Power calculation model training method, calculation method, device, equipment and medium

Abstract

The present invention relates to the field of new energy technology, and discloses a power calculation model training method, a calculation method, a device, an equipment and a medium. The present invention obtains historical power data, historical irradiance data and historical meteorological data of a target photovoltaic power station; calculates the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data; determines multiple historical difference time intervals of the target photovoltaic power station based on the historical difference degree; extracts the historical power data, historical irradiance data and historical meteorological data based on each historical difference time interval to obtain a training data set under each historical difference time interval; trains a preset power calculation model based on the training data set, and during the training process, learns the corresponding relationship between the historical irradiance data, historical meteorological data and historical power data under each historical difference degree to obtain a photovoltaic power calculation model corresponding to different historical difference time intervals.

Description

Technical Field

[0001] The present invention relates to the field of new energy technologies, and in particular to a power calculation model training method, a calculation method, a device, equipment and a medium. Background Art

[0002] At present, photovoltaic power stations often use the total irradiance of meteorological monitoring stations and adopt physical methods to convert the measured horizontal plane irradiance intensity into the inclined plane irradiance intensity of photovoltaic modules. Taking into account factors such as the location of the photovoltaic power station, the characteristics of different photovoltaic modules and the installation method, a photoelectric conversion model of photovoltaic cells is established to obtain the theoretical power of the photovoltaic power station.

[0003] When this method is used, the fixed irradiance value is usually in one-to-one correspondence with the theoretical power of the photovoltaic power station; under the same irradiance conditions, the output of the photovoltaic panels is also mapped one-to-one with the fixed setting output; however, since the irradiance in the four seasons of spring, summer, autumn and winter often fluctuates regularly, the irradiance in spring, summer, autumn and winter is the same but the actual output of the photovoltaic panels is not consistent, the calculation result is inaccurate when the above method is used for power calculation; and the output of the photovoltaic power station is different under different weather conditions. Therefore, the above method has the problem of relatively low overall accuracy when calculating power due to failure to consider seasonal and weather differences. Summary of the invention

[0004] In view of this, the present invention provides a power calculation model training method, calculation method, device, equipment and medium to solve the problem of relatively low overall accuracy due to failure to consider seasonal differences and weather differences when calculating power.

[0005] In a first aspect, the present invention provides a power calculation model training method, and the power calculation model construction method includes: obtaining historical power data, historical irradiance data and historical meteorological data of a target photovoltaic power station; calculating the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data; determining multiple historical difference time intervals of the target photovoltaic power station based on the historical difference degree; extracting historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals; training a preset power calculation model based on the training data set, and during the training process, learning the correspondence between the historical irradiance data, the historical meteorological data and the historical power data under each of the historical difference degrees to obtain a photovoltaic power calculation model corresponding to different historical difference time intervals.

[0006] In an optional embodiment, determining multiple historical difference time intervals of the photovoltaic power station based on the historical difference degree includes: clustering the difference degree to obtain clustering results; obtaining multiple historical difference degree groups with different similarities and their corresponding timestamps based on the clustering results; and determining the historical difference time interval based on the timestamps.

[0007] In an optional embodiment, clustering the difference degrees to obtain clustering results includes: intercepting the historical difference degrees according to a preset time scale to obtain a historical difference degree interval containing a timestamp; clustering the historical difference degree interval to obtain a clustering result. In an optional embodiment, determining multiple historical difference time intervals of the photovoltaic power station based on the historical difference degrees also includes: sorting the historical difference degrees by size to obtain a sorting result; dividing the historical difference degrees based on the sorting result to obtain historical difference degrees of different sizes and their corresponding time intervals; and using the time interval as the historical difference time interval.

[0008] In an optional embodiment, the power calculation model training method also includes: obtaining the irradiation change trend of the historical irradiance data according to a preset time dimension; determining the historical peak time and the historical trough time of the historical irradiance based on the irradiation change trend; determining the historical difference time interval based on the historical peak time, the historical trough time and the historical difference degree.

