Solar irradiance interval prediction method and device, computer device, and storage medium

Abstract

The application relates to the technical field of interval prediction, and discloses a solar irradiance interval prediction method and device, computer equipment and a storage medium, historical environment parameter data sets and a first environment parameter data set in a to-be-predicted period are acquired, and a first environment sample data set is determined; a target solar irradiance prediction model is established; the first environment parameter data set is input into the target solar irradiance prediction model, and a solar irradiance prediction result in the to-be-predicted period is obtained; based on the first environment sample data set, the solar irradiance prediction result in the to-be-predicted period is subjected to interval prediction by using an autonomous sampling method, and a target solar irradiance interval prediction result in the to-be-predicted period is obtained. The application considers the similarity and regularity of historical data and the complexity and accuracy of a model. Furthermore, the autonomous sampling method is used to perform interval prediction on the solar irradiance prediction result, and the reliability and confidence of the prediction result are effectively increased.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of interval prediction, and particularly relates to a solar irradiance interval prediction method and device, computer equipment and a storage medium. BACKGROUND

[0002] Accurate prediction of solar irradiance is the basis for evaluating power plant generation and ensuring power supply planning and more effective balancing of the power grid. At present, the methods for short-term solar irradiance prediction mainly include physical model method, statistical learning method, artificial neural network method, deep learning method and clustering analysis method. The physical model method has high computational complexity, but can provide high-precision prediction results; the statistical method does not require complex physical models and has high computational efficiency, but the prediction accuracy is low and is greatly affected by data quality and feature selection; the artificial neural network method can adapt to complex nonlinear relationships, but requires a large amount of data and computing resources, and is sensitive to network structure and parameter selection. The deep learning method requires a large amount of data and computing resources, but can provide high-precision prediction results.

[0003] However, the current short-term solar irradiance prediction methods are all based on single-point prediction, for example, the above methods usually give a certain prediction value, rather than a confidence interval of the prediction value, which cannot comprehensively reflect the uncertainty of the prediction result, resulting in a large prediction error. SUMMARY

[0004] Therefore, the present application provides a solar irradiance interval prediction method, device, computer equipment and storage medium to solve the problem that the current short-term solar irradiance prediction methods are all based on single-point prediction, which cannot comprehensively reflect the uncertainty of the prediction result, resulting in a large prediction error.

[0005] In a first aspect, the present application provides a solar irradiance interval prediction method for a photovoltaic power station, which comprises:

[0006] obtaining a historical environmental parameter data set and a first environmental parameter data set of the photovoltaic power station in a to-be-predicted period; determining a first environmental sample data set based on the historical environmental parameter data set and the first environmental parameter data set; establishing a target solar irradiance prediction model based on the first environmental sample data set; inputting the first environmental parameter data set into the target solar irradiance prediction model to obtain a solar irradiance prediction result in the to-be-predicted period; and performing interval prediction on the solar irradiance prediction result in the to-be-predicted period based on the first environmental sample data set by using an autonomous sampling method to obtain a target solar irradiance interval prediction result in the to-be-predicted period.

[0007] The solar irradiance interval prediction method provided by the application selects a first environmental parameter data set in a to-be-predicted period based on a historical environmental parameter data set, and predicts by using a target solar irradiance prediction model, which considers both the similarity and regularity of historical data and the complexity and accuracy of the model. Further, the solar irradiance prediction result is subjected to interval prediction by using an autonomous sampling method, thereby effectively increasing the reliability and confidence of the prediction result.

[0008] In an optional embodiment, the first environmental sample data set is determined based on the historical environmental parameter data set and the first environmental parameter data set, and the method comprises:

[0009] The historical environmental parameter data set is subjected to regression analysis, and a multiple linear regression function reflecting the relationship between the environmental parameter and the collection time is constructed; the second environmental parameter data set is determined based on the multiple linear regression function; and the first environmental sample data set is determined based on the historical environmental parameter data set and the second environmental parameter data set.

[0010] The application determines the second environmental parameter data set by constructing a multiple linear regression function reflecting the relationship between the environmental parameter and the collection time, considers the similarity and regularity of historical data, and provides data support for subsequent reduction of prediction error.

[0011] In an optional embodiment, the first environmental sample data set is determined based on the historical environmental parameter data set and the second environmental parameter data set, and the method comprises:

[0012] The Mahalanobis distance between the historical environmental parameter data corresponding to each preset collection time in the historical environmental parameter data set and the second environmental parameter data corresponding to the preset collection time in the second environmental parameter data set is calculated; the target period is determined based on each Mahalanobis distance; and the first environmental sample data set is determined in the historical environmental parameter data set based on the target period.

[0013] The application determines the target period and the first environmental sample data set in combination with the Mahalanobis distance, considers the correlation between various data features, and can effectively solve the problem of different measurement scales between different data features.

[0014] In an optional embodiment, the target solar irradiance prediction model is established based on the first environmental sample data set, and the method comprises:

[0015] The first environmental sample data subset and the second environmental sample data subset are determined based on the first environmental sample data set; the first environmental sample data subset is input into a preset neural network for training to obtain an initial solar irradiance prediction model; and the initial solar irradiance prediction model is verified by using the second environmental sample data subset to obtain the target solar irradiance prediction model.

[0016] The application establishes a target solar irradiance prediction model by using a first environmental sample data subset, and verifies the model by combining a second environmental sample data subset, considers the complexity and accuracy of the model, and improves the accuracy of the model prediction.

