A distributed photovoltaic power generation power prediction method and system based on AI

Abstract

This invention relates to the field of power generation prediction technology, and specifically discloses an AI-based method and system for predicting distributed photovoltaic (PV) power generation. The method includes: obtaining the correlation time of each type of environmental parameter between the distributed PV unit and each similar distributed PV unit based on all environmental parameters at each time point within a preset time period before the current time; obtaining the prediction matching degree of each correlated distributed PV unit based on the correlation time of all environmental parameters between the distributed PV unit and all similar distributed PV units; and obtaining the PV power generation prediction result of the distributed PV unit based on the prediction matching degree of all correlated distributed PV units. This invention achieves accurate prediction of PV power generation of distributed PV units with relatively little reference to previous data, effectively reducing the data volume requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of power generation prediction, and in particular to an AI-based distributed photovoltaic power generation prediction method and system. Background Art

[0002] At present, with the continuous growth of global energy demand and the enhancement of environmental protection awareness, distributed photovoltaic power generation has received widespread attention as a clean and renewable energy form. Metal roofs are a common form of construction, especially in industrial plants, warehouses and other buildings, providing an ideal space for the installation of distributed photovoltaic power generation systems. By installing photovoltaic power generation systems on metal roofs, it is possible to achieve effective utilization of solar energy, reduce dependence on traditional energy, and reduce energy consumption and environmental pollution. With the widespread application of distributed photovoltaic power generation systems, predicting photovoltaic power generation and improving the accuracy of prediction have become the key to ensuring the stability of the power grid. However, traditional photovoltaic power generation prediction methods often rely on a large amount of historical data, which may be limited in some cases. For example, newly installed distributed photovoltaic units or a large proportion of distributed photovoltaic power stations are difficult to be equipped with a complete online monitoring system due to their small installed capacity and wide distribution, resulting in data scarcity. In addition, the output power of photovoltaic power generation has the inherent property of random fluctuations. As the scale of photovoltaic power generation grid connection increases, the impact of this fluctuation on the frequency stability of the power system is becoming increasingly significant. Therefore, a more efficient and accurate prediction method is needed to adapt to distributed photovoltaic power generation systems under different data conditions.

[0003] However, the existing AI-based distributed photovoltaic power prediction methods and systems only establish multi-type photovoltaic power prediction models through multi-dimensional historical data, and select the corresponding prediction models according to the actual data to predict the photovoltaic power generation. It does not consider how to achieve accurate prediction of the photovoltaic power generation of distributed photovoltaic units with less reference to previous data, and does not reduce the requirements for data volume, so that effective prediction can be performed on distributed photovoltaic units with scarce data or newly installed. For example, the patent with the publication number of "CN116683427A" and the name of "distributed photovoltaic short-term power prediction method and device based on AI" includes the following steps: at the data source, multiple environmental factors affecting photovoltaic power generation are obtained, so as to improve the dimension of subsequent model training data based on this; at the same time, before training, the acquired historical data is classified; in this way, multi-dimensional historical data can be used to establish multi-type photovoltaic power prediction models, and the corresponding prediction models are selected according to the actual data to predict photovoltaic power generation. Therefore, the above patent can achieve accurate prediction of photovoltaic power generation, and its applicability and accuracy are greatly improved compared with traditional technologies, which is suitable for large-scale application and promotion in the field of photovoltaic power generation prediction. However, the patent only establishes multi-type photovoltaic power prediction models through multi-dimensional historical data, and selects the corresponding prediction models according to the actual data to predict photovoltaic power generation. It does not take into account the accurate prediction of photovoltaic power generation of distributed photovoltaic units with less previous data, and does not reduce the requirements for data volume, so that effective prediction can be made on distributed photovoltaic units with scarce data or newly installed.

[0004] Therefore, the present invention proposes an AI-based distributed photovoltaic power generation power prediction method and system. Summary of the invention

[0005] The present invention provides an AI-based distributed photovoltaic power generation power prediction method and system, which is used to obtain the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit according to the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, and realizes the screening of the specific moments when the change trend of each type of environmental parameter of each similar distributed photovoltaic unit within a preset time period before the current moment is similar to the change trend of the corresponding type of environmental parameter of the distributed photovoltaic unit at the current moment, and then obtains the predicted power matrix of the distributed photovoltaic unit according to the association moment of all types of environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, which is convenient for the subsequent calculation of the prediction matching degree, and according to the distributed photovoltaic unit The power matrix is ​​predicted to obtain the predicted matching degree of each associated distributed photovoltaic unit, so as to quantify the correlation of each associated distributed photovoltaic unit with respect to the distributed photovoltaic unit in terms of predicted power generation. Then, according to the predicted matching degree of all associated distributed photovoltaic units, the predicted distributed photovoltaic unit is obtained, so as to facilitate the determination of the prediction time of the subsequent distributed photovoltaic unit. According to the predicted distributed photovoltaic unit, the predicted time of the distributed photovoltaic unit is obtained, and finally, according to the predicted time of the distributed photovoltaic unit, the photovoltaic power generation power prediction result of the distributed photovoltaic unit is obtained. With reference to less previous data, the photovoltaic power generation power of the distributed photovoltaic unit is accurately predicted, which reduces the requirement for the amount of data and enables effective prediction on distributed photovoltaic units with scarce data or newly installed.

[0006] The present invention provides a distributed photovoltaic power generation power prediction method based on AI, comprising:

[0007] S1: Based on the characteristics of all similar battery components of the distributed photovoltaic units at the current moment, all similar distributed photovoltaic units are obtained;

[0008] S2: Based on all environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained;

[0009] S3: based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, a prediction power matrix of the distributed photovoltaic unit is obtained, and based on the prediction power matrix of the distributed photovoltaic unit, a prediction matching degree of each associated distributed photovoltaic unit is obtained;

[0010] S4: Based on the predicted matching degree of all associated distributed photovoltaic units, the predicted distributed photovoltaic unit is obtained, and based on the predicted distributed photovoltaic unit, the predicted time of the distributed photovoltaic unit is obtained, and based on the predicted time of the distributed photovoltaic unit, the photovoltaic power generation power prediction result of the distributed photovoltaic unit is obtained.

