Solar power output prediction method, device, and terminal

Abstract

The application is suitable for the technical field of new energy output prediction, and provides a solar power output prediction method, device, terminal and storage medium, the method comprising: obtaining solar power output historical data, screening information in the solar power output historical data to obtain input variables, inputting the input variables into a preset first prediction model to predict the solar power output, obtaining a first prediction result, based on the obtained first prediction result, constructing a corrected second prediction model to predict the solar power output, and combining the prediction results of the first prediction model and the second prediction model to obtain a target prediction result. The embodiment of the application obtains input variables with greater correlation with the prediction result, reduces sample redundancy, improves prediction efficiency, and the double-channel prediction makes the calculated solar power output prediction result more reliable.

Description

Technical Field

[0001] This invention belongs to the field of new energy power output prediction technology, and particularly relates to a solar power output prediction method, device and terminal. Background Technology

[0002] To address the shortage of traditional fossil fuels and the environmental pollution they cause, developing green energy has become a key focus of energy research. However, solar power output is significantly affected by weather and other factors, exhibiting strong intermittency and volatility. These characteristics pose a significant challenge to the power system after integrating a high proportion of renewable energy. Accurate forecasting of renewable energy power generation output can not only improve the operational efficiency of renewable energy power plants but also assist dispatching departments in adjusting operating modes, ensuring the safe, stable, and economical operation of the power system after the integration of a high proportion of renewable energy.

[0003] Traditional methods for predicting solar power output include modeling and irradiance prediction. However, modeling requires high accuracy of model parameters and is difficult to model, while irradiance prediction requires a large number of samples and a long training time. Therefore, improving the modeling efficiency and reliability of prediction results for solar power output has become one of the demands of photovoltaic power generation. Summary of the Invention

[0004] This invention provides a method, apparatus, and terminal for predicting solar power output, thereby improving the efficiency and accuracy of solar power output prediction.

[0005] In a first aspect, embodiments of the present invention provide a method for predicting solar power output, comprising:

[0006] Acquire historical solar power output data, filter the information in the historical solar power output data, and use the filtering results as input variables;

[0007] The input variables are input into a preset first prediction model to predict solar power output and obtain a first prediction result.

[0008] Based on the first prediction result, the input variables and the historical solar power output data are fitted and corrected to obtain a second prediction model. The input variables are then input into the target preset model to predict solar power output and obtain the second prediction result.

[0009] Calculate the deviation between the first prediction result and the second prediction result, and calculate the target prediction result of solar power output based on the deviation value.

[0010] As another embodiment of this application, the step of acquiring historical solar power output data and filtering the information in the historical solar power output data, using the filtering results as input variables, includes:

[0011] Based on the information in the historical solar power output data, a meteorological feature sequence and an output variable feature sequence are constructed.

[0012] The correlation coefficient between any two adjacent meteorological features in the meteorological feature sequence is calculated sequentially to obtain the first correlation coefficient sequence;

[0013] Set an upper limit for the correlation coefficient, and compare each of the first correlation coefficients in the first correlation coefficient sequence with the upper limit for the correlation coefficient. Determine the meteorological features corresponding to all correlation coefficients greater than the upper limit for the correlation coefficient as the dimensionality-reduced meteorological feature variables. If all the first correlation coefficients in the first correlation coefficient sequence are greater than the upper limit for the correlation coefficient, then determine the first meteorological feature in the meteorological feature sequence as the dimensionality-reduced meteorological feature variable.

[0014] Calculate the correlation coefficient between one output variable feature in the output variable feature sequence and the dimensionality-reduced meteorological feature variable in turn to obtain the second correlation coefficient sequence;

[0015] Set a lower limit for the correlation coefficient, and check one by one whether the second correlation coefficient in the second correlation coefficient sequence is lower than the lower limit for the correlation coefficient;

[0016] If the current second correlation coefficient is lower than the lower limit of the correlation coefficient, then the meteorological feature corresponding to the current second correlation coefficient is deleted from the meteorological feature sequence;

[0017] If the current second correlation coefficient is not lower than the lower limit of the correlation coefficient, then the meteorological characteristic variable corresponding to the current second correlation coefficient is determined as the input variable;

[0018] By detecting all the second correlation coefficients in the second correlation coefficient sequence using the method described above, all input variables are obtained.

