Copula structure-based net load prediction method and system

Abstract

The invention provides a net load prediction method and system based on a Copula structure. According to the implementation scheme, a dictionary learning method is adopted, new energy output and load of a first regional power grid at first time are learned, predicted new energy output and predicted load of the first regional power grid are obtained, and an initial predicted net load is determined; a Gaussian Copula structure is adopted, modeling is carried out on a prediction error for predicting new energy output and a prediction error for predicting a load, and a joint probability density function is obtained; performing prediction error sampling on the joint probability density function to obtain a new energy output error sample and a load error sample of the first regional power grid at the first time; determining a net load error based on the new energy output error sample and the load error sample; and based on the net load error, correcting the predicted net load to obtain a target predicted net load of the first regional power grid at the first time. According to the invention, the net load accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a net load prediction method and system based on a Copula structure. Background Art

[0002] With the large-scale access of renewable energy, load forecasting of traditional power systems faces many challenges. The volatility and uncertainty of renewable energy generation greatly reduce the prediction accuracy of power grid load. In order to ensure the safe and stable operation of the power system, it is necessary to accurately predict the net load, that is, to eliminate the impact of renewable energy generation such as wind power and solar power, and predict the actual net load that the power grid needs to dispatch without the support of these energy sources.

[0003] However, most traditional load forecasting methods rely on historical load data and simple time series models, which cannot fully consider the impact of renewable energy. Therefore, how to improve the accuracy, reliability and real-time performance of net load forecasting has become an important issue in the current power system scheduling and optimization. Summary of the invention

[0004] The present invention provides a net load prediction method and system based on a Copula structure, which can solve at least one of the above technical problems.

[0005] According to one aspect of the present invention, a net load prediction method based on a Copula structure is provided, comprising: Using a dictionary learning method, learning the new energy output and load of the first regional power grid at the first time, to obtain the predicted new energy output and predicted load of the first regional power grid at the first time; Determining a predicted net load of the first regional power grid within the first time based on the predicted new energy output and the predicted load; A Gaussian Copula structure is used to model the prediction error of the predicted new energy output and the prediction error of the predicted load to obtain a joint probability density function; Performing prediction error sampling on the joint probability density function to obtain a new energy output error sample and a load error sample of the first regional power grid at the first time; Determining a net load error based on the new energy output error sample and the load error sample; Based on the net load error, the predicted net load is corrected to obtain a target predicted net load of the first regional power grid at the first time.

[0006] According to another aspect of the present invention, there is provided a net load prediction device based on a Copula structure, comprising: An output and load forecasting module, configured to learn the new energy output and load of the first regional power grid at the first time by using a dictionary learning method, and obtain the predicted new energy output and predicted load of the first regional power grid at the first time; A net load determination module, configured to determine a predicted net load of the first regional power grid at the first time based on the predicted new energy output and the predicted load; A prediction error modeling module is used to model the prediction error of the predicted new energy output and the prediction error of the predicted load by using a Gaussian Copula structure to obtain a joint probability density function; A prediction error sampling module, used to perform prediction error sampling on the joint probability density function to obtain a new energy output error sample and a load error sample of the first regional power grid at the first time; A net load error determination module, configured to determine a net load error based on the new energy output error sample and the load error sample; A net load correction module is used to correct the predicted net load based on the net load error to obtain a target predicted net load of the first regional power grid at the first time.

[0007] According to another aspect of the present invention, a net load prediction system based on a Copula structure is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the processor is used to obtain the instructions from the memory and execute the instructions, so that the at least one processor can execute any net load prediction method based on a Copula structure described in any embodiment of the present invention.

[0008] By adopting the technical solution of the present invention, a dictionary learning method is used to learn the new energy output and load of the first regional power grid at the first time, and the predicted new energy output and predicted load of the first regional power grid at the first time are obtained. In this way, based on the predicted new energy output and the predicted load, the initial predicted net load can be determined. Then, by using the Gaussian Copula structure, the prediction error of the predicted new energy output and the prediction error of the predicted load are modeled to obtain a joint probability density function, and the prediction error sampling of the joint probability density function is performed to obtain the new energy output error sample and the load error sample of the first regional power grid at the first time. The corresponding net load error can be determined by using the new energy output error sample and the load error sample. Furthermore, based on the net load error, the initial predicted net load is corrected to accurately obtain the target predicted net load of the first regional power grid at the first time. Therefore, by adopting the technical solution of the present invention, the accuracy of net load prediction can be improved.

[0009] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings are used to better understand the present invention and do not constitute a limitation of the present invention. Figure 1 is a flow chart of a net load prediction method based on a Copula structure according to an embodiment of the present invention; Figure 2 This is a structural block diagram of a net load prediction device based on a Copula structure according to an embodiment of the present invention; Figure 3 The block diagram is a block diagram of an electronic device for implementing the method according to the embodiment of the present invention. DETAILED DESCRIPTION

[0011] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be appreciated by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present invention. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0012] Figure 1 It is a flow chart of a net load prediction method based on Copula structure according to an embodiment of the present invention.