[0011] Preferably, the AI-based distributed photovoltaic power generation prediction method, S1: based on the characteristics of all battery-like components of the distributed photovoltaic unit at the current moment, all similar distributed photovoltaic units are obtained, including:

[0012] Obtaining the characteristics of all types of battery components of the distributed photovoltaic unit at the current moment, wherein the characteristics of all types of battery components include the number of battery cells, the area of ​​the battery cells and the photoelectric conversion efficiency of the battery components;

[0013] Treat all distributed photovoltaic units connected to the AI-based distributed photovoltaic power generation prediction system as standby distributed photovoltaic units;

[0014] When the difference between the value of the product of the number of cells and the cell area of ​​each prepared distributed photovoltaic unit at the current moment and the value of the product of the number of cells and the cell area of ​​the distributed photovoltaic unit at the current moment is less than the preset first difference, and the difference between the photoelectric conversion efficiency of the battery assembly of the corresponding prepared distributed photovoltaic unit at the current moment and the photoelectric conversion efficiency of the battery assembly of the distributed photovoltaic unit at the current moment is less than the preset second difference, the corresponding prepared distributed photovoltaic unit is regarded as a similar distributed photovoltaic unit, and all similar distributed photovoltaic units are obtained.

[0015] Preferably, the distributed photovoltaic power generation power prediction method based on AI, S2: based on all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the associated moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, including:

[0016] Obtaining all class environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, wherein all class environmental parameters include solar radiation intensity, ambient temperature and ambient wind speed;

[0017] Taking each moment of the distributed photovoltaic unit in the preset time period before the current moment as the horizontal coordinate value, and each type of environmental parameter of the distributed photovoltaic unit at the corresponding moment in the preset time period before the current moment as the vertical coordinate value, each type of environmental parameter point of the distributed photovoltaic unit is obtained;

[0018] Take each moment of each similar distributed photovoltaic unit in the preset time period before the current moment as the horizontal coordinate value, and each type of environmental parameter of the corresponding moment of the similar distributed photovoltaic unit in the preset time period before the current moment as the vertical coordinate value, and obtain each type of environmental parameter point of each similar distributed photovoltaic unit;

[0019] Based on each type of environmental parameter point of the distributed photovoltaic unit and each type of environmental parameter point of each similar distributed photovoltaic unit, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained.

[0020] Preferably, the AI-based distributed photovoltaic power generation prediction method obtains a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on each type of environmental parameter point of the distributed photovoltaic unit and each type of environmental parameter point of each similar distributed photovoltaic unit, including:

[0021] Connect each type of environmental parameter point of the distributed photovoltaic unit in the order of the horizontal coordinate from small to large, and obtain the connection line of each type of environmental parameter of the distributed photovoltaic unit, and connect each type of environmental parameter point of each similar distributed photovoltaic unit in the order of the horizontal coordinate from small to large, and obtain the connection line of each type of environmental parameter of each similar distributed photovoltaic unit;

[0022] The images of the connection lines of each type of environmental parameters of the distributed photovoltaic unit and the connection lines of the corresponding type of environmental parameters of each similar distributed photovoltaic unit in the same coordinate system are regarded as the comparison diagram of the corresponding type of environmental parameters between the distributed photovoltaic unit and the corresponding similar distributed photovoltaic unit.

[0023] Preferably, the AI-based distributed photovoltaic power generation prediction method obtains the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, including:

[0024] Based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and the preset curve judgment model, all connecting line segments similar to the end line segments of the connecting line of each type of environmental parameter of each similar distributed photovoltaic unit are obtained as the associated line segments of the connecting line of the corresponding type of environmental parameter of the corresponding similar distributed photovoltaic unit;

[0025] The moments corresponding to all associated line segments of the connecting line of each type of environmental parameter of each similar distributed photovoltaic unit are regarded as the associated moments of the corresponding type of environmental parameters between the distributed photovoltaic unit and the corresponding similar distributed photovoltaic unit.

[0026] Preferably, the AI-based distributed photovoltaic power generation power prediction method, S3: based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, obtain the predicted power matrix of the distributed photovoltaic unit, and based on the predicted power matrix of the distributed photovoltaic unit, obtain the predicted matching degree of each associated distributed photovoltaic unit, including:

[0027] If the number of association moments of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is greater than zero, it is determined that the corresponding similar distributed photovoltaic units are associated in the corresponding type of environmental parameters;

[0028] If each similar distributed photovoltaic unit is associated in all class environmental parameters, the corresponding similar distributed photovoltaic unit is determined to be an associated distributed photovoltaic unit;

[0029] Obtaining a predicted power matrix of the distributed photovoltaic unit based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units;

[0030] Based on the predicted power matrix of the distributed photovoltaic units, the predicted matching degree of each associated distributed photovoltaic unit is obtained.

[0031] Preferably, the AI-based distributed photovoltaic power generation power prediction method obtains the predicted power matrix of the distributed photovoltaic unit based on the association time of all environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, including:

[0032] The association moment closest to the current moment in terms of time sequence among all association moments of each type of environmental parameter between the distributed photovoltaic unit and each associated distributed photovoltaic unit is regarded as the selected association moment of the corresponding type of environmental parameter between the distributed photovoltaic unit and the corresponding associated distributed photovoltaic unit;

[0033] All associated distributed photovoltaic units are ordinal-defined starting from 1 in order of distance from the distributed photovoltaic units from small to large, to obtain an ordinal definition result;

[0034] Based on the selected association moments and ordinal definition results of all class environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, which is:

[0035] ;

[0036] in, is the predicted power matrix of distributed photovoltaic units, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, where n is the number of associated distributed photovoltaic units.

[0037] Preferably, the AI-based distributed photovoltaic power generation power prediction method obtains the prediction matching degree of each associated distributed photovoltaic unit based on the prediction power matrix of the distributed photovoltaic unit, including:

[0038] ;

[0039] in, is the predicted matching degree of the associated distributed photovoltaic unit with ordinal number, is the value of the power generated by the distributed photovoltaic unit at the current moment, is the sum of the values ​​of the power generation at the selected association time of all class environmental parameters between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number, is the rank of the predicted power matrix of the distributed photovoltaic unit, is a distributed photovoltaic unit and the ordinal number is The value of the power generated at the selected associated moment of the solar radiation intensity between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient temperature between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient wind speed between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The maximum value of the power generation at the selected association moment of all types of environmental parameters between the associated distributed photovoltaic units, is the natural logarithm, and the value of the natural constant e is 2.718.