[0019] As another embodiment of this application, the step of deleting the meteorological feature corresponding to the current second correlation coefficient from the meteorological feature sequence if the current second correlation coefficient is lower than the lower limit of the correlation coefficient further includes:

[0020] If all the second correlation coefficients in the second correlation coefficient sequence are lower than the lower limit of the correlation coefficient, then jump to the step "Construct meteorological feature sequence and output variable feature sequence based on the information in the historical solar power output data" to reconstruct the meteorological feature sequence and execute subsequent steps until the input variable is determined.

[0021] As another embodiment of this application, the step of inputting the input variables into a preset first prediction model to predict solar power output and obtaining a first prediction result includes:

[0022] The input variables are input into the first prediction model to predict the solar power output, thereby obtaining the predicted solar power output value and the input variables of the first prediction model at the next moment.

[0023] As another embodiment of this application, after obtaining the first prediction result, the method further includes:

[0024] The predicted solar power output is combined with the historical solar power output data to obtain the combined historical solar power output data.

[0025] The input variable at the next moment is merged with the input variable to obtain the updated input variable;

[0026] Based on the merged historical solar power output data and the updated input variables, the parameters of the first prediction model are updated to obtain a new first prediction model for subsequent solar power output prediction.

[0027] As another embodiment of this application, the step of fitting and correcting the input variables and the historical solar power output data based on the first prediction result to obtain a second prediction model, and then performing solar power output prediction to obtain a second prediction result, includes:

[0028] Based on the merged historical solar power output data, a solar power output data matrix is ​​constructed;

[0029] Based on the updated input variables, construct the input variable matrix;

[0030] Based on the solar power output data matrix and the input variable matrix, the function relationship matrix is ​​calculated;

[0031] Based on the aforementioned functional relationship matrix, a second prediction model with multiple inputs and a single output is obtained;

[0032] The second prediction model is used to predict solar power output, and the second prediction result is obtained.

[0033] As another embodiment of this application, a function relationship matrix is ​​calculated based on the solar power output data matrix and the input variable matrix, including:

[0034] according to The function relationship matrix is ​​calculated.

[0035] in, Let X represent the obtained function relationship matrix, and let X represent the input variable matrix. T L represents the transpose of the input variable matrix, and L represents the solar power output data matrix.

[0036] Based on the aforementioned functional relationship matrix, a second prediction model with multiple inputs and a single output is obtained, including:

[0037] according to The second prediction model is obtained;

[0038] in, The values ​​represent the prediction results corresponding to the second prediction model, x1, x2, ..., x. n These represent the input variables for the solar power output prediction method.

[0039] As another embodiment of this application, the step of calculating the deviation value between the first prediction result and the second prediction result, and calculating the target prediction result of solar power output based on the deviation value, includes:

[0040] Calculate the deviation between the first prediction result and the second prediction result;

[0041] Set a deviation threshold and detect whether the deviation value exceeds the deviation threshold;

[0042] When the deviation value exceeds the deviation threshold, the first prediction result is output as the target prediction result for solar power output.

[0043] When the deviation value does not exceed the deviation threshold, the second prediction result is output as the target prediction result for solar power output.

[0044] Secondly, embodiments of the present invention provide a solar power output prediction device, comprising:

[0045] The data processing module is used to acquire historical solar power output data, filter the information in the historical solar power output data, and use the obtained filtering results as input variables.

[0046] The first prediction module is used to input the input variables into a preset first prediction model to predict solar power output and obtain a first prediction result.

[0047] The second prediction module is used to fit and correct the input variables and the historical solar power output data based on the first prediction result to obtain a second prediction model, and input the input variables into the target preset model to predict solar power output to obtain the second prediction result.

[0048] The result output module is used to calculate the deviation between the first prediction result and the second prediction result, and to calculate the target prediction result of solar power output based on the deviation value.

[0049] Thirdly, embodiments of the present invention provide a terminal including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the solar power output prediction method as described in the first aspect or any possible implementation thereof.

[0050] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: Compared with the prior art, the embodiments of the present invention filter the information in the historical data of solar power output to obtain input variables that are highly correlated with the prediction results, thereby improving the prediction efficiency. The input variables are input into a preset first prediction model to predict solar power output, and a first prediction result is obtained. Based on the obtained first prediction result, a modified second prediction model is constructed to predict solar power output, making the calculated solar power output prediction results more reliable. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart illustrating the implementation of the solar power output prediction method provided in this embodiment of the invention.