[0013] like Figure 1 As shown, the net load prediction method based on the Copula structure may include: S110, using a dictionary learning method to learn the new energy output and load of the first regional power grid at the first time, to obtain the predicted new energy output and predicted load of the first regional power grid at the first time; S120, determining a predicted net load of the first regional power grid at a first time based on the predicted new energy output and the predicted load; S130, using a Gaussian Copula structure, models the prediction error of the predicted renewable energy output and the prediction error of the predicted load to obtain a joint probability density function; S140, performing prediction error sampling on the joint probability density function to obtain a new energy output error sample and a load error sample of the first regional power grid at the first time; S150, determining a net load error based on the new energy output error sample and the load error sample; S160, based on the net load error, correct the predicted net load to obtain a target predicted net load of the first regional power grid at the first time.

[0014] Exemplarily, the new energy output may include wind power output and photovoltaic power output. In some examples, it may also include hydropower output, thermal power output, etc.

[0015] Exemplarily, the first time may be divided into multiple time moments under multiple dates.

[0016] It can be understood that the new energy output and load of the first regional power grid at the first time can be regarded as the actual measured values.

[0017] Exemplarily, before learning the new energy output and load of the first regional power grid at various times on various dates, new energy output data and load data reflecting different seasons and weather conditions can be selected, and outlier processing can be performed on the new energy output data and load data to obtain new energy output data and load data for dictionary learning.

[0018] Exemplarily, the method for handling outliers is: based on the new energy output and load output at each time point, the mean and standard deviation of the new energy output, as well as the mean and standard deviation of the load output are determined respectively, and the standard scores of the new energy output and the load at each time point are calculated using the mean and standard deviation corresponding to the new energy output and the load output respectively.

[0019] For example, the standard score can be calculated using the following formula: : ; in, Indicates the new energy output or load output at a certain point in time. Represents the mean value corresponding to the output of new energy or load, Indicates the standard deviation of renewable energy output or load output.

[0020] Next, if the standard fraction of renewable energy output or load at each time point If it is greater than a preset value, such as 3, the new energy output or load output at that time point is set to the corresponding mean value.

[0021] Exemplarily, the predicted load is subtracted from the predicted new energy output to obtain the predicted net load.

[0022] For example, if the new energy output includes wind power output and photovoltaic output, the predicted load is subtracted from the sum of the predicted wind power output and the predicted photovoltaic output to obtain the predicted net load, as shown in the following formula: ; in, express The predicted net load at the time, express The forecast load at the time, express Predicted wind power output at any given moment, express Predicted PV output at any given moment.

[0023] Exemplarily, the prediction error of the predicted new energy output can be obtained by subtracting the new energy output at the first time from the predicted new energy output. The prediction error of the predicted load can be obtained by subtracting the load at the first time from the predicted load. If the above-mentioned first time is divided into multiple moments under multiple dates, these actual output data or predicted output data can be constructed into a matrix, and then the corresponding calculations are performed. For example, the predicted new energy output matrix is ​​subtracted from the predicted new energy output matrix to obtain the prediction error matrix of the predicted new energy output matrix, and the predicted load matrix is ​​subtracted from the load matrix to obtain the prediction error matrix of the predicted load matrix.

[0024] Exemplarily, for example, if the new energy output includes wind power output and photovoltaic output, the prediction errors of the predicted load, predicted photovoltaic output, and predicted wind power output are calculated through the aforementioned steps. Since these prediction errors contain common dependent factors, such as climate, etc., through steps S130 to S140, the prediction errors with interdependent relationships are combined, and in the subsequent step S150, the initial prediction value is corrected, which can improve the accuracy of the net load prediction.

[0025] Exemplarily, a plurality of new energy output error samples and a plurality of load error samples may be extracted, the extracted samples are averaged, and then the mean of the new energy output error samples is subtracted from the mean of the load error samples to obtain a net load error.

[0026] For example, the Monte Carlo sampling method can be used to extract the new energy output error samples and the load error samples from the joint probability density function obtained in the previous step until a sufficient sample size is obtained. Of course, other sampling methods can also be used to extract the new energy output error samples and the load error samples from the joint probability density function.

[0027] It can be understood that the errors involved in the embodiments of the present invention are all prediction errors.

[0028] For example, gradually increase the number of samples of the new energy output error and load error samples, and when the fluctuation of the mean and variance of the new energy output error samples and the load error samples becomes very small and converges, stop increasing the number of samples and determine the minimum sample size. .

[0029] For example, for Samples , and ,calculate , as follows: ; in, express The moment The predicted net load error, express The moment Load error samples, express The moment Wind power output error samples, express The moment PV output error samples.

[0030] For example, the logarithmic value is negative The predicted net load errors at the time are averaged, and the absolute value of the average is added to the initial predicted net load to obtain the upper boundary of the predicted net load, as follows: ; ; in, express The upper bound of the predicted net load at the time, express The predicted net load at the time, Indicates that the value is negative The mean of the predicted net load error at time.