[0040] Preferably, the distributed photovoltaic power generation prediction method based on AI, S4: based on the prediction matching degree of all associated distributed photovoltaic units, obtain the predicted distributed photovoltaic unit, and based on the predicted distributed photovoltaic unit, obtain the predicted time of the distributed photovoltaic unit, and based on the predicted time of the distributed photovoltaic unit, obtain the photovoltaic power generation prediction result of the distributed photovoltaic unit, including:

[0041] The associated distributed photovoltaic unit with the largest prediction matching degree among all associated distributed photovoltaic units is regarded as the predicted distributed photovoltaic unit;

[0042] The selected associated time of the corresponding class environmental parameter with the largest power generation among the power generation at the selected associated time of all class environmental parameters between the distributed photovoltaic unit and the predicted distributed photovoltaic unit is regarded as the predicted time of the distributed photovoltaic unit;

[0043] The power generation power of the distributed photovoltaic unit at the prediction time is predicted as the photovoltaic power generation power prediction result of the distributed photovoltaic unit.

[0044] The present invention provides an AI-based distributed photovoltaic power generation prediction system, which is used to execute any one of the AI-based distributed photovoltaic power generation prediction methods in Embodiments 1 to 9, including:

[0045] A first processing module is used to obtain all similar distributed photovoltaic units based on the characteristics of all battery-like components of the distributed photovoltaic units at the current moment;

[0046] A second processing module is used to obtain a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, and obtain an association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology;

[0047] A calculation module, for obtaining a predicted power matrix of the distributed photovoltaic unit based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, and obtaining a predicted matching degree of each associated distributed photovoltaic unit based on the predicted power matrix of the distributed photovoltaic unit;

[0048] The prediction module is used to obtain the predicted distributed photovoltaic unit based on the predicted matching degree of all associated distributed photovoltaic units, and obtain the predicted time of the distributed photovoltaic unit based on the predicted distributed photovoltaic unit, and obtain the photovoltaic power generation power prediction result of the distributed photovoltaic unit based on the predicted time of the distributed photovoltaic unit.

[0049] The beneficial effects of the present invention compared with the prior art are as follows: based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and the specific moments when the change trend of each type of environmental parameter of each similar distributed photovoltaic unit within the preset time period before the current moment is similar to the change trend of the corresponding type of environmental parameter of the distributed photovoltaic unit at the current moment are screened out, and then according to the association moment of all types of environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, which is convenient for the subsequent calculation of the predicted matching degree, and according to the predicted power matrix of the distributed photovoltaic unit , obtain the predicted matching degree of each associated distributed photovoltaic unit, realize the quantification of the correlation of each associated distributed photovoltaic unit with respect to the distributed photovoltaic unit in terms of predicted power generation, and then obtain the predicted distributed photovoltaic unit according to the predicted matching degree of all associated distributed photovoltaic units, so as to facilitate the determination of the prediction time of the subsequent distributed photovoltaic unit, obtain the predicted time of the distributed photovoltaic unit according to the predicted distributed photovoltaic unit, and finally obtain the photovoltaic power generation power prediction result of the distributed photovoltaic unit according to the predicted time of the distributed photovoltaic unit, realize the accurate prediction of the photovoltaic power generation power of the distributed photovoltaic unit with reference to less previous data, reduce the requirement for data volume, and make effective prediction on the distributed photovoltaic units with scarce data or newly installed.

[0050] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the written application document.

[0051] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0053] Figure 1 This is a flow chart of a distributed photovoltaic power generation power prediction method based on AI in an embodiment of the present invention;

[0054] Figure 2 Schematic diagram of a distributed photovoltaic power generation prediction system based on AI in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention. Embodiment 1:

[0056] The present invention provides a distributed photovoltaic power generation power prediction method based on AI, referring to Figure 1 ,include:

[0057] S1: Based on the characteristics of all similar battery components of the distributed photovoltaic units at the current moment, all similar distributed photovoltaic units are obtained;

[0058] S2: Based on all environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained;

[0059] S3: based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, a prediction power matrix of the distributed photovoltaic unit is obtained, and based on the prediction power matrix of the distributed photovoltaic unit, a prediction matching degree of each associated distributed photovoltaic unit is obtained;

[0060] S4: Based on the predicted matching degree of all associated distributed photovoltaic units, the predicted distributed photovoltaic unit is obtained, and based on the predicted distributed photovoltaic unit, the predicted time of the distributed photovoltaic unit is obtained, and based on the predicted time of the distributed photovoltaic unit, the photovoltaic power generation power prediction result of the distributed photovoltaic unit is obtained.

[0061] In this embodiment, the distributed photovoltaic unit is any photovoltaic power generation unit that uses an AI-based distributed photovoltaic power generation power prediction system to implement photovoltaic power generation power prediction, wherein the distributed photovoltaic power generation unit is specifically a small modular, decentralized power generation unit with a power generation power ranging from several kilowatts to tens of megawatts, which is arranged around the user and has reliability and efficiency.

[0062] In this embodiment, similar distributed photovoltaic units are some distributed photovoltaic units that are similar in power generation component composition and structure, which are screened out from all distributed photovoltaic units connected to the AI-based distributed photovoltaic power generation power prediction system based on the characteristics of all battery-like components of the distributed photovoltaic units at the current moment, and any distributed photovoltaic unit individually corresponds to some similar distributed photovoltaic units.

[0063] In this embodiment, the preset time period is a time period preset to obtain all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units, and the preset time period in this embodiment is set to 20 hours.

[0064] In this embodiment, each moment is a moment selected from a preset time period before the current moment, and the time length between each adjacent moment is the same, and the time length between adjacent moments in this embodiment is 1 hour.

[0065] In this embodiment, the comparison chart of each type of environmental parameter is obtained based on all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, and can reflect the change trend of each type of environmental parameter of the distributed photovoltaic unit and each similar distributed photovoltaic unit.

[0066] In this embodiment, the AI ​​technology is a neural network model technology, and in this embodiment, it is specifically a preset curve judgment model.

[0067] In this embodiment, the associated moment of each type of environmental parameter is a specific moment in which the change trend of each type of environmental parameter of each similar distributed photovoltaic unit within a preset time period before the current moment is similar to the change trend of the corresponding type of environmental parameter of the distributed photovoltaic unit at the current moment.