[0053] Figure 2 This is a flowchart of obtaining input variables provided in an embodiment of the present invention;

[0054] Figure 3 This is a flowchart of obtaining the second prediction result provided in an embodiment of the present invention;

[0055] Figure 4 This is a flowchart of obtaining target detection results provided by an embodiment of the present invention;

[0056] Figure 5 This is a flowchart of the solar power output prediction device provided in the embodiments of the present invention;

[0057] Figure 6 This is a schematic diagram of the terminal provided in an embodiment of the present invention. Detailed Implementation

[0058] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0059] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0060] Figure 1 The following is a detailed flowchart illustrating the implementation process of a solar power output prediction method provided in an embodiment of the present invention.

[0061] Step 101: Obtain historical solar power output data, filter the information in the historical solar power output data, and use the filtered results as input variables.

[0062] In this embodiment, the historical solar power output data may include meteorological information such as date and time, daily irradiance, accumulated days, temperature, wind speed, wind direction, atmospheric pressure, rainfall, and average daily irradiance, as well as the corresponding solar power output information.

[0063] Optional, such as Figure 2 The process of filtering information from historical solar power output data, and using the filtered results as input variables, can include the following steps.

[0064] Step 201: Construct meteorological characteristic sequences and output variable characteristic sequences based on information from historical solar power output data.

[0065] Optionally, in this embodiment, a meteorological feature sequence is constructed based on meteorological information such as date and time, daily irradiance, accumulated days, temperature, wind speed, wind direction, atmospheric pressure, rainfall, and average daily irradiance from historical solar power output data, and an output variable feature sequence is constructed based on the meteorological information and solar power output information.

[0066] Step 202: Calculate the correlation coefficient between any two adjacent meteorological features in the meteorological feature sequence in turn to obtain the first correlation coefficient sequence.

[0067] Optionally, in this step, according to Calculate the meteorological characteristic sequence {x1, x2, ... x} sequentially. n The correlation coefficient between any two adjacent meteorological features in};

[0068] Where, f(x) i ,x i+1) represents the correlation coefficient between two adjacent meteorological features, m represents the number of historical data corresponding to the meteorological feature, and x represents the correlation coefficient between two adjacent meteorological features. i With x i+1 Indicates two adjacent meteorological features, Indicates meteorological characteristics x i The corresponding historical average value Indicates meteorological characteristics x i+1 The corresponding historical average, x i k Indicates meteorological characteristics x i The corresponding k-th historical data.

[0069] Step 203: Set an upper limit for the correlation coefficient, and compare each of the first correlation coefficients in the first correlation coefficient sequence with the upper limit for the correlation coefficient. Determine the meteorological features corresponding to all correlation coefficients greater than the upper limit for the correlation coefficient as the dimensionality-reduced meteorological feature variables. If all the first correlation coefficients in the first correlation coefficient sequence are greater than the upper limit for the correlation coefficient, then determine the first meteorological feature in the meteorological feature sequence as the dimensionality-reduced meteorological feature variable.

[0070] Optionally, in this step, the upper limit of the correlation coefficient is set to 90%, and the first correlation coefficient in the first correlation coefficient sequence is compared with the upper limit of the correlation coefficient one by one. The meteorological features corresponding to all correlation coefficients greater than 90% are determined as the meteorological feature variables after dimensionality reduction.

[0071] If all the first correlation coefficients in the first correlation coefficient sequence are greater than 90%, then the first meteorological feature in the meteorological feature sequence is determined as the dimensionality-reduced meteorological feature variable, replacing the other features in the meteorological feature sequence, in order to reduce sample redundancy.

[0072] Step 204: Calculate the correlation coefficient between one output variable feature in the output variable feature sequence and the reduced meteorological feature variable in turn to obtain the second correlation coefficient sequence.

[0073] Optionally, in this step, you can... Calculate the correlation coefficient between a feature of an output variable in the output variable feature sequence and the dimensionality-reduced meteorological feature variable;

[0074] Where f(x,y) represents the correlation coefficient between the dimensionality-reduced meteorological features and a feature of an output variable, y k This represents the k-th historical data point corresponding to the output variable feature.

[0075] Step 205: Set the lower limit of the correlation coefficient and check whether the second correlation coefficient in the second correlation coefficient sequence is lower than the lower limit of the correlation coefficient.