[0017] In an optional embodiment, based on the first environmental sample data set, the solar irradiance prediction result in the to-be-predicted period is interval predicted by using the autonomous sampling method to obtain the target solar irradiance interval prediction result in the to-be-predicted period, comprising:

[0018] At least one second environmental sample data set is determined in the first environmental sample data set by using the autonomous sampling method; at least one target solar irradiance prediction model is established based on each second environmental sample data set; the first environmental parameter data set is input into each target solar irradiance prediction model to obtain at least one solar irradiance prediction result; at least one confidence interval range and at least one initial solar irradiance interval prediction result in the to-be-predicted period are determined based on each solar irradiance prediction result and each preset interval prediction confidence value; the accuracy of each initial solar irradiance interval prediction result is determined based on each confidence interval range; and the target solar irradiance interval prediction result in the to-be-predicted period is determined based on each accuracy and each initial solar irradiance interval prediction result.

[0019] The autonomous sampling method is used to interval predict the solar irradiance prediction result, without making any assumption on the original first environmental sample data set, the distribution and parameters can be directly estimated from the first environmental sample data set, the uncertainty of the estimated quantity can be quantified by calculating the confidence interval range, and the reliability and confidence of the prediction result are effectively increased.

[0020] In an optional embodiment, at least one confidence interval range and at least one initial solar irradiance interval prediction result in the to-be-predicted period are determined based on each solar irradiance prediction result and each preset interval prediction confidence value, comprising:

[0021] The sample number of each second environmental sample data set is obtained; the solar irradiance average value and the solar irradiance standard deviation are determined based on each solar irradiance prediction result; at least one interval value is determined based on the solar irradiance standard deviation, each preset interval prediction confidence value and each sample number; the confidence interval range corresponding to each preset interval prediction confidence value is determined based on each interval value and the solar irradiance average value; and at least one initial solar irradiance interval prediction result in the to-be-predicted period is determined based on each confidence interval range.

[0022] The uncertainty of the estimated quantity is quantified by calculating the confidence interval range, and the reliability and confidence of the prediction result are effectively increased.

[0023] In an alternative embodiment, the accuracy of each initial solar irradiance interval prediction result is determined based on each confidence interval range, comprising:

[0024] The accuracy level of each solar irradiance interval prediction result corresponding to each confidence interval range is obtained by using a big data method; an evaluation index matrix is established based on each accuracy level; a solar irradiance matrix is established based on each solar irradiance prediction result; a membership matrix is established based on the evaluation index matrix and the solar irradiance matrix; and the accuracy of each initial solar irradiance interval prediction result is determined based on the membership matrix and each solar irradiance prediction result.

[0025] The present application quantifies the uncertainty of the estimator by calculating the confidence interval range, effectively increasing the reliability and confidence of the prediction result.

[0026] In a second aspect, the present application provides a solar irradiance interval prediction device for a photovoltaic power station, comprising:

[0027] The acquisition module is configured to acquire a historical environmental parameter data set and a first environmental parameter data set of the photovoltaic power station in a to-be-predicted period; the determination module is configured to determine a first environmental sample data set based on the historical environmental parameter data set and the first environmental parameter data set; the establishment module is configured to establish a target solar irradiance prediction model based on the first environmental sample data set; the input module is configured to input the first environmental parameter data set into the target solar irradiance prediction model to obtain a solar irradiance prediction result in the to-be-predicted period; and the prediction module is configured to perform interval prediction on the solar irradiance prediction result in the to-be-predicted period by using a self-sampling method based on the first environmental sample data set to obtain a target solar irradiance interval prediction result in the to-be-predicted period.

[0028] In a third aspect, the present application provides a computer device, comprising a memory and a processor, which are communicatively connected to each other, and the memory stores computer instructions, and the processor executes the computer instructions to perform the solar irradiance interval prediction method of the first aspect or any of the corresponding embodiments thereof.

[0029] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer instructions for causing a computer to execute the solar irradiance interval prediction method of the first aspect or any of the corresponding embodiments thereof. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the specific embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings described below are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0031] Figure 1 is a flowchart of a solar irradiance interval prediction method according to an embodiment of the present application;

[0032] Figure 2 is a flowchart of another solar irradiance interval prediction method according to an embodiment of the present application;

[0033] Figure 3 is a flowchart of still another solar irradiance interval prediction method according to an embodiment of the present application;

[0034] Figure 4 is a flowchart of a solar irradiance interval prediction method for a photovoltaic power station according to an embodiment of the present application;

[0035] Figure 5 is a structural block diagram of a solar irradiance interval prediction device according to an embodiment of the present application;

[0036] Figure 6 is a hardware structure schematic diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0038] Accurate prediction of solar irradiance is the basis for evaluating power plant generation and ensuring power supply planning and more effective balancing of the power grid.

[0039] Therefore, the embodiments of the present application provide a solar irradiance interval prediction method for a photovoltaic power station. By selecting a target period retrieval condition and using a target solar irradiance prediction model for prediction, the reliability and confidence of the prediction results are increased.

[0040] According to the embodiment of the present application, a solar irradiance interval prediction method is provided. It should be noted that the steps shown in the flowchart 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 here.

[0041] In the present embodiment, a solar irradiance interval prediction method is provided, which can be used in a photovoltaic power station, Figure 1 The flowchart of the solar irradiance interval prediction method according to the embodiment of the present application is shown in FIG. 1, which includes the following steps: Figure 1

[0042] In step S101, a historical environmental parameter data set and a first environmental parameter data set of the photovoltaic power station in a to-be-predicted period are obtained.

[0043] The historical environmental parameter data set represents a set of environmental parameters including solar irradiance, air temperature, humidity, etc. collected by the photovoltaic power station in a historical collection period.

[0044] The first environmental parameter data set represents a set of environmental parameters including solar irradiance, air temperature, humidity, etc. collected by the photovoltaic power station in the to-be-predicted period.