[0068] In this embodiment, the predicted power matrix of the distributed photovoltaic unit is a matrix obtained based on the association moments of all types of environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, and can be used to calculate the predicted matching degree of the associated distributed photovoltaic units.

[0069] In this embodiment, the predicted matching degree of the associated distributed photovoltaic unit is a numerical value obtained based on the predicted power matrix of the distributed photovoltaic unit, which can characterize the correlation of each associated distributed photovoltaic unit with respect to the distributed photovoltaic unit in terms of predicted power generation.

[0070] In this embodiment, the predicted distributed photovoltaic unit is an associated distributed photovoltaic unit selected from the associated distributed photovoltaic units and which can most accurately predict the power generation power of the distributed photovoltaic unit.

[0071] In this embodiment, the predicted moment of the distributed photovoltaic unit is a moment obtained based on the predicted distributed photovoltaic unit, which can be used to predict the power generation power of the distributed photovoltaic unit at the next moment after the current moment (in this embodiment, the time interval between the next moment and the current moment is 1 hour).

[0072] In this embodiment, the photovoltaic power generation prediction result of the distributed photovoltaic unit is the photovoltaic power generation power of the distributed photovoltaic unit predicted at the next moment after the current moment based on the prediction moment of the distributed photovoltaic unit.

[0073] The beneficial effects of the above technology are as follows: based on the comparison diagram of each type of environmental parameters between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameters between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and the specific moments when the change trend of each type of environmental parameters of each similar distributed photovoltaic unit within the preset time period before the current moment is similar to the change trend of the corresponding type of environmental parameters of the distributed photovoltaic unit at the current moment are screened out, and then according to the association moment of all types of environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, which is convenient for the subsequent calculation of the predicted matching degree, and the predicted power matrix of each distributed photovoltaic unit is obtained according to the predicted power matrix of the distributed photovoltaic unit. The prediction matching degree of each associated distributed photovoltaic unit is used to quantify the correlation of each associated distributed photovoltaic unit with respect to the distributed photovoltaic unit in terms of predicted power generation. Then, according to the prediction matching degree of all associated distributed photovoltaic units, the predicted distributed photovoltaic unit is obtained, which is convenient for determining the prediction time of the subsequent distributed photovoltaic unit. According to the predicted distributed photovoltaic unit, the prediction time of the distributed photovoltaic unit is obtained. Finally, according to the prediction time of the distributed photovoltaic unit, the prediction result of the photovoltaic power generation of the distributed photovoltaic unit is obtained. With reference to less previous data, the photovoltaic power generation of the distributed photovoltaic unit is accurately predicted, which reduces the requirement for the amount of data and enables effective prediction on distributed photovoltaic units with scarce data or newly installed. Embodiment 2:

[0074] Based on Example 1, the distributed photovoltaic power generation power prediction method based on AI, S1: based on the characteristics of all battery-like components of the distributed photovoltaic unit at the current moment, all similar distributed photovoltaic units are obtained, including:

[0075] Obtaining the characteristics of all types of battery components of the distributed photovoltaic unit at the current moment, wherein the characteristics of all types of battery components include the number of battery cells, the area of ​​the battery cells and the photoelectric conversion efficiency of the battery components;

[0076] Treat all distributed photovoltaic units connected to the AI-based distributed photovoltaic power generation prediction system as standby distributed photovoltaic units;

[0077] When the difference between the value of the product of the number of cells and the cell area of ​​each prepared distributed photovoltaic unit at the current moment and the value of the product of the number of cells and the cell area of ​​the distributed photovoltaic unit at the current moment is less than the preset first difference, and the difference between the photoelectric conversion efficiency of the battery assembly of the corresponding prepared distributed photovoltaic unit at the current moment and the photoelectric conversion efficiency of the battery assembly of the distributed photovoltaic unit at the current moment is less than the preset second difference, the corresponding prepared distributed photovoltaic unit is regarded as a similar distributed photovoltaic unit, and all similar distributed photovoltaic units are obtained.

[0078] In this embodiment, the number of cells refers to the solar panels (photovoltaic panels) installed in the distributed photovoltaic unit.

[0079] In this embodiment, the cell area is the total area of ​​the solar panels (photovoltaic panels) installed in the distributed photovoltaic unit.

[0080] In this embodiment, the photovoltaic conversion efficiency of the battery assembly is the efficiency of the solar panels (photovoltaic panels) installed in the distributed photovoltaic units in converting solar energy into electrical energy.

[0081] In this embodiment, the preset first difference is a first difference threshold value that is preset and used to determine whether the prepared distributed photovoltaic unit is a similar distributed photovoltaic unit.

[0082] In this embodiment, the preset second difference is a second difference threshold value that is preset and used to determine whether the prepared distributed photovoltaic unit is a similar distributed photovoltaic unit.

[0083] The beneficial effects of the above technology are: according to the characteristics of all battery-like components of the distributed photovoltaic unit at the current moment, the preset first difference and the preset second difference, all similar distributed photovoltaic units are obtained, and some distributed photovoltaic units that are similar in power generation component composition and structure are screened out from all distributed photovoltaic units connected to the AI-based distributed photovoltaic power generation prediction system. Embodiment 3:

[0084] On the basis of Example 1, the AI-based distributed photovoltaic power generation prediction method, S2: based on all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the associated moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, including:

[0085] Obtaining all class environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, wherein all class environmental parameters include solar radiation intensity, ambient temperature and ambient wind speed;

[0086] Taking each moment of the distributed photovoltaic unit in the preset time period before the current moment as the horizontal coordinate value, and each type of environmental parameter of the distributed photovoltaic unit at the corresponding moment in the preset time period before the current moment as the vertical coordinate value, each type of environmental parameter point of the distributed photovoltaic unit is obtained;

[0087] Take each moment of each similar distributed photovoltaic unit in the preset time period before the current moment as the horizontal coordinate value, and each type of environmental parameter of the corresponding moment of the similar distributed photovoltaic unit in the preset time period before the current moment as the vertical coordinate value, and obtain each type of environmental parameter point of each similar distributed photovoltaic unit;

[0088] Based on each type of environmental parameter point of the distributed photovoltaic unit and each type of environmental parameter point of each similar distributed photovoltaic unit, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained.