[0009] In an optional embodiment, the training of the preset power calculation model based on the training data set includes: extracting the historical irradiance data and the historical meteorological data in the training data set; calculating the horizontal plane total radiation data based on the historical irradiance data; extracting the historical temperature data and the historical humidity data in the historical meteorological data; calculating the historical temperature and humidity ratio data based on the historical temperature data and the historical humidity data; and inputting the horizontal plane total radiation data, the historical temperature and humidity ratio data, and the historical meteorological data into each preset power calculation model for model training.

[0010] In a second aspect, the present invention provides a photovoltaic power calculation method, which includes: obtaining the irradiance data and meteorological data of the target photovoltaic power station and their timestamps; selecting a corresponding photovoltaic power calculation model based on the historical difference time interval corresponding to the timestamp; wherein the photovoltaic power calculation model is trained by the corresponding historical meteorological data, historical irradiance data and historical power data in the historical difference time interval; inputting the irradiance data and the meteorological data into the photovoltaic power calculation model for calculation to obtain a photovoltaic power calculation result.

[0011] In a third aspect, the present invention provides a power calculation model training device, the device comprising: an acquisition module for acquiring historical power data, historical irradiance data and historical meteorological data of a target photovoltaic power station; a difference degree determination module for calculating the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data; a time interval determination module for determining multiple historical difference time intervals of the target photovoltaic power station based on the historical difference degree; a training data set determination module for extracting historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals; a model training module for training a preset power calculation model based on the training data set, and during the training process, learning the correspondence between the historical irradiance data, the historical meteorological data and the historical power data under each of the historical difference degrees to obtain a photovoltaic power calculation model corresponding to different historical difference time intervals.

[0012] In a fourth aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the power calculation model training method or the power calculation method of the above-mentioned first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0013] In a fifth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the power calculation model training method or the power calculation method of the above-mentioned first aspect or any corresponding embodiment thereof.

[0014] The present invention provides a power calculation model training method, the power calculation model construction method includes: obtaining historical power data, historical irradiance data and historical meteorological data of a target photovoltaic power station; calculating the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data; determining multiple historical difference time intervals of the target photovoltaic power station based on the historical difference degree; extracting historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals; training a preset power calculation model based on the training data set, and learning the corresponding relationship between the historical irradiance data, the historical meteorological data and the historical power data under each of the historical difference degrees during the training process, so as to obtain a photovoltaic power calculation model corresponding to different historical difference time intervals. In the above method of this embodiment, the photovoltaic power calculation model is determined by the data set divided by different historical difference time intervals, and the output time interval is determined by the historical difference degree of the target photovoltaic power station. This determination method can reflect the influence of irradiance data and power data caused by different time, and improves the accuracy of theoretical power calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 is a schematic diagram of irradiance and photovoltaic power generation according to an embodiment of the present invention;

[0017] Figure 2 According to an embodiment of the present invention, a 60000MJ / m 2 Schematic diagram of photovoltaic power generation corresponding to irradiance;

[0018] Figure 3 is a flowchart of a power calculation model training method according to an embodiment of the present invention;

[0019] Figure 4 is a flow chart of a power calculation method according to an embodiment of the present invention;

[0020] Figure 5 is a schematic diagram of a power calculation model training device according to an embodiment of the present invention;

[0021] Figure 6 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0023] According to an embodiment of the present invention, an embodiment of a power calculation model training method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0024] At present, photovoltaic power stations often use the total irradiance of meteorological monitoring stations and adopt physical methods to convert the measured horizontal plane irradiance intensity into the inclined plane irradiance intensity of photovoltaic modules. Taking into account factors such as the location of the photovoltaic power station, the characteristics of different photovoltaic modules and the installation method, a photoelectric conversion model of photovoltaic cells is established to obtain the theoretical power of the photovoltaic power station.