[0045] Specifically, the historical environmental parameter data set includes a plurality of data sub-libraries storing various types of environmental parameters, and each data sub-library corresponds to a category label.

[0046] According to the real-time environmental parameters collected by the photovoltaic power station, a real-time environmental parameter data set is constructed, and the categories of the environmental parameters in the real-time environmental parameter data set are determined, such as air temperature, humidity, wind speed and irradiance.

[0047] Further, the real-time environmental parameter data set is matched with the category labels of each data sub-library, and the environmental parameters in the real-time environmental parameter data set whose categories fail to match the category labels of each data sub-library are marked as abnormal data and the abnormal data in the real-time environmental parameter data set is deleted.

[0048] Finally, the real-time environmental parameter data set after deleting the abnormal data is taken as the first environmental parameter data set in the to-be-predicted period.

[0049] In step S102, based on the historical environmental parameter data set and the first environmental parameter data set, a first environmental sample data set is determined.

[0050] The first environmental sample data set represents a set of environmental parameter data in a collection period similar to the historical collection period corresponding to the historical environmental parameter data set in the to-be-predicted period.

[0051] ​Step S103, based on the first environment sample data set, a target solar irradiance prediction model is established.

[0052] Specifically, the model training with the first environment sample data set can obtain the target solar irradiance prediction model meeting the condition.

[0053] Step S104, the first environment parameter data set is input into the target solar irradiance prediction model, and a solar irradiance prediction result in the to-be-predicted period is obtained.

[0054] Specifically, the first environment parameter data set collected by the photovoltaic power station in the to-be-predicted period is input into the constructed target solar irradiance prediction model, and a corresponding solar irradiance prediction result in the to-be-predicted period can be output.

[0055] Step S105, based on the first environment sample data set, the solar irradiance prediction result in the to-be-predicted period is interval predicted by using the autonomous sampling method, and a target solar irradiance interval prediction result in the to-be-predicted period is obtained.

[0056] The autonomous sampling method represents a continuous or discontinuous sample collection mode without human intervention in a sampling process through a device according to a pre-prepared program.

[0057] In this embodiment, the solar irradiance prediction result in the to-be-predicted period is interval predicted by using the autonomous sampling method, which can directly estimate the distribution and parameters from the first environment sample data set, and effectively increases the reliability and confidence of the prediction result.

[0058] The solar irradiance interval prediction method provided in this embodiment is based on the first environment parameter data set selected from the historical environment parameter data set in the to-be-predicted period, and the target solar irradiance prediction model is used for prediction, which not only considers the similarity and regularity of historical data, but also considers the complexity and accuracy of the model. Further, the autonomous sampling method is used for interval prediction of the solar irradiance prediction result, which effectively increases the reliability and confidence of the prediction result.

[0059] In this embodiment, a solar irradiance interval prediction method is provided, which can be used for a photovoltaic power station, Figure 2 is a flowchart of the solar irradiance interval prediction method according to the embodiment of the present application, as Figure 2 shown, the flowchart includes the following steps:

[0060] Step S201, a historical environment parameter data set and a first environment parameter data set of a photovoltaic power station in a to-be-predicted period are obtained. For details, please refer to the step S101 of the embodiment shown in Figure 1 herein, which will not be repeated.

[0061] Step S202, determining the first environment sample dataset based on the historical environment parameter dataset and the first environment parameter dataset.

[0062] Specifically, the step S202 includes:

[0063] Step S2021, performing regression analysis based on the historical environment parameter dataset, and constructing a multiple linear regression function.

[0064] The multiple linear regression function reflects the relationship between the environment parameter and the collection time.

[0065] Specifically, the collection time of each type of environment parameter in the historical environment parameter dataset in a plurality of historical collection periods is taken as the independent variable, and each type of environment parameter at different collection times in the historical environment parameter dataset in the plurality of historical collection periods is taken as the dependent variable to perform regression analysis, and a multiple linear regression function reflecting the relationship between the environment parameter and the collection time is constructed according to the regression analysis result.

[0066] Step S2022, determining the second environment parameter dataset based on the multiple linear regression function.

[0067] Specifically, a change trend curve of each type of environment parameter changing with time in the to-be-predicted period is established according to the multiple linear regression function, the change trend curve of each type of environment parameter takes the collection time as the horizontal coordinate and each type of environment parameter as the vertical coordinate, and each type of environment parameter at different collection times in the change trend curve of each type of environment parameter is taken as the second environment parameter dataset.

[0068] Step S2023, determining the first environment sample dataset based on the historical environment parameter dataset and the second environment parameter dataset.

[0069] Specifically, the second environment parameter dataset is taken as a condition for determining the first environment sample dataset, and further, the first environment sample dataset can be obtained in the historical environment parameter dataset according to the condition.

[0070] In some optional embodiments, the step S2023 includes:

[0071] Step a1, calculating the Mahalanobis distance between the historical environment parameter data corresponding to each preset collection time in the historical environment parameter dataset and the second environment parameter data corresponding to the preset collection time in the second environment parameter dataset.

[0072] Step a2, determining a target period based on each Mahalanobis distance.

[0073] Step a3, determining the first environment sample dataset in the historical environment parameter dataset based on the target period.

[0074] Specifically, for each category of environmental parameter value in the historical environmental parameter data set, the Mahalanobis distance between the corresponding environmental parameter value in the second environmental parameter data set at the same collection time is obtained.

[0075] Then, the Mahalanobis distance values of all categories of environmental parameters in each historical collection period in the historical environmental parameter data set are accumulated to obtain a total Mahalanobis distance, and the historical collection period with the smallest total Mahalanobis distance is selected as a similar period, that is, a target period.

