction error exceeds the threshold, the photovoltaic output prediction model is iteratively optimized and predicted within the maximum optimization number range until the prediction error of the model is within the threshold range and then step S5 is entered. If the maximum optimization number has been exceeded and the prediction error of the model still exceeds the threshold, the warning mode is entered to indicate that abnormal weather exists and human intervention is required for prediction;

[0031] The prediction error is calculated as follows:

[0032] (7)

[0033] In the formula, Represents the prediction error, which is the difference between the predicted PV output and the average historical PV output; Indicates the real-time photovoltaic output forecast value; Indicates the same micro-meteorological conditions in history Average photovoltaic output under ; Indicates the minimum value of the error interval; Indicates the maximum value of the error interval; Indicates the current micro-meteorological status; Represents the current micro-meteorological state function determined by multiple environmental parameters; Indicates air pressure;

[0034] The formula for optimizing the photovoltaic output prediction model is as follows:

[0035] (8)

[0036] In the formula, Respectively represent the predicted photovoltaic output values ​​before and after optimization; represents the adjustment factor; Indicates the maximum number of optimizations set; Indicates the current number of optimizations; Indicates that a warning signal has been triggered.

[0037] Preferably, the method further comprises step S6, using and The two evaluation indicators comprehensively evaluate the prediction results of step S5;

[0038] (9)

[0039] In the formula, represents the root mean square error; represents the mean absolute error; Indicates predicted values; represents the number of samples; represents the target value; Represents the predicted value, which is the predicted value of photovoltaic output.

[0040] Preferably, the wind speed under the combined influence of micro-topography and overall topography is U The expression is as follows:

[0041] (10)

[0042] In the formula, Indicates the wind speed over the overall terrain; Indicates the reference wind speed when there is no terrain influence; Indicates terrain height, the absolute height of terrain relative to sea level; Indicates slope; Indicates the effect of terrain height and slope on wind speed; represents the height influence coefficient; Indicates the target height for wind speed calculation; Indicates the surface roughness length; represents the terrain height influence coefficient; Indicates the reference height, which is the base terrain height of the photovoltaic panel area; Represents the slope influence coefficient.

[0043] The photovoltaic output prediction method based on micro-topography provided by the present invention, through in-depth analysis of the micro-topography effect, performs multi-level clustering on the data related to photovoltaic power generation according to the direct and indirect effects of micro-topography on photovoltaic output, comprehensively considers the influence of various meteorological and micro-topography conditions on photovoltaic output, including meteorological factors such as temperature, wind speed, radiation intensity, and micro-topography factors such as slope, terrain undulation, and shielding effect, thereby establishing a photovoltaic output prediction model, so that the model can fully capture micro-topography information. In this case, the historical photovoltaic output data is used to optimize the model, so that the photovoltaic output prediction accuracy of the model is greatly improved. In summary, compared with the traditional photovoltaic output prediction method, the present invention not only considers meteorological conditions, but also fully considers the influence of micro-topography on photovoltaic output, and the accuracy of photovoltaic output prediction is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1This is a flow chart of a photovoltaic output prediction method based on micro-topography influence according to the present invention;

[0045] Figure 2 It is a flow chart of the multi-level clustering method in the present invention. DETAILED DESCRIPTION

[0046] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with embodiments and drawings. The contents mentioned in the implementation modes are not intended to limit the present invention.

[0047] like Figure 1 As shown, a photovoltaic output prediction method based on micro-topography influence specifically includes the following steps.

[0048] Step S1, obtaining the terrain and meteorological data of the photovoltaic power station, including longitude and latitude, weather, and altitude, and normalizing the data.

[0049] Preferably, the data is normalized and scaled to a uniform range to accelerate the learning process, avoid numerical problems, improve model accuracy, and ensure weight consistency. The normalization formula is as follows:

[0050] (10)

[0051] In the formula, is the value before normalization; is the normalized value; and are the maximum and minimum values ​​of the physical quantity to be normalized.