[0025] When this method is used, the fixed irradiance value is usually mapped one-to-one with the theoretical power of the photovoltaic power station; under the same irradiance, the output of the photovoltaic panel is also mapped one-to-one with the fixed setting output; however, the inventors have found that since the irradiance often fluctuates regularly in the four seasons of spring, summer, autumn and winter, the irradiance is the same in spring, summer, autumn and winter, but the actual photovoltaic panel output is not consistent, and the calculation result is inaccurate when the above method is used for power calculation; for example, Figure 1 is a schematic diagram of irradiance and photovoltaic power generation according to an embodiment of the present invention. Figure 2 According to an embodiment of the present invention, a 60000MJ / m 2 Schematic diagram of photovoltaic power generation under irradiance, such as Figure 1 , Figure 2 As shown, at 60000MJ / m 2 Under different irradiances, the corresponding photovoltaic power generation powers are 23800, 23000, 24500 and 23500 respectively, that is, the photovoltaic power generation powers are different under the same irradiance; and the difference in photovoltaic power generation powers is not greatly affected by temperature; for example, the irradiance is 60000MJ / m 2The times are around mid-February, around March 23, around May 23, and around August 23. The temperature in March is higher than that in February, but the photovoltaic power generation is lower than that in February; the temperature in February is lower than that in August, but the power generation is higher than that in August; therefore, the difference in power generation is not significantly affected by temperature changes.

[0026] Also, see Figure 1 , Figure 2 As shown, the temperature from January to March is generally higher than that from March to April, but the irradiance gradually decreases from March to April; the temperature from April to May is generally higher than that from January to April, but the irradiance is relatively low; the temperature in June rises slowly relative to the temperature in May, but the irradiance rises with a larger slope and a faster rate since June; further, July to September should be the period with the highest temperature throughout the year, but the irradiance shows a downward trend and is relatively low in May; therefore, there is no positive or negative correlation between irradiance and temperature; the inventors have found that factors such as weather conditions, irradiance conditions, ice coverage or dust accumulation of photovoltaic panels will affect the consistency between irradiance and power generation, but there is no linear influence relationship between the change in power generation and the above-mentioned single influencing factors or their combined influencing factors.

[0027] Based on this, the historical difference degrees between the historical power data and the historical irradiance data at different times are different, and the historical difference degrees are not related to time, temperature, irradiance, etc.; and the calculation method of using a fixed irradiance value and a one-to-one correspondence with the theoretical power of the photovoltaic power station to obtain the power data of the photovoltaic power station under fixed irradiance data is inaccurate.

[0028] Based on this, a power calculation model training method is provided in this embodiment. Figure 3 is a flow chart of a power calculation model training method according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0029] Step S101, obtaining historical power data, historical irradiance data and historical meteorological data of a target photovoltaic power station.

[0030] In this embodiment, the historical power data may be the power generation of the photovoltaic panel recorded by the photovoltaic panel control module under a preset time dimension; the power generation power may be a time-series power generation power with a timestamp; the historical irradiance data and the historical meteorological data may be recorded by instruments under a preset time dimension; as a possible implementation method, the preset time dimension may be 1, 2, 3, ... N years, where N is a positive integer, and the data resolution of the historical power data, historical irradiance data and historical meteorological data may be minutes; for example, the historical power data, historical irradiance data and historical meteorological data may be obtained by collecting data every 15 minutes.

[0031] Step S102: calculating the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data.

[0032] In this embodiment, after obtaining the historical irradiance data, the historical ideal power generation corresponding to the target photovoltaic power station under the historical irradiance data is calculated. The ideal power generation can be a mathematical calculation result under ideal conditions without considering factors such as weather influence and physical loss. After obtaining the historical theoretical power generation, the historical theoretical power generation and the historical irradiance data are divided to calculate the historical difference degree of the photovoltaic power station.

[0033] In this embodiment, the timestamps of the historical difference degrees match the timestamps of the historical power data and the historical irradiance data.

[0034] Step S103: determining a plurality of historical difference time intervals of the target photovoltaic power station based on the historical difference degree.