[0076] Optionally, in this step, the lower limit of the correlation coefficient is set to 10%, and it is detected whether there is a second correlation coefficient in the second correlation coefficient sequence that is lower than the lower limit of the correlation coefficient.

[0077] If the current second correlation coefficient is lower than the lower limit of the correlation coefficient, proceed to step 206; if the current second correlation coefficient is not lower than the lower limit of the correlation coefficient, proceed to step 207.

[0078] Step 206: If the current second correlation coefficient is lower than the lower limit of the correlation coefficient, then delete the meteorological feature corresponding to the current second correlation coefficient in the meteorological feature sequence.

[0079] Optionally, in this step, meteorological feature variables corresponding to correlation coefficients below 10% in the second correlation coefficient sequence are not used as input variables, and the meteorological features corresponding to the second correlation coefficient are deleted from the original meteorological feature sequence.

[0080] Optionally, if all second correlation coefficients in the second correlation coefficient sequence are below the lower limit of the correlation coefficient, then proceed to step 201, reconstruct the meteorological feature sequence, and execute subsequent steps until the input variable is determined.

[0081] Step 207: If the current second correlation coefficient is not lower than the lower limit of the correlation coefficient, then the meteorological characteristic variable corresponding to the current second correlation coefficient is determined as the input variable.

[0082] Optionally, in this step, meteorological characteristic variables corresponding to at least 10% of the second correlation coefficients in the second correlation coefficient sequence are determined as input variables.

[0083] By detecting all the second correlation coefficients in the second correlation coefficient sequence according to steps 205 to 207, all input variables are obtained.

[0084] Step 102: Input the input variables into the preset first prediction model to predict solar power output and obtain the first prediction result.

[0085] Optionally, in this step, based on historical solar power output data, the time series data of the input variables are input into the multilayer perceptron model. The multilayer perceptron is used to obtain the spatiotemporal relationship between the variables, and the solar power output is predicted to obtain the predicted value of solar power output within the prediction period and the input variables of the multilayer perceptron model at the next moment.

[0086] Optionally, after obtaining the first prediction result in step 102, the method further includes merging the predicted solar power output value within the obtained prediction time period with the historical solar power output data to obtain merged historical solar power output data, thereby supplementing the original historical solar power output data.

[0087] The input variables of the obtained multilayer perceptron model at the next moment are merged with the previously determined input variables to obtain the updated input variables.

[0088] Based on the merged historical solar power output data and the updated input variables, the parameters of the multilayer perceptron model are updated to obtain a new multilayer perceptron model for subsequent solar power output prediction.

[0089] Step 103: Based on the first prediction result, the input variables and historical solar power output data are fitted and corrected to obtain the second prediction model. The input variables are then input into the target preset model to predict solar power output and obtain the second prediction result.

[0090] Optional, such as Figure 3 The method shown involves fitting and correcting the input variables and historical solar power output data based on the first prediction result to obtain a second prediction model. The input variables are then input into the target preset model to predict solar power output and obtain the second prediction result. This process may include the following steps.

[0091] Step 301: Construct a solar power output data matrix based on the merged historical solar power output data.

[0092] Optionally, in this step, according to Construct a solar power output data matrix;

[0093] Where L represents the solar power output data matrix, L k This represents the solar power output data at time k.

[0094] Step 302: Construct the input variable matrix based on the updated input variables.

[0095] Optionally, in this step, according to Construct the input variable matrix;

[0096] Where X represents the input variable matrix, x mn This represents the data corresponding to the nth meteorological characteristic variable at the mth time point.

[0097] Step 303: Based on the solar power output data matrix and the input variable matrix, calculate the function relationship matrix.

[0098] Optionally, in this step, according to The function relationship matrix is ​​calculated;

[0099] in, Let X represent the obtained functional relationship matrix. T This represents the transpose of the input variable matrix.

[0100] Step 304: Based on the function relationship matrix, obtain the second prediction model with multiple inputs and single output.

[0101] Optionally, in this step, according to A second prediction model with multiple inputs and a single output is obtained;

[0102] in, The values ​​represent the prediction results corresponding to the second prediction model, x1, x2, ..., x. n These represent the input variables for the solar power output prediction method.

[0103] Step 305: Use the second prediction model to predict the solar power output and obtain the second prediction result.

[0104] Optionally, in this step, the updated input variables are input into the obtained second prediction model to obtain the corresponding solar power output prediction data.