[0052] Step S2, considering the direct and indirect effects of micro-topography on photovoltaic output, multi-level clustering is performed on the data obtained in step S1 to construct a photovoltaic output prediction model.

[0053] Microtopography affects photovoltaic output through direct and indirect channels. The direct impact is mainly reflected in factors such as solar radiation, temperature and wind speed. For example, hills and trees may block radiation, resulting in a reduction in the radiation received by photovoltaic panels. Indirect impacts are mainly caused by changes in topography and meteorological patterns. For example, factors such as wind corridor effect, vegetation cover and soil moisture will change the local temperature and humidity environment, thereby affecting photovoltaic output. Specifically, microtopography significantly changes microclimate conditions by affecting factors such as terrain undulation, slope, altitude, vegetation cover, and soil type in local areas. Undulating terrain can affect air flow. Valleys and low-lying areas are prone to accumulate cold air and form lower temperatures, while mountains and slopes cause hot air to rise and the temperature is relatively high. Different slopes receive different amounts of solar radiation. The temperature on the south slope is higher, while the temperature on the north slope is lower. As the altitude increases, the temperature gradually decreases, especially in mountainous areas, where climate change is more dramatic. Vegetation cover also plays an important role in local climate. Forest and grassland areas usually have higher humidity and lower temperatures, while desert or bare land areas have larger temperature differences. Water bodies such as lakes and rivers can form a stable microclimate by regulating humidity and temperature. In addition, especially in mountainous and valley areas, changes in wind speed and direction will also significantly affect local climate conditions such as temperature, humidity and precipitation. At the same time, environmental pollution, air quality and dust deposition will also cause dust to accumulate on the surface of photovoltaic panels, thereby reducing the absorption of solar radiation and further reducing power generation capacity. Further considering whether the photovoltaic power station is located on the coast or at sea, this difference will also affect the microclimate, such as: coastal and offshore areas have higher humidity and cloud cover, faster wind speed, and lower sunshine time and stability than inland areas. These microclimates further affect the stability and output efficiency of photovoltaic output.

[0054] Based on the above analysis, if Figure 2 As shown, multi-level clustering is performed on the data obtained in step S1, as follows:

[0055] The first layer is based on the spatial division of terrain characteristics: the region is divided into several sub-regions according to the overall terrain type and terrain parameters of the region. The micro-topography characteristics in each sub-region are similar, but the micro-topography of each sub-region is different; the overall terrain types include hills, plains, mountains, and plateaus, and the terrain parameters include altitude, slope, and relative undulation. The classification method of this layer is as follows:

[0056] (11)

[0057] In the formula, represents the altitude factor; represents the slope factor; Represents the relative fluctuation factor.

[0058] Mountainous areas: If the altitude is ≥1000, and the slope is >15 or the relative undulation is ≥500, it is classified as mountainous. This covers the situation of "high altitude and steep or undulating terrain" and does not conflict with plateaus.

[0059] Plateau: If the altitude is ≥1000, but the slope is ≤15 and the relative undulation is <500, it is classified as a plateau. Since "mountainous areas" have been determined in advance, only areas with an altitude of ≥1000 but a gentle slope and little undulation will fall into the plateau category.

[0060] Hills: When the conditions of "mountain" and "plateau" are not met (i.e. the altitude is <1000, or at least does not meet the mountain determination), areas with an altitude of 200-800 m, a slope of 5°-15°, and a relative undulation of 200-500 m will be classified as hills.

[0061] Plain: If the above terrain conditions are no longer met, the situation with altitude <200, slope <5°, and relative relief <200 will be classified as plain.

[0062] It is worth mentioning that the above four types of terrain are the most representative terrains. If it is considered that the photovoltaic power station may be located at a location with relatively unique terrain, the present invention can further subdivide the terrain according to actual conditions.