[0035] As mentioned above, the historical difference degrees of the historical power data and the historical irradiance data at different times are different, and the historical difference degrees are not related to time, temperature, irradiance, etc.; the calculation method of obtaining the power data of the photovoltaic power station under fixed irradiance data by using a one-to-one correspondence between a fixed irradiance value and the theoretical power of the photovoltaic power station is inaccurate, and factors such as weather conditions, irradiance conditions, and ice coverage or dust accumulation of photovoltaic panels under the same irradiance will affect the output of the photovoltaic panels, but there is no linear influence relationship between the change in power generation and the above-mentioned single influencing factors or their combined influencing factors; therefore, in the present invention, in order to solve this problem, after obtaining the historical difference degree, multiple historical difference time intervals of the target photovoltaic power station can be determined based on the historical difference degree, wherein different historical difference degrees can reflect the influence of factors such as different weather conditions, different irradiance conditions, different ice coverage or different dust accumulation states of photovoltaic panels caused by different times on the irradiance data and power data.

[0036] For example, the degree of clustering of historical difference degrees within the historical period can be analyzed to obtain the historical difference time intervals corresponding to different historical difference degrees, and different photovoltaic power calculation models can be trained for different historical difference time intervals to perform calculations, so that the photovoltaic theoretical power calculation can adapt to the changes in the historical difference degree.

[0037] Step S104, extracting historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals.

[0038] In this embodiment, after the historical difference time intervals corresponding to different historical difference degrees are obtained, the historical power data, the historical irradiance data and the historical meteorological data are extracted based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals; as a possible implementation method, the above data can be extracted through the time series correspondence between the output time interval and the historical power data, the historical irradiance data and the historical meteorological data to obtain a training data set; for example, when the historical difference degree is A, the output time interval corresponding to the output time interval is [A1 month, A2 month], [A3 month, A4 month]..., then the output time interval belonging to [A1 month, A The historical power data, historical irradiance data and historical meteorological data at the time of [A1 month, A2 month] and [A3 month, A4 month] are extracted as the training data set at the time of [A1 month, A2 month] and [A3 month, A4 month] with the historical difference degree of A; when the historical difference degree is B, the corresponding output time interval is [B1 month, B2 month], [B3 month, B4 month]…, then the historical power data, historical irradiance data and historical meteorological data belonging to the time of [B1 month, B2 month] and [B3 month, B4 month] are extracted as the training data set at the time of [B1 month, B2 month] and [B3 month, B4 month] with the historical difference degree of B….

[0039] Step S105, training the preset power calculation model based on the training data set, and during the training process, learning the correspondence between the historical irradiance data, the historical meteorological data and the historical power data under each historical difference degree, so as to obtain the photovoltaic power calculation model corresponding to different historical difference time intervals.

[0040] In this embodiment, the preset power calculation model can be a machine learning model or a deep learning model; as a possible implementation method, the preset power calculation model is an XGB model; specifically, XGB uses a decision tree (Classification and Regression Trees, CART) as a basic learner, wherein CART is a tree structure that predicts by recursively dividing features; in XGB, each decision tree is a weak classifier, and a strong classifier is constructed by integrating multiple decision trees; the residuals of all previous trees (or the gradient of the objective function) are corrected by continuously training new decision trees; each new decision tree will try to fit the error between the previous model prediction result and the actual value, and ultimately improve the performance of the overall model; in this embodiment, the XGB model is used to learn the correspondence between the historical irradiance data, the historical meteorological data and the historical power data under each of the historical difference degrees; during the training process, the historical irradiance data and the historical meteorological data are used as input to obtain the power data output by the model, and the model parameters of XGB are continuously adjusted according to the difference between the power data output by the model and the historical power data until the model converges.

[0041] The power calculation model training method of this embodiment obtains historical power data, historical irradiance data and historical meteorological data of the target photovoltaic power station; calculates the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data; determines multiple historical difference time intervals of the target photovoltaic power station based on the historical difference degree; extracts historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals; trains a preset power calculation model based on the training data set, and during the training process, learns the corresponding relationship between the historical irradiance data, the historical meteorological data and the historical power data under each of the historical difference degrees to obtain a photovoltaic power calculation model corresponding to different historical difference time intervals; in the above method of this embodiment, the photovoltaic power calculation model is determined by the data set divided by different historical difference time intervals, and the output time interval is determined by the historical difference degree of the target photovoltaic power station. This determination method can reflect the influence of factors such as different weather conditions, different irradiance conditions, different ice coverage or different dust accumulation conditions of photovoltaic panels caused by different times on irradiance data and power data, thereby improving the accuracy of theoretical power calculation.