[0076] Finally, the environmental parameter data corresponding to the target period in the historical environmental parameter data set is selected as the first environmental sample data set.

[0077] Step S203, based on the first environmental sample data set, a target solar irradiance prediction model is established.

[0078] Specifically, the above step S203 includes:

[0079] Step S2031, based on the first environmental sample data set, a first environmental sample data subset and a second environmental sample data subset are determined.

[0080] Specifically, the first environmental sample data set is divided into a first environmental sample data subset and a second environmental sample data subset. The first environmental sample data subset is used as a model training set, and the second environmental sample data subset is used as a model test set.

[0081] Step S2032, the first environmental sample data subset is input into a preset neural network for training to obtain an initial solar irradiance prediction model.

[0082] Specifically, the first environmental sample data subset is input into the preset neural network for training until the preset model loss function training is stable, and the initial solar irradiance prediction model is output, and the model parameters of the initial solar irradiance prediction model are saved.

[0083] Step S2033, the second environmental sample data subset is used to verify the initial solar irradiance prediction model to obtain a target solar irradiance prediction model.

[0084] Specifically, the output data matrix of the initial solar irradiance prediction model after iterative training in step S2032 is verified using the second environmental sample data subset, and if the verification is passed, the target solar irradiance prediction model is output.

[0085] If the verification fails, the model parameters are adjusted and the first environmental sample data subset is retrained until the output data matrix of the trained model can pass the similarity verification, and the target solar irradiance prediction model is output.

[0086] Step S204: Input the first environmental parameter dataset into the target solar irradiance prediction model to obtain the solar irradiance prediction results for the period to be predicted. For details, please refer to [link to details]. Figure 1 Step S104 of the illustrated embodiment will not be described again here.

[0087] Step S205: Based on the first environmental sample dataset, the solar irradiance prediction results for the prediction period are used to perform interval prediction using the autonomous sampling method, thus obtaining the target solar irradiance interval prediction results for the prediction period. For details, please refer to... Figure 1 Step S105 of the illustrated embodiment will not be described again here.

[0088] The solar irradiance interval prediction method provided in this embodiment determines the second environmental parameter dataset by constructing a multiple linear regression function that reflects the relationship between environmental parameters and collection time. It takes into account the similarity and regularity of historical data. Furthermore, it combines Mahalanobis distance to determine the target period and the first environmental sample dataset, taking into account the correlation between various data features. This method can effectively handle the problem of different measurement scales between different data features.

[0089] This embodiment provides a method for predicting solar irradiance intervals, which can be used in photovoltaic power plants. Figure 3 This is a flowchart of a solar irradiance interval prediction method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps:

[0090] Step S301: Obtain the historical environmental parameter dataset and the first environmental parameter dataset of the photovoltaic power plant within the predicted period. For details, please refer to [link to relevant documentation]. Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0091] Step S302: Based on the historical environmental parameter dataset and the first environmental parameter dataset, determine the first environmental sample dataset. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0092] Step S303: Based on the first environmental sample dataset, establish a target solar irradiance prediction model. For details, please refer to [link to relevant documentation]. Figure 3 Step S203 of the illustrated embodiment will not be described again here.

[0093] Step S304: Input the first environmental parameter dataset into the target solar irradiance prediction model to obtain the solar irradiance prediction results for the period to be predicted. For details, please refer to [link to details]. Figure 1 Step S104 of the illustrated embodiment will not be described again here.

[0094] At step S305, based on the first environment sample data set, an interval prediction of the solar irradiance prediction result in the to-be-predicted period is performed by using the bootstrap method, to obtain a target solar irradiance interval prediction result in the to-be-predicted period.

[0095] Specifically, the step S305 includes:

[0096] At step S3051, at least one second environment sample data set is determined in the first environment sample data set by using the bootstrap method.

[0097] Specifically, each type of environment parameter collected at different time in a similar period is taken as a sample data set, a sample data set extraction size m and an extraction times k are determined, a sample with a size of m is randomly extracted from the sample data set with replacement, and the extraction is repeated k times, each time the sample is marked as a bootstrap sample, and finally k different bootstrap samples are obtained; each bootstrap sample is taken as a second environment sample data set.

[0098] At step S3052, at least one target solar irradiance prediction model is established based on each second environment sample data set.

[0099] Specifically, by training each second environment sample data set, a target solar irradiance prediction model corresponding to each second environment sample data set can be established. The specific training process is described in detail in the step S203, which will not be repeated here.

[0100] At step S3053, the first environment parameter data set is input into each target solar irradiance prediction model, to obtain at least one solar irradiance prediction result.

[0101] Specifically, the first environment parameter data set is input into each target solar irradiance prediction model, to obtain the solar irradiance prediction result output by each target solar irradiance prediction model.

[0102] At step S3054, based on each solar irradiance prediction result and each preset interval prediction confidence value, at least one confidence interval range and at least one initial solar irradiance interval prediction result in the to-be-predicted period are determined.

[0103] The interval prediction means that an interval estimation is given to the prediction of future or unknown data under a certain confidence.

[0104] Specifically, by setting the preset interval prediction confidence value to different values, different confidence interval ranges can be obtained.

[0105] Further, different initial solar irradiance interval prediction results can be obtained according to different confidence interval ranges.

[0106] Step S3055, determining the accuracy of each initial solar irradiance interval prediction result based on each confidence interval range.

[0107] Specifically, according to the description of step S3054, the initial solar irradiance interval prediction result represents an interval estimation result given by the prediction of the solar irradiance prediction result under a certain confidence.

[0108] Therefore, the accuracy of each initial solar irradiance interval prediction result can be determined according to different confidence interval ranges.