[0105] Step 104: Calculate the deviation between the first prediction result and the second prediction result, and calculate the target prediction result of solar power output based on the deviation value.

[0106] Optional, such as Figure 4 The process of calculating the deviation between the first and second prediction results and then calculating the target prediction result for solar power output based on the deviation can include the following steps.

[0107] Step 401: Calculate the deviation between the first prediction result and the second prediction result.

[0108] Optionally, in this step, the deviation between the solar power output prediction data in the first prediction result and the solar power output prediction data in the second prediction result is calculated.

[0109] Step 402: Set the deviation threshold and check whether the deviation value exceeds the deviation threshold.

[0110] Optionally, in this step, a deviation threshold is set, and it is detected whether the deviation value between the first prediction result and the second prediction result exceeds the deviation threshold. If it exceeds the deviation threshold, step 403 is executed; if it does not exceed the deviation threshold, step 404 is executed.

[0111] Step 403: When the deviation value exceeds the deviation threshold, output the first prediction result as the target prediction result for solar power output.

[0112] Optionally, in this step, when the deviation between the first prediction result and the second prediction result exceeds the set deviation threshold, the solar power output data predicted by the multilayer perceptron model is output as the target prediction result.

[0113] Step 404: When the deviation value does not exceed the deviation threshold, output the second prediction result as the target prediction result for solar power output.

[0114] Optionally, in this step, if the deviation between the first prediction result and the second prediction result does not exceed the set deviation threshold, the corrected second prediction result will be output as the target prediction result by default.

[0115] The aforementioned solar power output prediction method obtains historical solar power output data and filters the information to obtain input variables that are highly correlated with the prediction results, thereby reducing sample redundancy and improving prediction efficiency. The input variables are then input into a preset first prediction model to predict solar power output and obtain a first prediction result. Based on the obtained first prediction result, a modified second prediction model is constructed to predict solar power output, making the calculated prediction results more reliable.

[0116] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0117] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0118] Figure 5 A schematic diagram of the solar power output prediction device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below:

[0119] like Figure 5 As shown, the solar power output prediction device 500 includes: a data processing module 501, a first prediction module 502, a second prediction module 503, and a result output module 504.

[0120] The data processing module 501 is used to acquire historical solar power output data, filter the information in the historical solar power output data, and use the obtained filtering results as input variables.

[0121] The first prediction module 502 is used to input the input variables into a preset first prediction model to predict the solar power output and obtain the first prediction result.

[0122] The second prediction module 503 is used to fit and correct the input variables and historical solar power output data based on the first prediction result to obtain a second prediction model, and input the input variables into the target preset model to predict solar power output to obtain the second prediction result.

[0123] The result output module 504 is used to calculate the deviation value between the first prediction result and the second prediction result, and to calculate the target prediction result of solar power output based on the deviation value.

[0124] Optionally, the data processing module 501 acquires historical solar power output data, filters the information in the historical solar power output data, and uses the filtered results as input variables, which can be used for:

[0125] Based on information from historical solar power output data, meteorological characteristic sequences and output variable characteristic sequences are constructed.

[0126] The correlation coefficient between any two adjacent meteorological features in the meteorological feature sequence is calculated sequentially to obtain the first correlation coefficient sequence.

[0127] Set an upper limit for the correlation coefficient, and compare each of the first correlation coefficients in the first correlation coefficient sequence with the upper limit for the correlation coefficient. The meteorological features corresponding to all correlation coefficients greater than the upper limit for the correlation coefficient are determined as the meteorological feature variables after dimensionality reduction. If all the first correlation coefficients in the first correlation coefficient sequence are greater than the upper limit for the correlation coefficient, then the first meteorological feature in the meteorological feature sequence is determined as the meteorological feature variable after dimensionality reduction.

[0128] The correlation coefficient between one output variable feature in the output variable feature sequence and the dimensionality-reduced meteorological feature variable is calculated sequentially to obtain the second correlation coefficient sequence.

[0129] Set a lower limit for the correlation coefficient and check whether the second correlation coefficient in the second correlation coefficient sequence is lower than the lower limit for the correlation coefficient.

[0130] If the current second correlation coefficient is lower than the lower limit of the correlation coefficient, then delete the meteorological feature corresponding to the current second correlation coefficient in the meteorological feature sequence.