[0042] In an optional embodiment, determining multiple historical difference time intervals of the photovoltaic power station based on the historical difference degree includes: clustering the difference degree to obtain clustering results; obtaining multiple historical difference degree groups with different similarities and their corresponding timestamps based on the clustering results; and determining the historical difference time interval based on the timestamps.

[0043] In this embodiment, after obtaining the degree of difference, the degree of difference is clustered; as a possible implementation method, K clustering, hierarchical clustering, etc. can be adopted; after obtaining the clustering results, a plurality of historical difference degree groups with different similarities and their corresponding timestamps are obtained based on the clustering results; for example, the historical difference degree groups obtained by clustering may be [70%, 71%), [71%, 74%), [74%, 77%), [77%, 78%), [78%, 80%), and the corresponding timestamps are the timestamps corresponding to the historical difference degrees; it should be understood that the timestamp may be a month, for example, if the daily historical difference degrees of the target photovoltaic power station in January, February and July are all [70%, 72%), then the historical difference time intervals with the historical difference degrees of [70%, 72%) are January, February and July.

[0044] In an optional implementation, clustering is performed on the difference degrees to obtain clustering results, including: intercepting the historical difference degrees according to a preset time scale to obtain a historical difference degree interval containing a timestamp; clustering is performed on the historical difference degree interval to obtain a clustering result.

[0045] In this embodiment, after obtaining the historical difference degree, the historical difference degree is intercepted according to a preset time scale to obtain a historical difference degree interval containing a timestamp; wherein the preset time scale may be greater than the data resolution of the historical power data and the historical irradiance data; exemplarily, the preset time scale may be set to days or months.

[0046] In an optional embodiment, the method of determining multiple historical difference time intervals of the photovoltaic power station based on the historical difference degree also includes: sorting the historical difference degrees by size to obtain a sorting result; dividing the historical difference degrees based on the sorting result to obtain historical difference degrees of different sizes and their corresponding time intervals; and using the time interval as the historical difference time interval.

[0047] In this embodiment, after the historical difference degrees are sorted by size, the historical difference degrees can be divided according to the sorting results to obtain historical difference degrees of different sizes and their corresponding time intervals; for example, after the historical difference degrees are sorted by size, the historical difference degrees can be divided into N levels, and different sizes of historical difference degrees and their corresponding time intervals are set under each level; this division method has less computational complexity than clustering.

[0048] In an optional embodiment, the power calculation model training method also includes: obtaining the irradiation change trend of the historical irradiance data according to a preset time dimension; determining the historical peak time and the historical trough time of the historical irradiance based on the irradiation change trend; determining the historical difference time interval based on the historical peak time, the historical trough time and the historical difference degree.

[0049] like Figure 1 As shown, the historical peaks of the historical irradiance data appear in March and June, and the historical trough appears in September. Near the peak in March, the historical difference degrees of February, March and April are similar to each other. At this time, February, March and April are divided into the first historical difference time interval; near the peak in June, the historical difference degrees of May, June and July are similar to each other, and are different from the historical difference degrees in the first historical difference time interval. At this time, May, June and July are divided into the second historical difference time interval; near the trough in September, the historical difference degrees of August, September, October and November are similar to each other, and are different from the historical difference degrees in the first historical difference time interval and the second historical difference time interval. At this time, August, September, October and November are divided into the third historical difference time interval; the historical difference degrees of November and January are similar to each other, and are different from the historical difference degrees in other historical difference time intervals. Therefore, November and January are divided into the fourth historical difference time interval.

[0050] In an optional embodiment, the training of the preset power calculation model based on the training data set includes: extracting the historical irradiance data and the historical meteorological data in the training data set; calculating the horizontal plane total radiation data based on the historical irradiance data; extracting the historical temperature data and the historical humidity data in the historical meteorological data; calculating the historical temperature and humidity ratio data based on the historical temperature data and the historical humidity data; and inputting the horizontal plane total radiation data, the historical temperature and humidity ratio data, and the historical meteorological data into each preset power calculation model for model training.