[0109] Step S3056, determining the target solar irradiance interval prediction result in the to-be-predicted period based on each accuracy and each initial solar irradiance interval prediction result.

[0110] Specifically, the initial solar irradiance interval prediction result with the highest accuracy is taken as the target solar irradiance interval prediction result in the to-be-predicted period.

[0111] In some optional embodiments, the above step S3054 comprises:

[0112] Step b1, obtaining the sample number of each second environment sample data set.

[0113] Step b2, determining the solar irradiance average value and the solar irradiance standard deviation based on each solar irradiance prediction result.

[0114] Step b3, determining at least one interval value based on the solar irradiance standard deviation, each preset interval prediction confidence value and each sample number.

[0115] Step b4, determining the confidence interval range corresponding to each preset interval prediction confidence value based on each interval value and the solar irradiance average value.

[0116] Step b5, determining at least one initial solar irradiance interval prediction result in the to-be-predicted period based on each confidence interval range.

[0117] First, according to the description of step S3051, the sample number of each second environment sample data set is k, and further, the k solar irradiance prediction results are sorted in descending order, and the solar irradiance average value v and the solar irradiance standard deviation s are calculated.

[0118] Secondly, the interval value is calculated by using the following relationship (1):

[0119]

[0120] wherein L represents an interval value; NORMSINV represents a standard normal distribution inverse function; represents a preset interval prediction confidence value; represents a confidence level.

[0121] Then, a confidence interval range corresponding to each preset interval prediction confidence value is determined as (v-L, v+L).

[0122] Finally, the uncertainty of the estimation quantity is quantified by the calculated confidence interval range, i.e., an initial solar irradiance interval prediction result in the to-be-predicted period can be determined by each confidence interval range.

[0123] In some optional embodiments, the above step S3055 comprises:

[0124] Step c1, the accuracy level of each solar irradiance interval prediction result corresponding to each confidence interval range is obtained by using a big data method.

[0125] Step c2, an evaluation index matrix is established based on each accuracy level.

[0126] Step c3, a solar irradiance matrix is established based on each solar irradiance prediction result.

[0127] Step c4, a membership matrix is established based on the evaluation index matrix and the solar irradiance matrix.

[0128] Step c5, the accuracy of each initial solar irradiance interval prediction result is determined based on the membership matrix and each solar irradiance prediction result.

[0129] First, the accuracy level of the solar irradiance interval prediction result corresponding to different confidence interval ranges is obtained by using a big data method, wherein the accuracy level can include 95%, 85%, 80%, 75%.

[0130] Second, an evaluation index matrix about prediction interval accuracy is established according to the accuracy level corresponding to different confidence interval ranges, and a solar irradiance matrix is established according to each solar irradiance prediction result predicted by the solar irradiance prediction model.

[0131] Then, the evaluation index matrix and the solar irradiance matrix are fused by using the following relation (2) to obtain a membership matrix representing the fuzzy relationship between the solar irradiance prediction result predicted by the solar irradiance prediction model and the prediction interval accuracy:

[0132] M = αM1 + βM2 (2)

[0133] In the formula, M represents a membership matrix; M1 represents an evaluation index matrix; M2 represents a solar irradiance matrix; and a and β represent weighted parameters for controlling the balance between the evaluation index matrix and the solar irradiance matrix in the membership matrix; "+" represents the addition of elements at corresponding positions of the evaluation index matrix and the solar irradiance matrix.

[0134] Finally, a corresponding confidence interval range can be determined based on each solar irradiance prediction result, and further, the accuracy of each initial solar irradiance interval prediction result can be determined in the membership matrix based on the confidence interval range.

[0135] The solar irradiance interval prediction method provided in the embodiment uses the autonomous sampling method to perform interval prediction on the solar irradiance prediction result, does not need to make any assumption on the original first environmental sample data set, can directly estimate the distribution and parameters from the first environmental sample data set, and can quantify the uncertainty of the estimated quantity by calculating the confidence interval range, thereby effectively increasing the reliability and confidence of the prediction result.

[0136] In an example, a solar irradiance interval prediction method for a photovoltaic power station is provided, as shown in the formula: Figure 4 The method comprises the following steps:

[0137] Step S1: acquiring real-time environmental parameters collected by the photovoltaic power station and marking the collection time, setting a collection period, and performing data cleaning on the real-time environmental parameters according to environmental parameter information in a historical database;

[0138] Step S2: setting a similar period retrieval condition according to a to-be-predicted period, extracting sample data satisfying the similar period retrieval condition from the historical database, taking the environmental parameters of each category of the selected similar period as input features and the solar irradiance as an output label, and constructing a solar irradiance prediction model;

[0139] Step S3: performing interval prediction on the prediction result by using the autonomous sampling method, and determining the prediction result accuracy according to the interval range in which the prediction result is located.

[0140] Further, the process of performing data cleaning on the real-time environmental parameters according to the environmental parameter information in the historical database comprises:

[0141] constructing a real-time environmental parameter data set according to the real-time environmental parameters collected by the photovoltaic power station, determining each environmental parameter category in the real-time environmental parameter data set; the historical database contains a plurality of data sub-libraries storing environmental parameters of various categories, and category labels of each data sub-library are generated; the real-time environmental parameter data set and the category labels of each data sub-library are matched, environmental parameters in the real-time environmental parameter data set for which the matching of the environmental parameter categories and the category labels of each data sub-library fails are marked as abnormal data, and the abnormal data in the real-time environmental parameter data set are deleted.