[0131] If the current second correlation coefficient is not lower than the lower limit of the correlation coefficient, then the meteorological characteristic variable corresponding to the current second correlation coefficient will be determined as the input variable;

[0132] By detecting all second correlation coefficients in the second correlation coefficient sequence in the same way as detecting the current second correlation coefficient, all input variables are obtained.

[0133] Optionally, in the data processing module 501, if the current second correlation coefficient is lower than the lower limit of the correlation coefficient, the meteorological feature corresponding to the current second correlation coefficient is deleted from the meteorological feature sequence. This can also be used for:

[0134] If all second correlation coefficients in the second correlation coefficient sequence are below the lower limit of the correlation coefficient, then proceed to step 201, reconstruct the meteorological feature sequence, and execute subsequent steps until the input variable is determined.

[0135] Optionally, in the first prediction module 502, the input variables are input into a preset first prediction model to predict solar power output, and a first prediction result is obtained, which can be used for:

[0136] The input variables are fed into the first prediction model to predict the solar power output, resulting in the predicted solar power output value and the input variables for the first prediction model at the next moment.

[0137] Optionally, after obtaining the first prediction result, the first prediction module 502 can also be used for:

[0138] The predicted solar power output is combined with the historical solar power output data to obtain the combined historical solar power output data.

[0139] The input variable for the next moment is merged with the input variable to obtain the updated input variable;

[0140] Based on the merged historical solar power output data and the updated input variables, the parameters of the first prediction model are updated to obtain a new first prediction model for subsequent solar power output prediction.

[0141] Optionally, in the second prediction module 503, based on the first prediction result, the input variables and historical solar power output data are fitted and corrected to obtain a second prediction model, and solar power output is predicted to obtain a second prediction result, which can be used for:

[0142] Construct a solar power output data matrix based on the merged historical solar power output data;

[0143] Construct the input variable matrix based on the updated input variables;

[0144] Based on the solar power output data matrix and the input variable matrix, the function relationship matrix is ​​calculated.

[0145] Based on the function relationship matrix, a second prediction model with multiple inputs and a single output is obtained;

[0146] The second prediction model is used to predict solar power output, and the second prediction result is obtained.

[0147] Optionally, the second prediction module 503 calculates a function relationship matrix based on the solar power output data matrix and the input variable matrix, which can be used for:

[0148] according to The function relationship matrix is ​​calculated;

[0149] in, Let X represent the obtained function relationship matrix, and let X represent the input variable matrix. TL represents the transpose of the input variable matrix, and L represents the solar power output data matrix.

[0150] Based on the function relationship matrix, a second prediction model with multiple inputs and a single output is obtained, including:

[0151] according to The second prediction model was obtained;

[0152] in, The values ​​represent the prediction results corresponding to the second prediction model, x1, x2, ..., x. n These represent the input variables for the solar power output prediction method.

[0153] Optionally, the result output module 504 calculates the deviation between the first prediction result and the second prediction result, and calculates the target prediction result of solar power output based on the deviation value, which can be used for:

[0154] Calculate the deviation between the first and second prediction results;

[0155] Set a deviation threshold and detect whether the deviation value exceeds the deviation threshold;

[0156] When the deviation value exceeds the deviation threshold, the first prediction result is output as the target prediction result for solar power output.

[0157] When the deviation value does not exceed the deviation threshold, the second prediction result is output as the target prediction result for solar power output.

[0158] The aforementioned solar power output prediction device acquires historical solar power output data and filters the information within it to obtain input variables that are highly correlated with the prediction results. This reduces sample redundancy and improves prediction efficiency. The input variables are then fed into a preset first prediction model to predict solar power output, obtaining a first prediction result. Based on the obtained first prediction result, a revised second prediction model is constructed to predict solar power output. This embodiment reduces sample redundancy and improves prediction efficiency by filtering input variables to eliminate features with low correlation to the output variables. Furthermore, by constructing a dual-channel model to predict solar power output, the reliability of the calculated prediction results is higher.

[0159] Figure 6 This is a schematic diagram of a terminal provided in an embodiment of the present invention. For example... Figure 6 As shown, the terminal 600 in this embodiment includes a processor 601, a memory 602, and a computer program 603 stored in the memory 602 and executable on the processor 601, such as a solar power output prediction program. When the processor 601 executes the computer program 603, it implements the steps described in the above-described solar power output prediction method embodiment, for example... Figure 1Steps 101 to 104 shown, or Figure 2 , Figure 3 as well as Figure 4 The steps shown indicate that when processor 601 executes computer program 603, it implements the functions of each module in the above-described device embodiments, for example... Figure 5 The functions of modules 501 to 504 are shown.