[0089] In this embodiment, the solar radiation intensity is the radiation energy vertically projected by the sun per unit time and per unit area at each moment in a preset time period before the current moment for the distributed photovoltaic unit or each similar distributed photovoltaic unit.

[0090] In this embodiment, the ambient temperature is the temperature of the atmospheric environment of the distributed photovoltaic unit or each similar distributed photovoltaic unit at each moment in a preset time period before the current moment.

[0091] In this embodiment, the ambient wind speed is the wind speed in the environment of the distributed photovoltaic unit or each similar distributed photovoltaic unit at each moment in a preset time period before the current moment.

[0092] The beneficial effects of the above technology are: according to all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, each type of environmental parameter point of the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and then according to each type of environmental parameter point of the distributed photovoltaic unit and each similar distributed photovoltaic unit, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained. This embodiment gives in detail a specific method for obtaining a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment. Embodiment 4:

[0093] On the basis of Example 3, the AI-based distributed photovoltaic power generation prediction method obtains a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on each type of environmental parameter point of the distributed photovoltaic unit and each type of environmental parameter point of each similar distributed photovoltaic unit, including:

[0094] Connect each type of environmental parameter point of the distributed photovoltaic unit in the order of the horizontal coordinate from small to large, and obtain the connection line of each type of environmental parameter of the distributed photovoltaic unit, and connect each type of environmental parameter point of each similar distributed photovoltaic unit in the order of the horizontal coordinate from small to large, and obtain the connection line of each type of environmental parameter of each similar distributed photovoltaic unit;

[0095] The images of the connection lines of each type of environmental parameters of the distributed photovoltaic unit and the connection lines of the corresponding type of environmental parameters of each similar distributed photovoltaic unit in the same coordinate system are regarded as the comparison diagram of the corresponding type of environmental parameters between the distributed photovoltaic unit and the corresponding similar distributed photovoltaic unit.

[0096] The beneficial effects of the above technology are as follows: this embodiment gives in detail a specific method for obtaining a comparison diagram of each type of environmental parameter between a distributed photovoltaic unit and each similar distributed photovoltaic unit based on each type of environmental parameter point of the distributed photovoltaic unit and each type of environmental parameter point of each similar distributed photovoltaic unit, which facilitates the subsequent determination of the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit. Embodiment 5:

[0097] On the basis of Example 4, the AI-based distributed photovoltaic power generation prediction method obtains the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, including:

[0098] Based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and the preset curve judgment model, all connecting line segments similar to the end line segments of the connecting line of each type of environmental parameter of each similar distributed photovoltaic unit are obtained as the associated line segments of the connecting line of the corresponding type of environmental parameter of the corresponding similar distributed photovoltaic unit;

[0099] The moments corresponding to all associated line segments of the connecting line of each type of environmental parameter of each similar distributed photovoltaic unit are regarded as the associated moments of the corresponding type of environmental parameters between the distributed photovoltaic unit and the corresponding similar distributed photovoltaic unit.

[0100] In this embodiment, the preset curve judgment model is composed of a large number of pre-collected distributed photovoltaic units and a large number of similar distributed photovoltaic units (corresponding to each distributed photovoltaic unit) as model inputs, and the connecting line segments of all connecting line segments of each type of environmental parameters of each similar distributed photovoltaic unit that are manually marked are similar to the end line segments of the connecting line of the corresponding type of environmental parameters of the distributed photovoltaic unit (corresponding to the corresponding similar distributed photovoltaic unit) as model outputs, and the trained comparison diagram of each type of environmental parameters between the input distributed photovoltaic unit and each similar distributed photovoltaic unit can output all connecting line segments of all connecting line segments of each type of environmental parameters of each similar distributed photovoltaic unit that are similar to the end line segments of the connecting line of the corresponding type of environmental parameters of the distributed photovoltaic unit.

[0101] In this embodiment, the terminal line segment is the portion of the connecting line of the distributed photovoltaic unit corresponding to the class environmental parameter, which is located between the current moment and the previous moment.

[0102] In this embodiment, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is the moment corresponding to all the associated line segments of the connecting line of each type of environmental parameter of each similar distributed photovoltaic unit (the method for determining the moment corresponding to the associated line segment is to take the moment closest to the midpoint of the associated line segment in time series as the moment corresponding to the corresponding associated line segment).

[0103] The beneficial effects of the above technology are: based on the comparison diagram of each type of environmental parameters between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameters between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and the specific moments when the change trend of each type of environmental parameters of each similar distributed photovoltaic unit within a preset time period before the current moment is similar to the change trend of the corresponding type of environmental parameters of the distributed photovoltaic unit at the current moment are screened out, which is convenient for the subsequent construction of the predicted power matrix of the distributed photovoltaic unit. Embodiment 6:

[0104] Based on Example 1, the AI-based distributed photovoltaic power generation power prediction method, S3: based on the association time of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, obtain the predicted power matrix of the distributed photovoltaic unit, and based on the predicted power matrix of the distributed photovoltaic unit, obtain the predicted matching degree of each associated distributed photovoltaic unit, including:

[0105] If the number of association moments of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is greater than zero, it is determined that the corresponding similar distributed photovoltaic units are associated in the corresponding type of environmental parameters;

[0106] If each similar distributed photovoltaic unit is associated in all class environmental parameters, the corresponding similar distributed photovoltaic unit is determined to be an associated distributed photovoltaic unit;

[0107] Obtaining a predicted power matrix of the distributed photovoltaic unit based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units;

[0108] Based on the predicted power matrix of the distributed photovoltaic units, the predicted matching degree of each associated distributed photovoltaic unit is obtained.