[0031] In this example, if the error of a predicted value is negative, it means that the predicted value is smaller than the measured value, and the absolute value of the error needs to be added to correct the predicted value.

[0032] In this example, the maximum possible net load demand is estimated by modifying the initial forecast net load by adding the mean absolute value of the negative net load errors.

[0033] For example, the logarithmic value is positive The predicted net load errors at the time are averaged, and the absolute value of the average is subtracted from the initial predicted net load to obtain the lower boundary of the predicted net load, as follows: ; ; in, express The lower bound of the forecast net load at time express The predicted net load at the time, Indicates that the value is positive The mean of the predicted net load error at time.

[0034] In this example, if the error of a predicted value is positive, it means that the predicted value is higher than the measured value, and the absolute value of the error needs to be subtracted to correct the predicted value.

[0035] In this example, the minimum possible net load demand is estimated by modifying the initial forecast net load by subtracting the mean absolute value of the positive net load errors.

[0036] For example, the revised predicted net load may be: ; in, express The predicted net load after time correction is The mean of the lower and upper bounds of the forecasted net load at the time.

[0037] According to the above implementation, a dictionary learning method is used to learn the new energy output and load of the first regional power grid at the first time, and the predicted new energy output and predicted load of the first regional power grid at the first time are obtained. In this way, based on the predicted new energy output and the predicted load, the initial predicted net load can be determined. Then, the prediction error of the predicted new energy output and the predicted load are modeled by using a Gaussian Copula structure to obtain a joint probability density function, and the prediction error is sampled for the joint probability density function to obtain the new energy output error sample and the load error sample of the first regional power grid at the first time. The corresponding net load error can be determined using the new energy output error sample and the load error sample. In this way, the net load error that takes into account the dependency between the new energy output and the load can be obtained. Furthermore, based on the net load error, the initial predicted net load is corrected to accurately obtain the target predicted net load of the first regional power grid at the first time. Therefore, the technical solution of the present invention can improve the accuracy of net load prediction.

[0038] In one embodiment, a dictionary learning method is used to learn the new energy output and load of the first regional power grid at each time on each date to obtain the predicted new energy output and predicted load of the first regional power grid at the first time, including: initializing the new energy output dictionary and the new energy output sparse matrix, as well as the load dictionary and the load sparse matrix based on the new energy output and load of the first regional power grid at each time on each date; for the new energy output and load, for each column and row, based on the matrix obtained by multiplying the dictionary of deleted columns with the sparse matrix of deleted corresponding rows, the dictionary is updated column by column and the sparse matrix is ​​updated row by row to obtain the new energy output dictionary after each column is updated and the new energy output sparse matrix after each row is updated, as well as the load dictionary after each column is updated and the load sparse matrix after each row is updated; based on the product of the updated new energy output dictionary after each column is updated and the new energy output sparse matrix after each row is updated, the predicted new energy output of the first regional power grid at the first time is determined; based on the product of the updated load dictionary after each column is updated and the load sparse matrix after each row is updated, the predicted load of the first regional power grid at the first time is determined.

[0039] In one embodiment, based on the new energy output and load of the first regional power grid at each time on each date within the first time, the new energy output dictionary and the new energy output sparse matrix, as well as the load dictionary and the load sparse matrix are initialized, including: based on the new energy output and load of the first regional power grid at each time on each date, respectively determining the new energy output sample matrix and the load sample matrix, wherein each row in the new energy output sample matrix and the load sample matrix represents each time, and each column represents each date; based on the former in the left singular matrix after the new energy output sample matrix is ​​decomposed column vectors to determine the new energy output dictionary; based on the standard normal distribution of each column output in the new energy output sample matrix, determine the non-zero elements of the corresponding column in the new energy output sparse matrix; based on the front of the left singular matrix after the load sample matrix is ​​decomposed column vectors to determine the load dictionary; based on the standard normal distribution of the forces in each column of the load sample matrix, the non-zero elements of the corresponding column in the load sparse matrix are determined.

[0040] Exemplarily, the new energy output sample matrix may include OK The new energy output of the series, among which The columns include Dates from the first moment to the The new energy output at this moment, The rows include dates from 1 to 2. Date The new energy output at each moment. The new energy output sample matrix can include the wind power output sample matrix and the photovoltaic output sample matrix. is a positive integer, Less than or equal to , is a positive integer, Less than or equal to .

[0041] For example, the load sample matrix may include OK The load of the column, where The columns include Dates from the first moment to the The load at the moment, The rows include the dates from the 1st to the next From time to Date load at a moment.

[0042] Exemplarily, the new energy output dictionary may include OK The load dictionary can include OK The column load, where .