[0063] The second layer is the local pattern division based on meteorological variables: on the basis of the first layer of terrain feature division, combined with the distribution characteristics of local meteorological variables, each sub-area is divided into several micro-meteorological pattern categories, and the meteorological characteristics and terrain characteristics within each pattern are similar; the meteorological variables include precipitation, temperature, wind speed, and solar radiation.

[0064] The third layer refines the sub-array division range within the photovoltaic power station by combining the detailed characteristics of the micro-topography and the factors affecting the performance of the photovoltaic power station components: On the basis of the second layer, the influence of the micro-topography undulations, local slope and orientation on the light reception is evaluated, and the calculation accuracy of the photovoltaic output prediction is optimized. At the same time, the influence of the shading effect and the thermal effect between equipment on the photovoltaic performance, as well as the contribution of the surface albedo to the radiation are considered. The photovoltaic power generation and temperature changes in each sub-area are analyzed in depth, and each sub-area is further subdivided according to these influencing factors to further improve the accuracy of the photovoltaic output prediction.

[0065] The corresponding photovoltaic output prediction model is constructed based on multi-level clustering to capture the impact of different micro-topography on photovoltaic output. The expression of the photovoltaic output prediction model is as follows:

[0066] (1)

[0067] In the formula, Indicates the predicted value of photovoltaic output; Represents the terrain factors that affect the PV output prediction; Represents the micro-meteorological state factor affected by micro-topography; Indicates photovoltaic panels; Indicates the total number of photovoltaic panels; Indicates The effective area of ​​the photovoltaic panels; Indicates arrival The solar radiation intensity on the surface of a photovoltaic panel; Indicates the function affecting the photovoltaic module;

[0068] (2)

[0069] In the formula, They are respectively expressed as weight coefficients of slope, aspect, altitude and surface albedo; represents the orientation factor; represents the surface albedo factor;

[0070] (3)

[0071] In the formula, Respectively expressed as The weight coefficient of Indicates temperature; Indicates humidity; Indicates the wind speed of the micro-topography; represents solar radiation; Indicates the regional base temperature; It represents the vertical temperature lapse rate; Indicates terrain elevation; Slope influence coefficient; Indicates the solar azimuth; Indicates the aspect angle of the ground slope; Average precipitation for the region; represents humidity impact factor; represents the adjustment factor used to control the rate at which humidity changes with altitude; Indicates height; Indicates the height of features associated with terrain; Indicates the reference wind speed when there is no micro-topography influence; Indicates the elevation growth rate coefficient; It indicates the gain or attenuation effect of micro-topography wind speed as the terrain height changes; Indicates the reference radiation intensity when there is no terrain influence;

[0072] (4)

[0073] In the formula, Represents the total area of ​​PV panels; represents the cloud occlusion factor; represents the inter-plate occlusion factor; Represents the coincidence factor. When cloud occlusion and inter-panel occlusion are completely independent, the coincidence factor ,When the cloud occlusion and the inter-board occlusion overlap, the smaller part of the two occlusion ratios is taken as the overlapping part; It represents the length of the cloud layer, which represents the horizontal scale of the cloud layer blocking the photovoltaic panel; Indicates the total time; Indicates the duration of the occlusion; Indicates the wind speed under the combined influence of micro-topography and overall topography; represents the inclination angle of the photovoltaic panel, Indicates the sun's altitude angle. When there is occlusion When , it returns a positive value, that is, The actual value calculated; when there is no occlusion When , return 0;

[0074] (10)

[0075] In the formula, Indicates the wind speed over the overall terrain; Indicates the reference wind speed when there is no terrain influence; Indicates terrain height, the absolute height of terrain relative to sea level; Indicates slope; Indicates the effect of terrain height and slope on wind speed; represents the height influence coefficient; Indicates the target height for wind speed calculation, usually relative to the ground (e.g., the installation height of the wind speed sensor or the target height for wind speed prediction); Indicates the surface roughness length; represents the terrain height influence coefficient; Indicates the reference height, which is the base terrain height of the photovoltaic panel area; Represents the slope influence coefficient.