[0109] The beneficial effects of the above technology are: according to the association time of all types of environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, all associated distributed photovoltaic units are obtained, and then according to the association time of all types of environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, which is convenient for the subsequent calculation of the predicted matching degree, and finally the predicted matching degree of each associated distributed photovoltaic unit is obtained according to the predicted power matrix of the distributed photovoltaic unit. Embodiment 7:

[0110] On the basis of Example 6, the AI-based distributed photovoltaic power generation prediction method obtains the predicted power matrix of the distributed photovoltaic unit based on the association time of all environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, including:

[0111] The association moment closest to the current moment in terms of time sequence among all association moments of each type of environmental parameter between the distributed photovoltaic unit and each associated distributed photovoltaic unit is regarded as the selected association moment of the corresponding type of environmental parameter between the distributed photovoltaic unit and the corresponding associated distributed photovoltaic unit;

[0112] All associated distributed photovoltaic units are ordinal-defined starting from 1 in order of distance from the distributed photovoltaic units from small to large, to obtain an ordinal definition result;

[0113] Based on the selected association moments and ordinal definition results of all class environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, which is:

[0114] ;

[0115] in, is the predicted power matrix of distributed photovoltaic units, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, where n is the number of associated distributed photovoltaic units.

[0116] The beneficial effect of the above technology is: according to the association moment of all types of environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, which is convenient for the subsequent calculation of the predicted matching degree. This embodiment gives in detail a specific method for constructing the predicted power matrix of the distributed photovoltaic unit.

[0117] Embodiment 8:

[0118] On the basis of Example 7, the AI-based distributed photovoltaic power generation prediction method obtains the prediction matching degree of each associated distributed photovoltaic unit based on the prediction power matrix of the distributed photovoltaic unit, including:

[0119] ;

[0120] in, The ordinal number is The predicted matching degree of the associated distributed photovoltaic units, is the value of the power generated by the distributed photovoltaic unit at the current moment, is a distributed photovoltaic unit and the ordinal number is The sum of the values ​​of the power generation at the selected association moment of all types of environmental parameters between the associated distributed photovoltaic units, is the rank of the predicted power matrix of the distributed photovoltaic unit, is a distributed photovoltaic unit and the ordinal number is The value of the power generated at the selected associated moment of the solar radiation intensity between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient temperature between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient wind speed between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The maximum value of the power generation at the selected association moment of all types of environmental parameters between the associated distributed photovoltaic units, is the natural logarithm, and the value of the natural constant e is 2.718.

[0121] The beneficial effects of the above technology are: based on the predicted power matrix of the distributed photovoltaic units, the predicted matching degree of each associated distributed photovoltaic unit is obtained, and the correlation of each associated distributed photovoltaic unit with respect to the distributed photovoltaic unit in terms of predicted power generation is quantified, which provides a basis for selecting the predicted distributed photovoltaic units and further improves the pertinence and accuracy of the prediction.

[0122] Embodiment 9:

[0123] Based on Example 1, the distributed photovoltaic power generation prediction method based on AI, S4: based on the prediction matching degree of all associated distributed photovoltaic units, obtain the predicted distributed photovoltaic unit, and based on the predicted distributed photovoltaic unit, obtain the predicted time of the distributed photovoltaic unit, and based on the predicted time of the distributed photovoltaic unit, obtain the photovoltaic power generation prediction result of the distributed photovoltaic unit, including:

[0124] The associated distributed photovoltaic unit with the largest prediction matching degree among all associated distributed photovoltaic units is regarded as the predicted distributed photovoltaic unit;

[0125] The selected associated time of the corresponding class environmental parameter with the largest power generation among the power generation at the selected associated time of all class environmental parameters between the distributed photovoltaic unit and the predicted distributed photovoltaic unit is regarded as the predicted time of the distributed photovoltaic unit;

[0126] The power generation power of the distributed photovoltaic unit at the prediction time is predicted as the photovoltaic power generation power prediction result of the distributed photovoltaic unit.

[0127] The beneficial effects of the above technology are as follows: based on the predicted matching degree of all associated distributed photovoltaic units, the predicted distributed photovoltaic unit is obtained, and the associated distributed photovoltaic unit that can most accurately predict the power generation power of the distributed photovoltaic unit is accurately selected, and then based on the predicted distributed photovoltaic unit, the predicted time of the distributed photovoltaic unit is obtained, and finally based on the predicted time of the distributed photovoltaic unit, the photovoltaic power generation power prediction result of the distributed photovoltaic unit is obtained. With reference to less previous data, the photovoltaic power generation power of the distributed photovoltaic unit is accurately predicted, the requirement for data volume is reduced, and effective prediction can be performed on distributed photovoltaic units with scarce data or newly installed.

[0128] Embodiment 10:

[0129] The present invention provides an AI-based distributed photovoltaic power generation prediction system, which is used to execute any one of the AI-based distributed photovoltaic power generation prediction methods in embodiments 1 to 9, referring to Figure 2 ,include:

[0130] A first processing module is used to obtain all similar distributed photovoltaic units based on the characteristics of all battery-like components of the distributed photovoltaic units at the current moment;

[0131] A second processing module is used to obtain a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, and obtain an association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology;

[0132] A calculation module, for obtaining a predicted power matrix of the distributed photovoltaic unit based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, and obtaining a predicted matching degree of each associated distributed photovoltaic unit based on the predicted power matrix of the distributed photovoltaic unit;

[0133] The prediction module is used to obtain the predicted distributed photovoltaic unit based on the predicted matching degree of all associated distributed photovoltaic units, and obtain the predicted time of the distributed photovoltaic unit based on the predicted distributed photovoltaic unit, and obtain the photovoltaic power generation power prediction result of the distributed photovoltaic unit based on the predicted time of the distributed photovoltaic unit.

[0134] The beneficial effects of the above technology are as follows: based on the comparison diagram of each type of environmental parameters between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameters between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and the specific moments when the change trend of each type of environmental parameters of each similar distributed photovoltaic unit within the preset time period before the current moment is similar to the change trend of the corresponding type of environmental parameters of the distributed photovoltaic unit at the current moment are screened out, and then according to the association moment of all types of environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, which is convenient for the subsequent calculation of the predicted matching degree, and the predicted power matrix of each distributed photovoltaic unit is obtained according to the predicted power matrix of the distributed photovoltaic unit. The prediction matching degree of each associated distributed photovoltaic unit is used to quantify the correlation of each associated distributed photovoltaic unit with respect to the distributed photovoltaic unit in terms of predicted power generation. Then, according to the prediction matching degree of all associated distributed photovoltaic units, the predicted distributed photovoltaic unit is obtained, which is convenient for determining the prediction time of the subsequent distributed photovoltaic unit. According to the predicted distributed photovoltaic unit, the prediction time of the distributed photovoltaic unit is obtained. Finally, according to the prediction time of the distributed photovoltaic unit, the prediction result of the photovoltaic power generation of the distributed photovoltaic unit is obtained. With reference to less previous data, the photovoltaic power generation of the distributed photovoltaic unit is accurately predicted, which reduces the requirement for the amount of data and enables effective prediction on distributed photovoltaic units with scarce data or newly installed.