[0043] For example, both the new energy output sparse matrix and the load sparse matrix can be OK The process of initializing these two matrices can be: For sparse matrices ,set up The sparsity of is 0.1, which means that the matrix 10% of the elements in each column are non-zero. The position of the non-zero element in each column is randomly selected, and the value of the non-zero element at that position is generated from the standard normal distribution of the corresponding column elements in the corresponding sample matrix, so as to ensure the uniformity of the elements in the initialized sparse matrix.

[0044] Exemplarily, the above-mentioned new energy output sample matrix can be divided into a photovoltaic output sample matrix and a wind power output sample matrix. Similarly, the photovoltaic output dictionary and the wind power output dictionary, as well as the photovoltaic output sparse matrix and the wind power output sparse matrix can be calculated according to the above example.

[0045] According to the above implementation, the new energy output dictionary and the new energy output sparse matrix, as well as the load dictionary and the load sparse matrix are initialized, which can ensure the uniformity of the elements during initialization.

[0046] In one embodiment, for the new energy output and load, the dictionary is updated column by column and the sparse matrix is ​​updated row by row based on the matrix obtained by multiplying the dictionary with the deleted columns and the sparse matrix with the corresponding rows deleted, including: Column and new energy output sparse matrix The rows are deleted to obtain the new energy output intermediate dictionary and the new energy output intermediate sparse matrix; the new energy output sample matrix is ​​subtracted from the product between the new energy output intermediate dictionary and the new energy output intermediate sparse matrix to obtain the new energy output error matrix; based on the first column vector in the left singular matrix after the new energy output error matrix is ​​decomposed, the first column vector in the new energy output dictionary is updated. Column elements; based on the product of the first singular value in the diagonal matrix after the new energy output error matrix is ​​decomposed and the first column vector in the right singular matrix, update the first column vector in the new energy output sparse matrix Row element.

[0047] In one embodiment, the method further comprises: Column and load output sparse matrix The rows are deleted to obtain the load output intermediate dictionary and the load output intermediate sparse matrix; the load output sample matrix is ​​subtracted from the product between the load output intermediate dictionary and the load output intermediate sparse matrix to obtain the load output error matrix; based on the first column vector in the left singular matrix after the load output error matrix is ​​decomposed, the first column vector in the load output dictionary is updated. Column elements; based on the product of the first singular value in the diagonal matrix after the load output error matrix is ​​decomposed and the first column vector in the right singular matrix, update the first column vector in the load output sparse matrix. Row elements, where is a positive integer, Less than or equal to .

[0048] For example, removing the dictionary The Column, get dictionary , remove the sparse matrix The Rows, get a sparse matrix , and calculate the error matrix ,in, is the sample matrix.

[0049] For example, the error matrix Do singular value decomposition, that is ,in, is a left singular matrix, is a diagonal matrix, is the right singular matrix. Then, take the left singular matrix Update the dictionary with the first column in The Column elements can ensure that the update direction of the dictionary atoms can be updated along the direction most relevant to the original data. And, the diagonal matrix The first singular value and right singular matrix of The result of multiplying the first column of The row elements, which ensures that the matrix is ​​sparse No. The coefficients of the rows and the updated dictionary Corresponding.

[0050] For example, from Start executing the above steps, and after each execution of the above steps Add 1 until each column vector and sparse matrix in the dictionary All row vectors in have been updated.

[0051] Exemplarily, the product of the updated new energy output dictionary of each column and the updated new energy output sparse matrix of each row is used as the predicted new energy output of the first regional power grid at the first time. For example, the product of the updated wind power output dictionary and the updated wind power output sparse matrix is ​​used as the predicted wind power output. For another example, the product of the updated photovoltaic output dictionary and the updated photovoltaic output sparse matrix is ​​used as the predicted photovoltaic output.

[0052] It can be understood that the predicted new energy output at the first time here can be a matrix, and the matrix includes the predicted new energy output at each time on each date in the first time.

[0053] Exemplarily, the product of the updated load dictionary in each column and the updated load sparse matrix in each row is used as the predicted load of the first regional power grid at the first time.

[0054] It can be understood that the predicted load at the first time here can be a matrix, which includes the load at each time on each date in the first time.

[0055] In one embodiment, predicting new energy output includes predicting photovoltaic output and predicting wind power output. A Gaussian Copula structure is used to model the prediction error of predicting new energy output and the prediction error of predicting load to obtain a joint probability density function, including: based on a first marginal distribution function of the prediction error of predicting wind power output, a second marginal distribution function of the prediction error of predicting photovoltaic output, and a third marginal distribution function of predicting load, a first conditional distribution function of wind power output error, a second conditional distribution function of photovoltaic output error, and a joint conditional distribution function between wind power output error and photovoltaic output error under a specified load error is determined; based on the first marginal distribution function, the second marginal distribution function, the third marginal distribution function, the first conditional distribution function, the second conditional distribution function, and the inverse cumulative distribution function of the standard normal distribution corresponding to each of the joint conditional distribution function, a first dependency relationship between load error and wind power output error, a second dependency relationship between load error and photovoltaic output error, and a conditional dependency relationship between wind power output error and photovoltaic output error under a specified load error is determined; based on the first dependency relationship, the second dependency relationship, and the conditional dependency relationship, a joint probability density function is determined.