[0051] Since the incident angle of sunlight is different in different seasons, the degree of radiation received by photovoltaic panels under the same irradiance is also different. To solve this problem, in this embodiment, the horizontal total radiation data is calculated based on the historical irradiance data to minimize the impact of different light angles on the theoretical power calculation results.

[0052] In an optional embodiment, after obtaining the training data sets under each of the historical difference time intervals, the method further includes: extracting sub-historical irradiance data in each of the training data sets; sorting each sub-historical irradiance data, extracting the sub-historical irradiance data within a preset interval before the sequence and averaging them as weather reference irradiance data; dividing the training data set based on the size of each sub-historical irradiance data and the weather reference irradiance data to obtain sub-training data sets under different weather types; and performing model training based on the sub-training data sets.

[0053] In the present invention, in order to consider the impact of the output conditions of photovoltaic panels in different weather conditions on power calculation, the weather conditions of the training data set are divided according to the irradiance data; it should be understood that different cloud cover under different weather conditions will result in different irradiances, and the irradiances on sunny days, rainy days and cloudy days should decrease in turn; based on this, in the present invention, the weather types can be divided by sorting the sub-historical irradiance data; specifically, after extracting the sub-historical irradiance data in each training data set, each sub-historical irradiance data is sorted, and the top 20% of the daily irradiance is extracted and averaged to obtain the weather benchmark irradiance data, and further, the irradiance corresponding to sunny weather should be the weather benchmark irradiance, the irradiance corresponding to cloudy weather should be 0.5 to 0.9 times the weather benchmark irradiance data, and the irradiance corresponding to rainy weather should be less than 0.5 times the weather benchmark irradiance data, thereby obtaining sub-training data sets under different weather types according to the above division criteria.

[0054] In a second aspect, the present invention provides a photovoltaic power calculation method. Figure 4 is a flow chart of a photovoltaic power calculation method according to an embodiment of the present invention. Figure 4 As shown, the photovoltaic power calculation method includes:

[0055] S201, obtaining irradiance data and meteorological data of a target photovoltaic power station and their timestamps.

[0056] In this embodiment, the irradiance data and meteorological data can be measured data or predicted data; the measured data can be collected in advance by the instrument, and the predicted data can be determined by the weather forecast; when the irradiance data and meteorological data are measured data, the photovoltaic power calculation method is used to calculate the theoretical power of the target photovoltaic power station; when the irradiance data and meteorological data are predicted data, the photovoltaic power calculation method is used to calculate the future predicted power of the target photovoltaic power station.

[0057] S202, selecting a corresponding photovoltaic power calculation model based on the historical difference time interval corresponding to the timestamp; wherein the photovoltaic power calculation model is trained by the historical meteorological data, historical irradiance data and historical power data corresponding to the historical difference time interval.

[0058] In this embodiment, the photovoltaic power calculation model corresponding to the historical difference time interval of the irradiance data and the meteorological data is selected, which can take into account the difference in the degree of historical difference at the time of the input data, and has the advantage of more accurate theoretical power calculation.

[0059] S203, inputting the irradiance data and the meteorological data into the photovoltaic power calculation model for calculation to obtain a photovoltaic power calculation result.

[0060] The photovoltaic power calculation method of this embodiment obtains the irradiance data and meteorological data of the target photovoltaic power station and their timestamps; selects the corresponding photovoltaic power calculation model based on the historical difference time interval corresponding to the timestamp; wherein the photovoltaic power calculation model is trained by the historical meteorological data, historical irradiance data and historical power data corresponding to the historical difference time interval; the irradiance data and the meteorological data are input into the photovoltaic power calculation model for calculation to obtain the photovoltaic power calculation result; the photovoltaic power calculation model is determined by the data set divided by different historical difference time intervals, and the output time interval is determined by the historical difference degree of the target photovoltaic power station. This determination method can reflect the influence of factors such as different weather conditions, different irradiance conditions, different ice coverage or different dust accumulation states of photovoltaic panels caused by different times on the irradiance data and power data. When predicting, the correspondence between the meteorological data, irradiance data and power data in the historical difference time interval corresponding to the timestamp to which the target photovoltaic power station currently belongs can be considered, which has the advantage of more accuracy when predicting.