[0160] For example, computer program 603 can be divided into one or more modules / units, one or more modules are stored in memory 602 and executed by processor 601 to complete the present invention. One or more modules can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 603 in terminal 600. For example, computer program 603 can be divided into... Figure 5 Modules 501 to 504 are shown.

[0161] Terminal 600 may include, but is not limited to, processor 601 and memory 602. Those skilled in the art will understand that... Figure 6 This is merely an example of terminal 600 and does not constitute a limitation on terminal 600. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal may also include input / output devices, network access devices, buses, etc.

[0162] The processor 601 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0163] The memory 602 can be an internal storage unit of the terminal 600, such as the hard disk or RAM of the terminal 600. The memory 602 can also be an external storage device of the terminal 600, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal 600. Furthermore, the memory 602 can include both internal storage units and external storage devices of the terminal 600. The memory 602 is used to store computer programs and other programs and data required by the terminal. The memory 602 can also be used to temporarily store data that has been output or will be output.

[0164] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0165] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0166] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0167] In the embodiments provided by this invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0168] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0169] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0170] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, it can implement the steps of the various embodiments of the solar power output prediction method described above.

[0171] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0172] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for predicting solar power output, characterized in that, include: Acquire historical solar power output data, filter the information in the historical solar power output data, and use the filtering results as input variables; The input variables are input into a preset first prediction model to predict solar power output and obtain a first prediction result. Based on the first prediction result, the input variables and the historical solar power output data are fitted and corrected to obtain a second prediction model. The input variables are then input into the second prediction model to predict solar power output and obtain a second prediction result. Calculate the deviation between the first prediction result and the second prediction result, and calculate the target prediction result of solar power output based on the deviation value; The process of acquiring historical solar power output data and filtering the information within that data, using the filtering results as input variables, includes: Based on the information in the historical solar power output data, a meteorological feature sequence and an output variable feature sequence are constructed. The correlation coefficient between any two adjacent meteorological features in the meteorological feature sequence is calculated sequentially to obtain the first correlation coefficient sequence; Set an upper limit for the correlation coefficient, and compare each of the first correlation coefficients in the first correlation coefficient sequence with the upper limit for the correlation coefficient. Determine the meteorological features corresponding to all correlation coefficients greater than the upper limit for the correlation coefficient as the dimensionality-reduced meteorological feature variables. If all the first correlation coefficients in the first correlation coefficient sequence are greater than the upper limit for the correlation coefficient, then determine the first meteorological feature in the meteorological feature sequence as the dimensionality-reduced meteorological feature variable. Calculate the correlation coefficient between one output variable feature in the output variable feature sequence and the dimensionality-reduced meteorological feature variable in turn to obtain the second correlation coefficient sequence; Set a lower limit for the correlation coefficient, and check one by one whether the second correlation coefficient in the second correlation coefficient sequence is lower than the lower limit for the correlation coefficient; If the current second correlation coefficient is lower than the lower limit of the correlation coefficient, then the meteorological feature corresponding to the current second correlation coefficient is deleted from the meteorological feature sequence; If the current second correlation coefficient is not lower than the lower limit of the correlation coefficient, then the meteorological characteristic variable corresponding to the current second correlation coefficient is determined as the input variable; By detecting all the second correlation coefficients in the second correlation coefficient sequence using the method described above, all input variables are obtained.

2. The solar power output prediction method according to claim 1, characterized in that, The step of deleting the meteorological feature corresponding to the current second correlation coefficient from the meteorological feature sequence if the current second correlation coefficient is lower than the lower limit of the correlation coefficient further includes: If all the second correlation coefficients in the second correlation coefficient sequence are lower than the lower limit of the correlation coefficient, then proceed to step "Construct meteorological feature sequence and output variable feature sequence based on the information in the historical solar power output data" to reconstruct the meteorological feature sequence and execute subsequent steps until the input variable is determined.

3. The solar power output prediction method according to claim 2, characterized in that, The input variables are input into a preset first prediction model to predict solar power output, and a first prediction result is obtained, including: The input variables are input into the first prediction model to predict the solar power output, thereby obtaining the predicted solar power output value and the input variables of the first prediction model at the next moment.