[0056] For example, the load prediction error can be obtained by subtracting the actual load from the predicted load obtained above: , subtracting the actual wind power output from the predicted wind power output obtained above can obtain the wind power output prediction error The photovoltaic output prediction error can be obtained by subtracting the actual photovoltaic output from the predicted photovoltaic output .Right now, , , The predicted values ​​are , , The difference from the corresponding actual value.

[0057] Exemplarily, in order to construct the joint probability density function of these three prediction errors, in this example, let is the load forecast error , wind power output prediction error , PV output prediction error The three-dimensional joint distribution function of is as follows: ; in, , , are the marginal distribution functions of load forecast error, wind power output forecast error and photovoltaic output forecast error, respectively. It is a Copula function, which is used to describe the dependency relationship between the three.

[0058] In this example, Gaussian Copula is used for modeling, and the covariance between the sample data of these prediction errors is used to estimate the correlation coefficient between the corresponding two variables. Furthermore, these correlation coefficients can be used to capture the linear dependency between load, wind power, and photovoltaic power, thereby obtaining the joint probability density function between the three prediction error variables.

[0059] For example, according to the marginal distribution of three random variables (load forecast error, wind power output forecast error, and photovoltaic output forecast error), the respective cumulative distribution functions are calculated, and then converted into standard normal variables through the inverse cumulative distribution function of the standard normal distribution. In this way, the scales between different variables can be unified, making the contributions of different variables in the model more balanced.

[0060] Exemplarily, the above function may include the following: , , , , and .

[0061] in, It satisfies the standard normal distribution A random variable. is the inverse cumulative distribution function of the standard normal distribution. To predict the wind power output The first marginal distribution function of The prediction error of photovoltaic power The second marginal distribution function of is the prediction error of the predicted wind load The third marginal distribution function of .

[0062] in, and Respectively represent the load error When the wind power output error Conditional distribution function and photovoltaic output error The conditional distribution function of . Indicates the load error When the wind power output error The error with photovoltaic output The joint conditional distribution function of the joint.

[0063] Exemplarily, the three-dimensional joint distribution function is triple-partially derived to obtain the joint probability density function, as shown in the following formula: ; in, represents the Copula density function, , , The prediction error , , The probability density function of .

[0064] For example, the dependencies between the three variables are Decomposed into the above three dependencies, we get the new joint probability density function as follows: ; ; ; ; In one embodiment, the first dependency is: ; in, Indicates the first dependency, represents the correlation coefficient matrix between load error and wind power output error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the third marginal distribution function, Represents the inverse cumulative distribution function of the standard normal distribution corresponding to the first marginal distribution function.

[0065] In one implementation, the second dependency is: ; in, represents the second dependency, which represents the correlation coefficient matrix between load error and photovoltaic output error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the third marginal distribution function, Represents the inverse cumulative distribution function of the standard normal distribution corresponding to the second marginal distribution function.

[0066] Exemplarily, the correlation coefficient matrix between load error and wind power output error is: .

[0067] Exemplarily, the correlation coefficient matrix between load error and photovoltaic output error is: .

[0068] in, is the correlation coefficient between load error and wind power output error, as follows: ; in, is the correlation coefficient between load error and photovoltaic output error, as follows: ; Among them, the correlation coefficient between wind power output error and photovoltaic output error is as follows: ; in, represents the covariance of load forecast error and wind power output forecast error, represents the covariance of load forecast error and PV output forecast error, represents the covariance of wind power output prediction error and photovoltaic output prediction error, , , They represent the standard deviation of load forecast error, wind power output forecast error, and photovoltaic output forecast error respectively.

[0069] It can be understood from the above analysis that the forecast errors of the above-mentioned outputs and loads can be matrices, including the forecast errors at each time on each date in the first time, and the covariance between them can be obtained by calculating the load forecast error matrix and the wind power output forecast error matrix. The standard deviation of the corresponding variable can be obtained by calculating each forecast error matrix.

[0070] In one implementation, the above conditional dependency is: ; in, Represents conditional dependencies, represents the correlation coefficient matrix between wind power output error and photovoltaic output error under a given load error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the first conditional distribution function, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the second conditional distribution function, Represents the inverse cumulative distribution function of the standard normal distribution corresponding to the joint conditional distribution function.

[0071] Exemplarily, the correlation coefficient matrix between wind power output error and photovoltaic output error under a given load error is: ; in, is the correlation coefficient between wind power output error and load error under given load error conditions, is the correlation coefficient between the photovoltaic output error and the load error under given load error conditions, It is the correlation coefficient between wind power output error and photovoltaic output error under given load error conditions.

[0072] For example, the correlation coefficient between the wind power output error and the load error under a given load error condition is as follows: ; For example, the correlation coefficient between the photovoltaic output error and the load error under a given load error condition is as follows: ; For example, the correlation coefficient between the wind power output error and the photovoltaic output error under a given load error condition is as follows: .