[0061] This invention provides a power calculation model training device, such as Figure 5 As shown, including:

[0062] An acquisition module 501 is used to acquire historical power data, historical irradiance data and historical meteorological data of a target photovoltaic power station;

[0063] A difference degree determination module 502 is used to calculate the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data;

[0064] A time interval determination module 503, configured to determine a plurality of historical difference time intervals of the target photovoltaic power station based on the historical difference degree;

[0065] A training data set determination module 504 is used to extract historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals;

[0066] The model training module 505 is used to train the preset power calculation model based on the training data set. During the training process, the correspondence between the historical irradiance data, the historical meteorological data and the historical power data under each historical difference degree is learned to obtain the photovoltaic power calculation model corresponding to different historical difference time intervals.

[0067] It should be noted here that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiments.

[0068] It should be noted that the above modules as part of the device can be implemented by software or hardware, wherein the hardware environment includes a network environment.

[0069] An embodiment of the present invention also provides a computer device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus, the memory is used to store a computer program; the processor is used to execute a method in any of the above embodiments by running the computer program stored in the memory.

[0070] Figure 6 is a structural block diagram of an optional computer device according to an embodiment of the present application, such as Figure 6 As shown, it includes a processor 10, a communication interface 20, a memory 30 and a communication bus 40, wherein the processor 10, the communication interface 20 and the memory 30 communicate with each other through the communication bus 40, wherein:

[0071] A memory 30, for storing computer programs;

[0072] The processor 10 is used to implement the method for constructing a disease database as described in any of the above embodiments when executing the computer program stored in the memory 30.

[0073] Optionally, in this embodiment, the communication bus may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0074] The communication interface is used for communication between the above-mentioned computer device and other devices.

[0075] The memory may include RAM, or may include non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0076] The above-mentioned processor can be a general-purpose processor, which can include but not be limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; it can also be DSP (Digital Signal Processing), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0077] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment will not be described in detail here.

[0078] It can be understood by those skilled in the art that Figure 6 The structure shown is for illustration only. The device for implementing any one of the methods in the above embodiments may be a terminal device, which may be a smart phone (such as an Android phone, an IOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, or other terminal devices. Figure 6 It does not limit the structure of the above electronic device. For example, the terminal device may also include Figure 6 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 6Different configurations shown.

[0079] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which can include: a flash drive, ROM, RAM, a magnetic disk or an optical disk, etc.

[0080] As an exemplary embodiment, the present application also provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute any one of the method steps of the present embodiment when running.

[0081] Optionally, in this embodiment, the above-mentioned storage medium can be used to execute the program code of the method steps of the embodiment of the present application.

[0082] Optionally, in this embodiment, the storage medium may be located on at least one network device among a plurality of network devices in the network shown in the above embodiment.

[0083] Optionally, in this embodiment, the storage medium is configured to store data for executing the method in the above embodiment.

[0084] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, which will not be described in detail in this embodiment.

[0085] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a ROM, a RAM, a mobile hard disk, a magnetic disk, or an optical disk.

[0086] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0087] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which may be personal computers, servers or network devices, etc.) to execute all or part of the steps of the method in the above embodiments.

[0088] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0089] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0090] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in this embodiment.

[0091] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0092] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

[0093] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0094] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A power calculation model training method, characterized in that: The power calculation model construction method comprises: Obtain historical power data, historical irradiance data and historical meteorological data of the target photovoltaic power station; Calculating the historical difference degree between the historical power data and the historical irradiance data, wherein the historical difference degree is: calculating the historical theoretical power generation corresponding to the target photovoltaic power station under the historical irradiance data, and dividing the historical theoretical power generation by the same historical irradiance data to calculate the historical difference degree of the photovoltaic power station; Determining a plurality of historical difference time intervals of the target photovoltaic power station based on the historical difference degree; Extracting historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals; The preset power calculation model is trained based on the training data set. During the training process, the correspondence between the historical irradiance data, the historical meteorological data and the historical power data under each historical difference degree is learned to obtain a photovoltaic power calculation model corresponding to different historical difference time intervals.