[0076] (5)

[0077] In the formula, Indicates the amount of radiation intensity fluctuation; Indicates the maximum radiation intensity;

[0078] (6)

[0079] In the formula, Indicates The reflectivity of the surface of a photovoltaic panel reflects the part of the incident light that is reflected but not absorbed; Indicates The working temperature of photovoltaic modules Efficiency under Indicates Block PV panel system efficiency; Indicates reference temperature Efficiency under represents the temperature coefficient; Indicates the ambient temperature; represents the temperature rise caused by solar radiation; Indicates the temperature rise caused by thermal effects between devices; Indicates the maximum operating temperature of the PV panel.

[0080] Step S3, in order to more accurately identify weather changes, such as: in plain terrain, clouds move quickly and a shorter prediction time scale is used. This step S3 will set the prediction time scale of the corresponding micro-topography photovoltaic output according to the moving speed of clouds in different sub-areas, use sensors to monitor the meteorological data of the photovoltaic power station at the set prediction time scale, and calculate the current photovoltaic output prediction value according to the photovoltaic output prediction model.

[0081] Step S4, collect historical photovoltaic outputs of different categories according to the multi-layer clustering method of step S2, compare the photovoltaic output prediction value obtained in step S3 with the historical photovoltaic output of the same type, and if the prediction error is within the threshold range, proceed to step S5; if the prediction error exceeds the threshold, iteratively optimize and predict the photovoltaic output prediction model within the maximum optimization number range until the prediction error of the model is within the threshold range and then proceed to step S5; if the maximum optimization number has been exceeded and the prediction error of the model still exceeds the threshold, enter the early warning mode, indicating that abnormal weather exists and human intervention in the prediction is required.

[0082] The calculation formula of the prediction error is as follows:

[0083] (7)

[0084] In the formula, Represents the prediction error, which is the difference between the predicted PV output and the average historical PV output; Indicates the real-time photovoltaic output forecast value; Indicates the same micro-meteorological conditions in history Average photovoltaic output under ; Indicates the minimum value of the error interval; Indicates the maximum value of the error interval; Indicates the current micro-meteorological status; Represents the current micro-meteorological state function determined by multiple environmental parameters; Indicates air pressure;

[0085] The formula for optimizing the photovoltaic output prediction model is as follows:

[0086] (8)

[0087] In the formula, Respectively represent the predicted photovoltaic output values ​​before and after optimization; represents the adjustment factor; Indicates the maximum number of optimizations set; Indicates the current number of optimizations; Indicates that a warning signal has been triggered.

[0088] Step S5, using the optimized photovoltaic output prediction model to predict the photovoltaic output of the photovoltaic power station.

[0089] Step S6, using RMSE and MAE The two evaluation indicators comprehensively evaluate the prediction results of step S5;

[0090] (9)

[0091] In the formula, represents the root mean square error; represents the mean absolute error; Indicates predicted values; represents the number of samples; represents the target value; Represents the predicted value, which is the predicted value of photovoltaic output.

[0092] The above embodiments are preferred implementation schemes of the present invention. In addition, the present invention may also be implemented in other ways. Any obvious replacement without departing from the concept of the present technical solution is within the protection scope of the present invention.

[0093] In order to make it easier for ordinary technicians in the field to understand the improvements of the present invention over the prior art, some drawings and descriptions of the present invention have been simplified, and for the sake of clarity, some other elements are omitted in this application document. Ordinary technicians in the field should realize that these omitted elements may also constitute the content of the present invention.