[0142] Further, the process of setting the similar period retrieval condition according to the to-be-predicted period includes:

[0143] Taking the collection time of each type of environmental parameter in the historical database in several historical collection periods as the independent variable, and taking each type of environmental parameter at different collection times in the historical database in several historical collection periods as the dependent variable, regression analysis is performed to construct a multiple linear regression function representing the mutual relationship between the environmental parameter and the collection time. According to the multiple linear regression function, a change trend curve of each type of environmental parameter in the to-be-predicted period is established over time. The change trend curve of each type of environmental parameter takes the collection time as the horizontal coordinate and each type of environmental parameter as the vertical coordinate. Each type of environmental parameter at different collection times in the change trend curve of each type of environmental parameter is taken as the similar period retrieval condition.

[0144] Further, the process of extracting sample data satisfying the similar period retrieval condition from the historical database includes:

[0145] The values of each type of environmental parameter at different collection times in the similar period retrieval condition are determined, and the values of each type of environmental parameter at different collection times in several historical collection periods in the historical database are obtained. For each type of environmental parameter value in the historical collection period, the Mahalanobis distance between it and the corresponding environmental parameter value at the same collection time in the similar period retrieval condition is calculated. The Mahalanobis distance values of all types of environmental parameters in each historical collection period are accumulated to obtain the total Mahalanobis distance. The historical collection period with the smallest total Mahalanobis distance is selected as the similar period.

[0146] Further, the process of constructing a solar irradiance prediction model with each type of environmental parameter in the selected similar period as the input feature and the solar irradiance as the output label includes:

[0147] The solar irradiance prediction model is constructed based on the RBN neural network. A historical data set is constructed according to the values of each type of environmental parameter at different collection times in the similar period, and the historical data set is divided into a training set and a test set. The solar irradiance prediction model is trained in real time through the training set until the loss function of the solar irradiance prediction model is stable, and the model parameters are saved. Then, the output data matrix of the iteratively trained solar irradiance prediction model is verified for similarity through the test set. In the process of establishing the solar irradiance prediction model, the environmental parameters include temperature, humidity, wind speed, and irradiance, etc. 1000 groups of historical collection period data information about temperature, humidity, wind speed, and irradiance are obtained. 950 groups of data are taken as the training set, and 50 groups of data are taken as the test set. The solar irradiance prediction model is trained until it is qualified.

[0148] The various types of environmental parameters at different collection times in the to-be-predicted period are input into the solar irradiance prediction model verified by the test set, and the solar irradiance at different collection times in the to-be-predicted period is obtained according to the output layer of the solar irradiance prediction model.

[0149] Further, the process of interval prediction of the prediction result by using the self-sampling method comprises:

[0150] The various types of environmental parameters at different collection times in similar periods are used as sample data sets, the sample data set extraction size m and the extraction times k are determined, a sample with a size of m is randomly extracted with replacement from the sample data set, the extraction is repeated k times, the sample of each extraction is marked as a Bootstrap sample, and finally k different Bootstrap samples are obtained; each Bootstrap sample is used as a training set, and k solar irradiance prediction models are established according to the solar irradiance prediction model construction method; the various types of environmental parameters at different collection times in the to-be-predicted period are input into the k solar irradiance prediction models to obtain k solar irradiance prediction results; the k solar irradiance prediction results are sorted in descending order of solar irradiance, and the solar irradiance mean value v and the solar irradiance standard deviation s of the k solar irradiance prediction results are calculated to determine the interval prediction confidence According to the confidence The interval value L is determined according to the solar irradiance standard deviation s and the number of k samples, and the confidence is determined according to the interval value L and the solar irradiance mean value v to determine the interval range of the confidence; by setting different values of the confidence, the interval range of different confidences can be obtained.

[0151] The interval value L and the confidence The specific process of the interval range is described in the above step b5.

[0152] Further, the process of determining the prediction result accuracy according to the interval range of the prediction result comprises:

[0153] The accuracy level of the solar irradiance prediction result corresponding to the interval range of different confidences is obtained by using the big data method, and the accuracy level includes 95%, 85%, 80%, and 75%;

[0154] According to the precision level corresponding to the interval range of different confidence levels, an evaluation index matrix about the prediction interval precision is established, and a solar irradiance matrix is established according to the solar irradiance predicted by the solar irradiance prediction model; a membership matrix representing the fuzzy relationship between the solar irradiance predicted by the solar irradiance prediction model and the prediction result precision is established according to the evaluation index matrix and the solar irradiance matrix, and the specific process refers to the above formula (2); and the prediction solar irradiance precision of the to-be-predicted period is obtained according to the membership matrix and the solar irradiance predicted by the solar irradiance prediction model.

[0155] In the embodiment, a solar irradiance interval prediction device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be described again. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware is also possible and is contemplated.

[0156] The embodiment provides a solar irradiance interval prediction device, which is used for a photovoltaic power station; as shown in the figure, the device comprises: Figure 5

[0157] The obtaining module 501 is configured to obtain a historical environmental parameter data set and a first environmental parameter data set of the photovoltaic power station in a to-be-predicted period.

[0158] The determining module 502 is configured to determine a first environmental sample data set based on the historical environmental parameter data set and the first environmental parameter data set.

[0159] The establishing module 503 is configured to establish a target solar irradiance prediction model based on the first environmental sample data set.

[0160] The input module 504 is configured to input the first environmental parameter data set into the target solar irradiance prediction model to obtain a solar irradiance prediction result in the to-be-predicted period.

[0161] The prediction module 505 is configured to perform interval prediction on the solar irradiance prediction result in the to-be-predicted period by using the self-sampling method based on the first environmental sample data set to obtain a target solar irradiance interval prediction result in the to-be-predicted period.

[0162] In some optional embodiments, the determining module 502 comprises:

[0163] The analysis and construction unit is configured to perform regression analysis based on the historical environmental parameter data set, and construct a multiple linear regression function, the multiple linear regression function reflecting the relationship between the environmental parameters and the collection time.