[0073] According to the above implementation, the Gaussian Copula structure can accurately describe the dependency between the output prediction errors and between the output prediction errors and the load prediction errors, thereby accurately describing the joint probability density function formed by the output prediction errors and the load prediction errors.

[0074] Figure 2 It is a structural block diagram of a net load prediction device based on a Copula structure according to an embodiment of the present invention.

[0075] like Figure 2 As shown, the net load prediction device based on the Copula structure may include: The output and load prediction module 210 is used to learn the new energy output and load of the first regional power grid at the first time by using a dictionary learning method to obtain the predicted new energy output and predicted load of the first regional power grid at the first time; A net load determination module 220, configured to determine a predicted net load of the first regional power grid within the first time based on the predicted new energy output and the predicted load; A prediction error modeling module 230 is used to model the prediction error of the predicted new energy output and the prediction error of the predicted load by using a Gaussian Copula structure to obtain a joint probability density function; A prediction error sampling module 240 is used to perform prediction error sampling on the joint probability density function to obtain a new energy output error sample and a load error sample of the first regional power grid at the first time; A net load error determination module 250, configured to determine a net load error based on the new energy output error sample and the load error sample; The net load correction module 260 is used to correct the predicted net load based on the net load error to obtain a target predicted net load of the first regional power grid at the first time.

[0076] In one implementation, the output and load prediction module 210 includes: An initialization unit, configured to initialize a new energy output dictionary and a new energy output sparse matrix, as well as a load dictionary and a load sparse matrix based on the new energy output and load of the first regional power grid at each time on each date within the first time; A matrix updating unit, for the new energy output and the load, based on the matrix obtained by multiplying the dictionary of deleted columns with the sparse matrix of deleted corresponding rows, updates the dictionary columns and updates the sparse matrix rows, so as to obtain the new energy output dictionary with updated columns and the new energy output sparse matrix with updated rows, as well as the load dictionary with updated columns and the load sparse matrix with updated rows; A first prediction unit, configured to determine a predicted renewable energy output of the first regional power grid at the first time based on a product of the updated renewable energy output dictionary of each column and the updated renewable energy output sparse matrix of each row; The second prediction unit is used to determine the predicted load of the first regional power grid at the first time based on the product of the updated load dictionary of each column and the updated load sparse matrix of each row.

[0077] In one implementation, the initialization unit is specifically used to: Based on the new energy output and load of the first regional power grid at each time on each date, determine a new energy output sample matrix and a load sample matrix respectively, wherein each row in the new energy output sample matrix and the load sample matrix represents each time, and each column represents each date; Based on the former in the left singular matrix after the new energy output sample matrix is ​​decomposed column vectors to determine the new energy output dictionary, where is a positive integer; Based on the standard normal distribution of each column output in the new energy output sample matrix, determine the non-zero elements of the corresponding column in the new energy output sparse matrix; Based on the load sample matrix decomposition of the left singular matrix A column vector, determining the load dictionary; Based on the standard normal distribution of each column force in the load sample matrix, the non-zero elements of the corresponding column in the load sparse matrix are determined.

[0078] In one implementation, the matrix updating unit is specifically configured to: The new energy output dictionary The columns and the new energy output sparse matrix Delete the rows to obtain the intermediate dictionary of new energy output and the intermediate sparse matrix of new energy output; Subtract the product of the new energy output intermediate dictionary and the new energy output intermediate sparse matrix from the new energy output sample matrix to obtain a new energy output error matrix; Based on the first column vector in the left singular matrix after the new energy output error matrix is ​​decomposed, the first column vector in the new energy output dictionary is updated. Column elements; Based on the product of the first singular value in the diagonal matrix after the new energy output error matrix is ​​decomposed and the first column vector in the right singular matrix, the first singular value in the new energy output sparse matrix is ​​updated. row element; The load output dictionary The column and the load output sparse matrix Delete the rows to obtain the load output intermediate dictionary and load output intermediate sparse matrix; Subtract the product of the load output intermediate dictionary and the load output intermediate sparse matrix from the load output sample matrix to obtain a load output error matrix; Based on the first column vector in the left singular matrix after the load output error matrix is ​​decomposed, update the first column vector in the load output dictionary. Column elements; Based on the product of the first singular value in the diagonal matrix after the decomposition of the load output error matrix and the first column vector in the right singular matrix, the first singular value in the load output sparse matrix is ​​updated. Row elements, where is a positive integer, Less than or equal to .