Claims

1. A photovoltaic output prediction method based on micro-topography, characterized in that: include: Step S1, obtaining the terrain and meteorological data of the photovoltaic power station, including longitude and latitude, weather, and altitude, and normalizing the data; Step S2, considering the direct and indirect effects of micro-topography on photovoltaic output, multi-level clustering is performed on the data obtained in step S1 to construct a photovoltaic output prediction model; The multi-level clustering method is as follows: The first layer is spatial division based on terrain characteristics: the region is divided into several sub-regions according to the overall terrain type and terrain parameters of the region. The micro-topography characteristics in each sub-region are similar, but the micro-topography of each sub-region is different; the overall terrain types include hills, plains, mountains, and plateaus, and the terrain parameters include altitude, slope, and relative undulation; The second layer is the local pattern division based on meteorological variables: on the basis of the first layer of terrain feature division, combined with the distribution characteristics of local meteorological variables, each sub-area is divided into several micro-meteorological pattern categories, and the meteorological characteristics and terrain characteristics in each pattern are similar; the meteorological variables include precipitation, temperature, wind speed, and solar radiation; The third layer combines the detailed features of micro-topography and the factors affecting the performance of PV power station components to refine the sub-array division range within the PV power station: Based on the second layer, the influence of the micro-topography undulations, local slope and orientation on light reception is evaluated, the influence of the shielding effect and the thermal effect between devices on the PV performance is considered, and the contribution of the surface albedo to radiation is considered. The PV power generation and temperature changes of each sub-area are analyzed, and each sub-area is further subdivided; Step S3, monitoring the meteorological data of the photovoltaic power station, and calculating the current photovoltaic output prediction value according to the photovoltaic output prediction model; Step S4, collecting historical photovoltaic outputs of different categories according to the multi-layer clustering method of step S2, comparing the photovoltaic output prediction value obtained in step S3 with the historical photovoltaic outputs of the same category, and optimizing the photovoltaic output prediction model according to the comparison result; Step S5, using the optimized photovoltaic output prediction model to predict the photovoltaic output of the photovoltaic power station.

2. The photovoltaic output prediction method based on micro-topography influence according to claim 1 is characterized in that: In step S2, a corresponding photovoltaic output prediction model is constructed according to multi-level clustering to capture the impact of different micro-topography on photovoltaic output. The expression of the photovoltaic output prediction model is as follows: （1） In the formula, Indicates the predicted value of photovoltaic output; Represents the terrain factors that affect the PV output prediction; Represents the micro-meteorological state factor affected by micro-topography; Indicates photovoltaic panels; Indicates the total number of photovoltaic panels; Indicates The effective area of ​​the photovoltaic panels; Indicates arrival The solar radiation intensity on the surface of a photovoltaic panel; Indicates the function affecting the photovoltaic module; （2） In the formula, They are respectively expressed as weight coefficients of slope, aspect, altitude and surface albedo; represents the slope factor; represents the orientation factor; represents the altitude factor; represents the surface albedo factor; （3） In the formula, Respectively expressed as The weight coefficient of Indicates temperature; Indicates humidity; Indicates the wind speed of the micro-topography; represents solar radiation; Indicates the regional base temperature; It represents the vertical temperature lapse rate; Indicates terrain elevation; Slope influence coefficient; Indicates the solar azimuth; Indicates the aspect angle of the ground slope; Average precipitation for the region; represents humidity impact factor; represents the adjustment factor used to control the rate at which humidity changes with altitude; Indicates height; Indicates the height of features associated with terrain; Indicates the reference wind speed when there is no micro-topography influence; Indicates the elevation growth rate coefficient; It indicates the gain or attenuation effect of micro-topography wind speed as the terrain height changes; Indicates the reference radiation intensity when there is no terrain influence; （4） In the formula, Represents the total area of ​​PV panels; represents the cloud occlusion factor; represents the inter-plate occlusion factor; Represents the coincidence factor. When cloud occlusion and inter-panel occlusion are completely independent, the coincidence factor ,When the cloud occlusion and the inter-board occlusion overlap, the smaller part of the two occlusion ratios is taken as the overlapping part; It represents the length of the cloud layer, which represents the horizontal scale of the cloud layer blocking the photovoltaic panel; Indicates the total time; Indicates the duration of the occlusion; Indicates the wind speed under the combined influence of micro-topography and overall topography; represents the inclination angle of the photovoltaic panel, Indicates the sun's altitude angle. When there is occlusion When , it returns a positive value; when there is no occlusion When , return 0; （5） In the formula, Indicates the amount of radiation intensity fluctuation; Indicates the maximum radiation intensity; （6） In the formula, Indicates The reflectivity of the surface of a photovoltaic panel reflects the part of the incident light that is reflected but not absorbed; Indicates The working temperature of photovoltaic modules Efficiency under Indicates Block PV panel system efficiency; Indicates reference temperature Efficiency under represents the temperature coefficient; Indicates the ambient temperature; represents the temperature rise caused by solar radiation; Indicates the temperature rise caused by thermal effects between devices; Indicates the maximum operating temperature of the PV panel.