[0164] ​The first determining unit is configured to determine the second environmental parameter dataset based on a multivariate linear regression function.

[0165] The second determining unit is configured to determine the first environmental sample dataset based on the historical environmental parameter dataset and the second environmental parameter dataset.

[0166] In some optional embodiments, the second determining unit comprises:

[0167] The calculating sub-unit is configured to calculate Mahalanobis distances between the historical environmental parameter data corresponding to each preset collection time in the historical environmental parameter dataset and the second environmental parameter data corresponding to the preset collection time in the second environmental parameter dataset.

[0168] The first determining sub-unit is configured to determine the target period based on each Mahalanobis distance.

[0169] The second determining sub-unit is configured to determine the first environmental sample dataset in the historical environmental parameter dataset based on the target period.

[0170] In some optional embodiments, the establishing module 503 comprises:

[0171] The third determining unit is configured to determine the first environmental sample data subset and the second environmental sample data subset based on the first environmental sample dataset.

[0172] The first input unit is configured to input the first environmental sample data subset into a preset neural network for training, to obtain an initial solar irradiance prediction model.

[0173] The verifying unit is configured to verify the initial solar irradiance prediction model by using the second environmental sample data subset, to obtain a target solar irradiance prediction model.

[0174] In some optional embodiments, the prediction module 505 comprises:

[0175] The fourth determining unit is configured to determine at least one second environmental sample dataset in the first environmental sample dataset by using an autonomous sampling method.

[0176] The establishing unit is configured to establish at least one target solar irradiance prediction model based on each second environmental sample dataset.

[0177] The second input unit is configured to input the first environmental parameter dataset into each target solar irradiance prediction model respectively, to obtain at least one solar irradiance prediction result.

[0178] The fifth determining unit is configured to determine at least one confidence interval range and at least one initial solar irradiance interval prediction result in a to-be-predicted period, based on each solar irradiance prediction result and each preset interval prediction confidence value.

[0179] The sixth determining unit is configured to determine the accuracy of each initial solar irradiance interval prediction result based on each confidence interval range.

[0180] The seventh determining unit is configured to determine a target solar irradiance interval prediction result in the to-be-predicted period based on each accuracy and each initial solar irradiance interval prediction result.

[0181] In some optional embodiments, the fifth determining unit comprises:

[0182] The first obtaining subunit is configured to obtain the sample quantity of each second environmental sample data set.

[0183] The third determining subunit is configured to determine the solar irradiance mean value and the solar irradiance standard deviation based on each solar irradiance prediction result.

[0184] The fourth determining subunit is configured to determine at least one interval value based on the solar irradiance standard deviation, each preset interval prediction confidence value and each sample quantity.

[0185] The fifth determining subunit is configured to determine the confidence interval range corresponding to each preset interval prediction confidence value based on each interval value and the solar irradiance mean value.

[0186] The sixth determining unit is configured to determine at least one initial solar irradiance interval prediction result in the to-be-predicted period based on each confidence interval range.

[0187] In some optional embodiments, the sixth determining unit comprises:

[0188] The second obtaining subunit is configured to obtain the accuracy level of each solar irradiance interval prediction result corresponding to each confidence interval range by using a big data method.

[0189] The first establishing subunit is configured to establish an evaluation index matrix based on each accuracy level.

[0190] The second establishing subunit is configured to establish a solar irradiance matrix based on each solar irradiance prediction result.

[0191] The third establishing subunit is configured to establish a membership matrix based on the evaluation index matrix and the solar irradiance matrix.

[0192] The seventh determining subunit is configured to determine the accuracy of each initial solar irradiance interval prediction result based on the membership matrix and each solar irradiance prediction result.

[0193] Further function descriptions of the above-mentioned modules and units are the same as those of the above-mentioned corresponding embodiments, which will not be repeated here.

[0194] The solar irradiance interval prediction device in the embodiment is presented in the form of functional units, where the units refer to ASIC (Application Specific Integrated Circuit) circuits, processors and memories that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0195] The embodiment of the present application also provides a computer device having the above Figure 5 solar irradiance interval prediction device.

[0196] Please refer to Figure 6 , Figure 6 is a structural schematic diagram of a computer device provided by an optional embodiment of the present application, as Figure 6 shown, the computer device includes one or more processors 10, memories 20, and interfaces for connecting components, including high-speed interfaces and low-speed interfaces. Various components are communicatively connected through different buses, and can be installed on a common mainboard or in other manners as needed. The processor can process instructions executed in the computer device, including instructions stored in the memory or the memory to display graphical information on a GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, multiple processors and / or buses can be used with multiple memories and multiple memories, if necessary. Similarly, multiple computer devices can be connected, each providing part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 The processor 10 is taken as an example in the embodiment.

[0197] The processor 10 can be a central processor, a network processor, or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a general array logic, or any combination thereof.

[0198] The memory 20 stores instructions executable by the at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0199] The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system, application programs required for at least one function, and the like. The data storage area can store data created according to the use of the computer device, and the like. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory such as at least one of a magnetic disk storage device, a flash memory device, or other non-transitory solid state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely from the processor 10, which can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0200] The memory 20 can include a volatile memory such as a random access memory, and can further include a non-volatile memory such as a flash memory, a hard disk, or a solid state disk, and a combination thereof.

[0201] The computer device further includes a communication interface 30 for communication of the computer device with other devices or communication networks.