[0079] In one embodiment, the predicted new energy output includes predicted photovoltaic output and predicted wind power output, and the prediction error modeling module 230 includes: a distribution function determination unit, configured to determine a first conditional distribution function of a wind power output error, a second conditional distribution function of a photovoltaic output error, and a joint conditional distribution function between a wind power output error and a photovoltaic output error under a specified load error based on a first marginal distribution function of a prediction error of the predicted wind power output, a second marginal distribution function of a prediction error of the predicted photovoltaic output, and a third marginal distribution function of the predicted load; a dependency determination unit, configured to determine a first dependency relationship between a load error and a wind power output error, a second dependency relationship between a load error and a photovoltaic output error, and a conditional dependency relationship between a wind power output error and a photovoltaic output error under a specified load error based on the first marginal distribution function, the second marginal distribution function, the third marginal distribution function, the first conditional distribution function, the second conditional distribution function, and the inverse cumulative distribution function of the standard normal distribution corresponding to each of the joint conditional distribution function; A density function determination unit is used to determine the joint probability density function based on the first dependency, the second dependency and the conditional dependency.

[0080] In one embodiment, the first dependency relationship is: ; in, represents the first dependency, represents the correlation coefficient matrix between load error and wind power output error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the third marginal distribution function, The inverse cumulative distribution function of the standard normal distribution corresponding to the first marginal distribution function is represented.

[0081] In one implementation, the second dependency is: ; in, represents the second dependency, represents the correlation coefficient matrix between the load error and the photovoltaic output error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the third marginal distribution function, The inverse cumulative distribution function of the standard normal distribution corresponding to the second marginal distribution function is represented.

[0082] In one embodiment, the conditional dependency is: ; in, represents the conditional dependency, represents the correlation coefficient matrix between wind power output error and photovoltaic output error under a given load error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the first conditional distribution function, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the second conditional distribution function, The inverse cumulative distribution function of the standard normal distribution corresponding to the joint conditional distribution function is represented.

[0083] For the description of specific functions and examples of each module and submodule of the system in the embodiment of the present invention, reference can be made to the relevant description of the corresponding steps in the above method embodiment, which will not be repeated here.

[0084] In the technical solution of the present invention, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0085] According to an embodiment of the present invention, the present invention also provides a system and a readable storage medium.

[0086] Figure 3 A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0087] like Figure 3 As shown, the electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 to a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0088] Multiple components in the electronic device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0089] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as a net load prediction method based on a Copula structure. For example, in some embodiments, the net load prediction method based on a Copula structure may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the net load prediction method based on the Copula structure described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the net load prediction method based on the Copula structure in any other appropriate manner (for example, by means of firmware).

[0090] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0091] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0092] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0093] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0094] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0095] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0096] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0097] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A net load forecasting method based on Copula structure, characterized in that: include: Using a dictionary learning method, learning the new energy output and load of the first regional power grid at the first time, to obtain the predicted new energy output and predicted load of the first regional power grid at the first time; Determining a predicted net load of the first regional power grid within the first time based on the predicted new energy output and the predicted load; A Gaussian Copula structure is used to model the prediction error of the predicted new energy output and the prediction error of the predicted load to obtain a joint probability density function; Performing prediction error sampling on the joint probability density function to obtain a new energy output error sample and a load error sample of the first regional power grid at the first time; Determining a net load error based on the new energy output error sample and the load error sample; Based on the net load error, the predicted net load is corrected to obtain a target predicted net load of the first regional power grid at the first time.

2. The method according to claim 1, characterized in that The method of adopting a dictionary learning method to learn the new energy output and load of the first regional power grid at the first time, and obtaining the predicted new energy output and predicted load of the first regional power grid at the first time, includes: Initialize a new energy output dictionary and a new energy output sparse matrix, as well as a load dictionary and a load sparse matrix based on the new energy output and load of the first regional power grid at each time on each date within the first time; For the new energy output and the load, based on the matrix obtained by multiplying the dictionary of deleted columns with the sparse matrix of deleted corresponding rows, the dictionary is updated column by column and the sparse matrix is ​​updated row by row, so as to obtain the new energy output dictionary after each column is updated and the new energy output sparse matrix after each row is updated, as well as the load dictionary after each column is updated and the load sparse matrix after each row is updated; Determine the predicted renewable energy output of the first regional power grid at the first time based on the product of the updated renewable energy output dictionary of each column and the updated renewable energy output sparse matrix of each row; Based on the product of the updated load dictionary of each column and the updated load sparse matrix of each row, the predicted load of the first regional power grid at the first time is determined.

3. The method according to claim 2, characterized in that The new energy output and load of the first regional power grid at each time on each date within the first time, initializing the new energy output dictionary and the new energy output sparse matrix, as well as the load dictionary and the load sparse matrix, include: Based on the new energy output and load of the first regional power grid at each time on each date, determine a new energy output sample matrix and a load sample matrix respectively, wherein each row in the new energy output sample matrix and the load sample matrix represents each time, and each column represents each date; Based on the former in the left singular matrix after the new energy output sample matrix is ​​decomposed column vectors to determine the new energy output dictionary, where is a positive integer; Based on the standard normal distribution of each column output in the new energy output sample matrix, determine the non-zero elements of the corresponding column in the new energy output sparse matrix; Based on the load sample matrix decomposition of the left singular matrix A column vector, determining the load dictionary; Based on the standard normal distribution of each column force in the load sample matrix, the non-zero elements of the corresponding column in the load sparse matrix are determined.