4. The solar power output prediction method according to claim 3, characterized in that, After obtaining the first prediction result, the process also includes: The predicted solar power output is combined with the historical solar power output data to obtain the combined historical solar power output data. The input variable at the next moment is merged with the input variable to obtain the updated input variable; Based on the merged historical solar power output data and the updated input variables, the parameters of the first prediction model are updated to obtain a new first prediction model for subsequent solar power output prediction.

5. The solar power output prediction method according to claim 4, characterized in that, Based on the first prediction result, the input variables and the historical solar power output data are fitted and corrected to obtain a second prediction model, and solar power output is predicted to obtain a second prediction result, including: Based on the merged historical solar power output data, a solar power output data matrix is ​​constructed; Based on the updated input variables, construct the input variable matrix; Based on the solar power output data matrix and the input variable matrix, the function relationship matrix is ​​calculated; Based on the aforementioned functional relationship matrix, a second prediction model with multiple inputs and a single output is obtained; The second prediction model is used to predict solar power output, and the second prediction result is obtained.

6. The solar power output prediction method according to claim 5, characterized in that, Based on the solar power output data matrix and input variable matrix, a function relationship matrix is ​​calculated, including: according to The function relationship matrix is ​​calculated. in, This represents the obtained functional relationship matrix. Represents the input variable matrix. This represents the transpose of the input variable matrix. Represents a matrix of solar power output data; Based on the aforementioned functional relationship matrix, a second prediction model with multiple inputs and a single output is obtained, including: according to The second prediction model is obtained; in, This represents the prediction result corresponding to the second prediction model. These represent the input variables for the solar power output prediction method.

7. The solar power output prediction method according to claim 6, characterized in that, The step of calculating the deviation between the first prediction result and the second prediction result, and calculating the target prediction result of solar power output based on the deviation value, includes: Calculate the deviation between the first prediction result and the second prediction result; Set a deviation threshold and detect whether the deviation value exceeds the deviation threshold; When the deviation value exceeds the deviation threshold, the first prediction result is output as the target prediction result for solar power output. When the deviation value does not exceed the deviation threshold, the second prediction result is output as the target prediction result for solar power output.

8. A solar power output detection device, characterized in that, include: The data processing module is used to acquire historical solar power output data, filter the information in the historical solar power output data, and use the obtained filtering results as input variables. The process of acquiring historical solar power output data and filtering the information within that data, using the filtered results as input variables, includes: constructing a meteorological feature sequence and an output variable feature sequence based on the information in the historical solar power output data; sequentially calculating the correlation coefficient between any two adjacent meteorological features in the meteorological feature sequence to obtain a first correlation coefficient sequence; setting an upper limit for the correlation coefficient, and comparing each first correlation coefficient in the first correlation coefficient sequence with the upper limit, identifying all meteorological features corresponding to correlation coefficients greater than the upper limit as dimensionality-reduced meteorological feature variables; if all first correlation coefficients in the first correlation coefficient sequence are greater than the upper limit, then determining the first meteorological feature in the meteorological feature sequence. The meteorological feature variables are dimensionality reduced. The correlation coefficient between an output variable feature in the output variable feature sequence and the dimensionality-reduced meteorological feature variables is calculated sequentially to obtain a second correlation coefficient sequence. A lower limit for the correlation coefficient is set, and each second correlation coefficient in the second correlation coefficient sequence is checked to see if it is lower than the lower limit. If the current second correlation coefficient is lower than the lower limit, the meteorological feature corresponding to the current second correlation coefficient is deleted from the meteorological feature sequence. If the current second correlation coefficient is not lower than the lower limit, the meteorological feature variable corresponding to the current second correlation coefficient is determined as an input variable. All second correlation coefficients in the second correlation coefficient sequence are checked using the above method to obtain all input variables. The first prediction module is used to input the input variables into a preset first prediction model to predict solar power output and obtain a first prediction result. The second prediction module is used to fit and correct the input variables and the historical solar power output data based on the first prediction result to obtain a second prediction model, and input the input variables into the second prediction model to predict solar power output to obtain a second prediction result. The result output module is used to calculate the deviation between the first prediction result and the second prediction result, and to calculate the target prediction result of solar power output based on the deviation value.

9. A terminal, comprising a memory and a processor, the memory for storing a computer program, the processor for calling and running the computer program stored in the memory, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

Images