[0202] The embodiments of the present application also provide a computer readable storage medium. The above-described method according to the embodiments of the present application can be implemented in hardware, firmware, or as computer code recorded on a storage medium, or be stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded to a local storage medium through a network, and thus the method described herein can be processed by such software using a general purpose computer, a special purpose processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, and the like. Further, the storage medium can include a combination of the above-mentioned storage media. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, which, when accessed and executed by the computer, the processor, or the hardware, implements the method shown in the above-described embodiments.

[0203] Although the embodiments of the present application have been described with reference to the accompanying drawings, various modifications and changes can be suggested to one skilled in the art, and it is intended that the present application encompass such modifications and changes as fall within the scope of the appended claims.

Claims

1. A solar irradiance interval prediction method for a photovoltaic power plant; characterized in that, The method comprises: acquiring a historical environmental parameter dataset and a first environmental parameter dataset of the photovoltaic power station in a to-be-predicted period; determining a first environmental sample dataset based on the historical environmental parameter dataset; establishing a target solar irradiance prediction model based on the first environmental sample dataset; determining at least one second environmental sample dataset in the first environmental sample dataset by using an autonomous sampling method; establishing at least one target solar irradiance prediction model based on each second environmental sample dataset; inputting the first environmental parameter dataset into each target solar irradiance prediction model respectively to obtain at least one solar irradiance prediction result; determining at least one confidence interval range and at least one initial solar irradiance interval prediction result in the to-be-predicted period based on each solar irradiance prediction result and each preset interval prediction confidence value; determining the accuracy of each initial solar irradiance interval prediction result based on each confidence interval range; determining a target solar irradiance interval prediction result in the to-be-predicted period based on each accuracy and each initial solar irradiance interval prediction result.

2. The method of claim 1, wherein, Determining a first environmental sample dataset based on the historical environmental parameter dataset comprises: performing regression analysis based on the historical environmental parameter dataset and constructing a multiple linear regression function, wherein the multiple linear regression function reflects the relationship between environmental parameters and collection time; determining a second environmental parameter dataset based on the multiple linear regression function; determining the first environmental sample dataset based on the historical environmental parameter dataset and the second environmental parameter dataset.

3. The method of claim 2, wherein, Determining the first environmental sample dataset based on the historical environmental parameter dataset and the second environmental parameter dataset comprises: calculating the Mahalanobis distance between the historical environmental parameter data corresponding to each preset collection time in the historical environmental parameter dataset and the second environmental parameter data corresponding to the preset collection time in the second environmental parameter dataset; determining a target period based on each Mahalanobis distance; determining the first environmental sample dataset in the historical environmental parameter dataset based on the target period.

4. The method of claim 1, wherein, Establishing a target solar irradiance prediction model based on the first environmental sample dataset comprises: determining a first environmental sample data subset and a second environmental sample data subset based on the first environmental sample dataset; inputting the first environmental sample data subset into a preset neural network for training to obtain an initial solar irradiance prediction model; verifying the initial solar irradiance prediction model by using the second environmental sample data subset to obtain the target solar irradiance prediction model.

5. The method of claim 1, wherein, Determining at least one confidence interval range and at least one initial solar irradiance interval prediction result in the to-be-predicted period based on each solar irradiance prediction result and each preset interval prediction confidence value comprises: acquiring the sample number of each second environmental sample dataset; determining a solar irradiance average value and a solar irradiance standard deviation based on each solar irradiance prediction result; determining at least one interval value based on the standard deviation of the solar irradiance, the confidence value of each of the preset intervals, and the number of each of the samples; determining the confidence interval range corresponding to each of the preset interval prediction confidence values based on each of the interval values and the average value of the solar irradiance; determining at least one initial solar irradiance interval prediction result in the to-be-predicted period based on each of the confidence interval ranges.

6. The method of claim 1, wherein, determining the accuracy of each of the initial solar irradiance interval prediction results based on each of the confidence interval ranges, including: obtaining the accuracy level of each of the solar irradiance interval prediction results corresponding to each of the confidence interval ranges by using a big data method; establishing an evaluation index matrix based on each of the accuracy levels; establishing a solar irradiance matrix based on each of the solar irradiance prediction results; establishing a membership matrix based on the evaluation index matrix and the solar irradiance matrix; determining the accuracy of each of the initial solar irradiance interval prediction results based on the membership matrix and each of the solar irradiance prediction results.

7. A solar irradiance interval prediction device for a photovoltaic power plant; characterized in that, The device comprises: an acquisition module configured to acquire a historical environmental parameter dataset and a first environmental parameter dataset of the photovoltaic power station in a to-be-predicted period; a determination module configured to determine a first environmental sample dataset based on the historical environmental parameter dataset; an establishment module configured to establish a target solar irradiance prediction model based on the first environmental sample dataset; a fourth determination unit configured to determine at least one second environmental sample dataset in the first environmental sample dataset by using a self-sampling method; an establishment unit configured to establish at least one target solar irradiance prediction model based on each of the second environmental sample datasets; a second input unit configured to input the first environmental parameter dataset into each of the target solar irradiance prediction models to obtain at least one solar irradiance prediction result; a fifth determination unit configured to determine at least one confidence interval range and at least one initial solar irradiance interval prediction result in the to-be-predicted period based on each of the solar irradiance prediction results and each preset interval prediction confidence value; a sixth determination unit configured to determine the accuracy of each of the initial solar irradiance interval prediction results based on each of the confidence interval ranges; a seventh determination unit configured to determine a target solar irradiance interval prediction result in the to-be-predicted period based on each of the accuracy and each of the initial solar irradiance interval prediction results.

8. A computer device, comprising: comprise: a memory and a processor, which are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the solar irradiance interval prediction method in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions, and the computer instructions are used to make a computer execute the solar irradiance interval prediction method in any one of claims 1 to 6.

Images