4. The method according to claim 3, characterized in that For the new energy output and the load, updating the dictionary columns and updating the sparse matrix rows based on matrices obtained by multiplying the dictionary with deleted columns and the sparse matrix with deleted corresponding rows respectively, includes: The new energy output dictionary The columns and the sparse matrix of the new energy output Delete the rows to obtain the intermediate dictionary of new energy output and the intermediate sparse matrix of new energy output; Subtract the product of the new energy output intermediate dictionary and the new energy output intermediate sparse matrix from the new energy output sample matrix to obtain a new energy output error matrix; Based on the first column vector in the left singular matrix after the new energy output error matrix is ​​decomposed, the first column vector in the new energy output dictionary is updated. Column elements; Based on the product of the first singular value in the diagonal matrix after the new energy output error matrix is ​​decomposed and the first column vector in the right singular matrix, the first singular value in the new energy output sparse matrix is ​​updated. row element; The load output dictionary The column and the load output sparse matrix Delete the rows to obtain the load output intermediate dictionary and load output intermediate sparse matrix; Subtract the product of the load output intermediate dictionary and the load output intermediate sparse matrix from the load output sample matrix to obtain a load output error matrix; Based on the first column vector in the left singular matrix after the load output error matrix is ​​decomposed, update the first column vector in the load output dictionary. Column elements; Based on the product of the first singular value in the diagonal matrix after the decomposition of the load output error matrix and the first column vector in the right singular matrix, the first singular value in the load output sparse matrix is ​​updated. Row elements, where is a positive integer, Less than or equal to .

5. The method according to claim 1, characterized in that The predicted new energy output includes predicted photovoltaic output and predicted wind power output. The Gaussian Copula structure is used to model the prediction error of the predicted new energy output and the prediction error of the predicted load to obtain a joint probability density function, including: Based on the first marginal distribution function of the prediction error of the predicted wind power output, the second marginal distribution function of the prediction error of the predicted photovoltaic output, and the third marginal distribution function of the predicted load, determine the first conditional distribution function of the wind power output error, the second conditional distribution function of the photovoltaic output error, and the joint conditional distribution function between the wind power output error and the photovoltaic output error under the specified load error; Based on the inverse cumulative distribution function of the standard normal distribution corresponding to each of the first marginal distribution function, the second marginal distribution function, the third marginal distribution function, the first conditional distribution function, the second conditional distribution function, and the joint conditional distribution function, determine a first dependency relationship between the load error and the wind power output error, a second dependency relationship between the load error and the photovoltaic output error, and a conditional dependency relationship between the wind power output error and the photovoltaic output error under a specified load error; Based on the first dependency, the second dependency and the conditional dependency, the joint probability density function is determined.

6. The method according to claim 5, characterized in that The first dependency is: ； in, represents the first dependency, represents the correlation coefficient matrix between load error and wind power output error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the third marginal distribution function, The inverse cumulative distribution function of the standard normal distribution corresponding to the first marginal distribution function is represented.

7. The method according to claim 5, characterized in that The second dependency is: ； in, represents the second dependency, represents the correlation coefficient matrix between the load error and the photovoltaic output error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the third marginal distribution function, The inverse cumulative distribution function of the standard normal distribution corresponding to the second marginal distribution function is represented.

8. The method according to claim 5, characterized in that The conditional dependency is: ； in, represents the conditional dependency, represents the correlation coefficient matrix between wind power output error and photovoltaic output error under a given load error, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the first conditional distribution function, represents the inverse cumulative distribution function of the standard normal distribution corresponding to the second conditional distribution function, The inverse cumulative distribution function of the standard normal distribution corresponding to the joint conditional distribution function is represented.

9. A net load prediction device based on Copula structure, characterized in that: include: An output and load forecasting module, configured to learn the new energy output and load of the first regional power grid at the first time by using a dictionary learning method, and obtain the predicted new energy output and predicted load of the first regional power grid at the first time; A net load determination module, configured to determine a predicted net load of the first regional power grid at the first time based on the predicted new energy output and the predicted load; A prediction error modeling module is used to model the prediction error of the predicted new energy output and the prediction error of the predicted load by using a Gaussian Copula structure to obtain a joint probability density function; A prediction error sampling module, used to perform prediction error sampling on the joint probability density function to obtain a new energy output error sample and a load error sample of the first regional power grid at the first time; A net load error determination module, configured to determine a net load error based on the new energy output error sample and the load error sample; A net load correction module is used to correct the predicted net load based on the net load error to obtain a target predicted net load of the first regional power grid at the first time.

10. A net load forecasting system based on Copula structure, characterized in that: include: at least one processor, and a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the processor is used to obtain the instructions from the memory and execute the instructions, so that the at least one processor can execute the method according to any one of claims 1 to 8.

Images