2. The power calculation model training method according to claim 1, characterized in that: The determining of a plurality of historical difference time intervals of the photovoltaic power station based on the historical difference degree includes: Performing clustering processing on the degree of difference to obtain a clustering result; Based on the clustering result, a plurality of historical difference degree groups with different similarities and their corresponding timestamps are obtained; The historical difference time interval is determined based on the timestamp.

3. The power calculation model training method according to claim 2, characterized in that: The clustering process is performed on the difference degrees to obtain a clustering result, including: The historical difference degree is intercepted according to a preset time scale to obtain a historical difference degree interval containing a timestamp; Clustering is performed on the historical difference degree intervals to obtain a clustering result.

4. The power calculation model training method according to claim 1, characterized in that: The determining of a plurality of historical difference time intervals of the photovoltaic power station based on the historical difference degree further includes: Sorting the historical differences to obtain a sorting result; Dividing the historical difference degrees based on the sorting results to obtain historical difference degrees divided into different sizes and their corresponding time intervals; The time interval is used as the historical difference time interval.

5. The power calculation model training method according to claim 1, characterized in that: The power calculation model training method also includes: Obtaining the irradiance change trend of the historical irradiance data according to a preset time dimension; Determine the historical peak time and the historical trough time of the historical irradiance based on the irradiance variation trend; The historical difference time interval is determined based on the historical peak time, the historical trough time, and the historical difference degree.

6. The power calculation model training method according to claim 1, characterized in that: The training of the preset power calculation model based on the training data set includes: In the training data set, extracting the historical irradiance data and the historical meteorological data; Calculate horizontal plane total radiation data based on the historical irradiance data; Extracting historical temperature data and historical humidity data from the historical meteorological data; Calculating historical temperature-humidity ratio data based on the historical temperature data and the historical humidity data; The horizontal plane total radiation data, the historical temperature-humidity ratio data and the historical meteorological data are input into each preset power calculation model for model training.

7. A photovoltaic power calculation method, characterized in that: The photovoltaic power calculation method comprises: Obtain the irradiance data and meteorological data of the target PV power station and their timestamps; A corresponding photovoltaic power calculation model is selected based on the historical difference time interval corresponding to the timestamp; wherein the photovoltaic power calculation model is trained by the historical meteorological data, historical irradiance data and historical power data corresponding to the historical difference time interval; the photovoltaic power calculation model is obtained based on the power calculation model training method according to any one of claims 1 to 6; The irradiance data and the meteorological data are input into the photovoltaic power calculation model for calculation to obtain a photovoltaic power calculation result.

8. A power calculation model training device, characterized in that: The device comprises: An acquisition module is used to acquire historical power data, historical irradiance data and historical meteorological data of a target photovoltaic power station; A difference degree determination module is used to calculate the historical difference degree of the target photovoltaic power station based on the historical power data and the historical irradiance data; the historical difference degree is: calculating the historical theoretical power generation corresponding to the target photovoltaic power station under the historical irradiance data, dividing the historical theoretical power generation by the same historical irradiance data to calculate the historical difference degree of the photovoltaic power station; A time interval determination module, configured to determine a plurality of historical difference time intervals of the target photovoltaic power station based on the historical difference degree; A training data set determination module, used to extract historical power data, historical irradiance data and historical meteorological data based on each of the historical difference time intervals to obtain a training data set under each of the historical difference time intervals; A model training module is used to train a preset power calculation model based on the training data set. During the training process, the correspondence between the historical irradiance data, the historical meteorological data and the historical power data under each historical difference degree is learned to obtain a photovoltaic power calculation model corresponding to different historical difference time intervals.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the power calculation model training method described in any one of claims 1 to 6 or the photovoltaic power calculation method described in claim 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the power calculation model training method described in any one of claims 1 to 6 or the photovoltaic power calculation method described in claim 7.

Images