3. The photovoltaic output prediction method based on micro-topography influence according to claim 2 is characterized in that: In step S3, the prediction time scale of the photovoltaic output of the corresponding micro-topography is set according to the moving speed of the cloud layer in different sub-areas, the meteorological data of the photovoltaic power station is monitored at the set prediction time scale, and the current photovoltaic output prediction value is calculated according to the photovoltaic output prediction model.

4. The photovoltaic output prediction method based on micro-topography influence according to claim 3 is characterized in that: In the step S4, the photovoltaic output prediction value obtained in step S3 is compared with the historical photovoltaic output of the same type. If the prediction error is within the threshold range, the process proceeds to step S5; if the prediction error exceeds the threshold, the photovoltaic output prediction model is iteratively optimized and predicted within the maximum optimization times, until the prediction error of the model is within the threshold range, and then the process proceeds to step S5. If the maximum optimization times have been exceeded and the prediction error of the model still exceeds the threshold, the process enters the early warning mode, indicating that abnormal weather exists and human intervention is required for prediction; The prediction error is calculated as follows: （7） In the formula, Represents the prediction error, which is the difference between the predicted PV output and the average historical PV output; Indicates the real-time photovoltaic output forecast value; Indicates the same micro-meteorological conditions in history Average photovoltaic output under Indicates the minimum value of the error interval; Indicates the maximum value of the error interval; Indicates the current micro-meteorological status; Represents the current micro-meteorological state function determined by multiple environmental parameters; Indicates air pressure; The formula for optimizing the photovoltaic output prediction model is as follows: （8） In the formula, Respectively represent the predicted photovoltaic output values ​​before and after optimization; represents the adjustment factor; Indicates the maximum number of optimizations set; Indicates the current number of optimizations; Indicates that a warning signal has been triggered.

5. The photovoltaic output prediction method based on micro-topography influence according to claim 4 is characterized in that: The method further comprises step S6, using RMSE and MAE The two evaluation indicators comprehensively evaluate the prediction results of step S5; （9） In the formula, represents the root mean square error; represents the mean absolute error; Indicates predicted values; represents the number of samples; represents the target value; Represents the predicted value, which is the predicted value of photovoltaic output.

6. The photovoltaic output prediction method based on micro-topography influence according to claim 5 is characterized by: Wind speed under the combined influence of micro-topography and overall topography The expression is as follows: （10） In the formula, Indicates the wind speed over the overall terrain; Indicates the reference wind speed when there is no terrain influence; Indicates terrain height, the absolute height of terrain relative to sea level; Indicates slope; Indicates the effect of terrain height and slope on wind speed; represents the height influence coefficient; Indicates the target height for wind speed calculation; Indicates the surface roughness length; represents the terrain height influence coefficient; Indicates the reference height, which is the base terrain height of the photovoltaic panel area; Represents the slope influence coefficient.

Images