[0135] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention, and the present invention is also intended to include these changes and modifications.

Claims

1. A distributed photovoltaic power generation prediction method based on AI, characterized in that: include: S1: Based on the characteristics of all similar battery components of the distributed photovoltaic units at the current moment, all similar distributed photovoltaic units are obtained; S2: Based on all environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained; S3: based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, a prediction power matrix of the distributed photovoltaic unit is obtained, and based on the prediction power matrix of the distributed photovoltaic unit, a prediction matching degree of each associated distributed photovoltaic unit is obtained; S4: based on the prediction matching degree of all associated distributed photovoltaic units, a predicted distributed photovoltaic unit is obtained, and based on the predicted distributed photovoltaic unit, a predicted time of the distributed photovoltaic unit is obtained, and based on the predicted time of the distributed photovoltaic unit, a photovoltaic power generation power prediction result of the distributed photovoltaic unit is obtained; Wherein, S3: based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, a predicted power matrix of the distributed photovoltaic unit is obtained, and based on the predicted power matrix of the distributed photovoltaic unit, a predicted matching degree of each associated distributed photovoltaic unit is obtained, including: If the number of association moments of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is greater than zero, the corresponding similar distributed photovoltaic units are determined to be associated in the corresponding type of environmental parameters; if each similar distributed photovoltaic unit is associated in all types of environmental parameters, the corresponding similar distributed photovoltaic unit is determined to be an associated distributed photovoltaic unit; based on the association moments of all types of environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained; based on the predicted power matrix of the distributed photovoltaic unit, the predicted matching degree of each associated distributed photovoltaic unit is obtained; Based on the association moments of all environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, a predicted power matrix of the distributed photovoltaic unit is obtained, including: The association moment closest to the current moment in time sequence among all association moments of each type of environmental parameter between the distributed photovoltaic unit and each associated distributed photovoltaic unit is regarded as the selected association moment of the corresponding type of environmental parameter between the distributed photovoltaic unit and the corresponding associated distributed photovoltaic unit; all associated distributed photovoltaic units are ordinal defined starting from 1 in order of distance from the distributed photovoltaic unit from small to large, and the ordinal definition result is obtained; based on the selected association moments of all types of environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units and the ordinal definition results, the predicted power matrix of the distributed photovoltaic unit is obtained, which is: ； in, is the predicted power matrix of distributed photovoltaic units, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, where n is the number of associated distributed photovoltaic units; Based on the predicted power matrix of the distributed photovoltaic units, the predicted matching degree of each associated distributed photovoltaic unit is obtained, including: ； in, is the predicted matching degree of the associated distributed photovoltaic unit with ordinal number, is the value of the power generated by the distributed photovoltaic unit at the current moment, is a distributed photovoltaic unit and the ordinal number is The sum of the values ​​of the power generation at the selected associated moment of all types of environmental parameters between the associated distributed photovoltaic units, is the rank of the predicted power matrix of the distributed photovoltaic unit, is a distributed photovoltaic unit and the ordinal number is The value of the power generated at the selected associated moment of the solar radiation intensity between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient temperature between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient wind speed between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The maximum value of the power generation at the selected association moment of all types of environmental parameters between the associated distributed photovoltaic units, is the natural logarithm, and the value of the natural constant e is 2.

718.

2. The AI-based distributed photovoltaic power generation prediction method according to claim 1 is characterized in that: S1: Based on the characteristics of all similar battery components of the distributed photovoltaic unit at the current moment, all similar distributed photovoltaic units are obtained, including: Obtaining the characteristics of all types of battery components of the distributed photovoltaic unit at the current moment, wherein the characteristics of all types of battery components include the number of battery cells, the area of ​​the battery cells and the photoelectric conversion efficiency of the battery components; Treat all distributed photovoltaic units connected to the AI-based distributed photovoltaic power generation prediction system as standby distributed photovoltaic units; When the difference between the value of the product of the number of cells and the cell area of ​​each prepared distributed photovoltaic unit at the current moment and the value of the product of the number of cells and the cell area of ​​the distributed photovoltaic unit at the current moment is less than the preset first difference, and the difference between the photoelectric conversion efficiency of the battery assembly of the corresponding prepared distributed photovoltaic unit at the current moment and the photoelectric conversion efficiency of the battery assembly of the distributed photovoltaic unit at the current moment is less than the preset second difference, the corresponding prepared distributed photovoltaic unit is regarded as a similar distributed photovoltaic unit, and all similar distributed photovoltaic units are obtained.

3. The AI-based distributed photovoltaic power generation prediction method according to claim 1 is characterized in that: S2: Based on all environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the associated moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, including: Obtaining all class environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, wherein all class environmental parameters include solar radiation intensity, ambient temperature and ambient wind speed; Taking each moment of the distributed photovoltaic unit in the preset time period before the current moment as the horizontal coordinate value, and each type of environmental parameter of the distributed photovoltaic unit at the corresponding moment in the preset time period before the current moment as the vertical coordinate value, each type of environmental parameter point of the distributed photovoltaic unit is obtained; Take each moment of each similar distributed photovoltaic unit in the preset time period before the current moment as the horizontal coordinate value, and each type of environmental parameter of the corresponding moment of the similar distributed photovoltaic unit in the preset time period before the current moment as the vertical coordinate value, and obtain each type of environmental parameter point of each similar distributed photovoltaic unit; Based on each type of environmental parameter point of the distributed photovoltaic unit and each type of environmental parameter point of each similar distributed photovoltaic unit, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, and based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained.

4. The AI-based distributed photovoltaic power generation prediction method according to claim 3 is characterized in that: Based on each type of environmental parameter point of the distributed photovoltaic unit and each type of environmental parameter point of each similar distributed photovoltaic unit, a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, including: Connect each type of environmental parameter point of the distributed photovoltaic unit in the order of the horizontal coordinate from small to large, and obtain the connection line of each type of environmental parameter of the distributed photovoltaic unit, and connect each type of environmental parameter point of each similar distributed photovoltaic unit in the order of the horizontal coordinate from small to large, and obtain the connection line of each type of environmental parameter of each similar distributed photovoltaic unit; The images of the connection lines of each type of environmental parameters of the distributed photovoltaic unit and the connection lines of the corresponding type of environmental parameters of each similar distributed photovoltaic unit in the same coordinate system are regarded as the comparison diagram of the corresponding type of environmental parameters between the distributed photovoltaic unit and the corresponding similar distributed photovoltaic unit.

5. The AI-based distributed photovoltaic power generation prediction method according to claim 4 is characterized in that: Based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology, the association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is obtained, including: Based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and the preset curve judgment model, all connecting line segments similar to the end line segments of the connecting line of each type of environmental parameter of each similar distributed photovoltaic unit are obtained as the associated line segments of the connecting line of the corresponding type of environmental parameter of the corresponding similar distributed photovoltaic unit; The moments corresponding to all associated line segments of the connecting line of each type of environmental parameter of each similar distributed photovoltaic unit are regarded as the associated moments of the corresponding type of environmental parameters between the distributed photovoltaic unit and the corresponding similar distributed photovoltaic unit.

6. The AI-based distributed photovoltaic power generation prediction method according to claim 1 is characterized in that: S4: Based on the prediction matching degree of all associated distributed photovoltaic units, a predicted distributed photovoltaic unit is obtained, and based on the predicted distributed photovoltaic unit, a predicted time of the distributed photovoltaic unit is obtained, and based on the predicted time of the distributed photovoltaic unit, a photovoltaic power generation power prediction result of the distributed photovoltaic unit is obtained, including: The associated distributed photovoltaic unit with the largest prediction matching degree among all associated distributed photovoltaic units is regarded as the predicted distributed photovoltaic unit; The selected associated time of the corresponding class environmental parameter with the largest power generation among the power generation at the selected associated time of all class environmental parameters between the distributed photovoltaic unit and the predicted distributed photovoltaic unit is regarded as the predicted time of the distributed photovoltaic unit; The power generation power of the distributed photovoltaic unit at the prediction time is predicted as the photovoltaic power generation power prediction result of the distributed photovoltaic unit.

7. An AI-based distributed photovoltaic power generation prediction system, characterized in that: A distributed photovoltaic power generation power prediction method based on AI for executing any one of claims 1 to 6, comprising: A first processing module is used to obtain all similar distributed photovoltaic units based on the characteristics of all battery-like components of the distributed photovoltaic units at the current moment; A second processing module is used to obtain a comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on all types of environmental parameters of the distributed photovoltaic unit and all similar distributed photovoltaic units at each moment in a preset time period before the current moment, and obtain an association moment of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit based on the comparison diagram of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit and AI technology; A calculation module, for obtaining a predicted power matrix of the distributed photovoltaic unit based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, and obtaining a predicted matching degree of each associated distributed photovoltaic unit based on the predicted power matrix of the distributed photovoltaic unit; A prediction module is used to obtain a predicted distributed photovoltaic unit based on the prediction matching degree of all associated distributed photovoltaic units, and obtain a predicted time of the distributed photovoltaic unit based on the predicted distributed photovoltaic unit, and obtain a photovoltaic power generation power prediction result of the distributed photovoltaic unit based on the predicted time of the distributed photovoltaic unit; Among them, based on the association moments of all class environmental parameters between the distributed photovoltaic unit and all similar distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained, and based on the predicted power matrix of the distributed photovoltaic unit, the predicted matching degree of each associated distributed photovoltaic unit is obtained, including: If the number of association moments of each type of environmental parameter between the distributed photovoltaic unit and each similar distributed photovoltaic unit is greater than zero, the corresponding similar distributed photovoltaic units are determined to be associated in the corresponding type of environmental parameters; if each similar distributed photovoltaic unit is associated in all types of environmental parameters, the corresponding similar distributed photovoltaic unit is determined to be an associated distributed photovoltaic unit; based on the association moments of all types of environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, the predicted power matrix of the distributed photovoltaic unit is obtained; based on the predicted power matrix of the distributed photovoltaic unit, the predicted matching degree of each associated distributed photovoltaic unit is obtained; Based on the association moments of all environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units, a predicted power matrix of the distributed photovoltaic unit is obtained, including: The association moment closest to the current moment in time sequence among all association moments of each type of environmental parameter between the distributed photovoltaic unit and each associated distributed photovoltaic unit is regarded as the selected association moment of the corresponding type of environmental parameter between the distributed photovoltaic unit and the corresponding associated distributed photovoltaic unit; all associated distributed photovoltaic units are ordinal defined starting from 1 in order of distance from the distributed photovoltaic unit from small to large, and the ordinal definition result is obtained; based on the selected association moments of all types of environmental parameters between the distributed photovoltaic unit and all associated distributed photovoltaic units and the ordinal definition results, the predicted power matrix of the distributed photovoltaic unit is obtained, which is: ； in, is the predicted power matrix of distributed photovoltaic units, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 1, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number 2, is the value of the power generation at the selected associated moment of the solar radiation intensity between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association time of the ambient temperature between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, is the value of the power generation at the selected association moment of the ambient wind speed between the distributed photovoltaic unit and the associated distributed photovoltaic unit with ordinal number n, where n is the number of associated distributed photovoltaic units; Based on the predicted power matrix of the distributed photovoltaic units, the predicted matching degree of each associated distributed photovoltaic unit is obtained, including: ； in, is the predicted matching degree of the associated distributed photovoltaic unit with ordinal number, is the value of the power generated by the distributed photovoltaic unit at the current moment, is a distributed photovoltaic unit and the ordinal number is The sum of the values ​​of the power generation at the selected associated moment of all types of environmental parameters between the associated distributed photovoltaic units, is the rank of the predicted power matrix of the distributed photovoltaic unit, is a distributed photovoltaic unit and the ordinal number is The value of the power generated at the selected associated moment of the solar radiation intensity between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient temperature between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The value of the power generation at the selected associated moment of the ambient wind speed between the associated distributed photovoltaic units, is a distributed photovoltaic unit and the ordinal number is The maximum value of the power generation at the selected association moment of all types of environmental parameters between the associated distributed photovoltaic units, is the natural logarithm, and the value of the natural constant e is 2.718.

Images