Multi-region new energy installed capacity joint prediction method, system and device considering region coupling factor, and storage medium

Abstract

The invention relates to the technical field of new energy capacity prediction, in particular to a multi-region new energy installed capacity joint prediction method, system and device considering region coupling factors and a storage medium. Related data of the new energy development influence factors are obtained through analysis; performing data preprocessing on the related data of the new energy development influence factors, and constructing an influence factor set of the new energy installed capacity considering the regional coupling factors for the preprocessed data; constructing a joint prediction model of multi-region new energy installation, and training the joint prediction model based on the influence factor set to obtain a corrected prediction model; based on different new energy development scenes, respectively determining related multi-scene influence factor data sets; predicting the installed capacity of the new energy under the multi-scene influence factor data set by using the corrected prediction model; the fluctuation problem in the installed capacity of the new energy is solved, and the prediction accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy capacity prediction, and in particular to a method, system, device and storage medium for joint prediction of multi-region new energy installed capacity taking into account regional coupling factors. Background Art

[0002] With the acceleration of global energy transformation, China has made significant progress in wind and solar power generation, especially in the northwest region, which is rich in natural resources. The installed capacity of wind and solar power generation has continued to rise. However, due to the differences in natural resource distribution, power demand and transmission capacity in various regions, achieving multi-regional joint forecasting and fully considering the mutual influence between regions has become a core challenge in the forecast of new energy installed capacity.

[0003] Among the related technologies, the prediction of new energy installed capacity is mostly concentrated in a single region, lacking a systematic analysis of cross-regional coordinated development and mutual influence between regions. For example, traditional capacity prediction methods such as grey models and artificial neural networks have shown certain advantages in short-term predictions in a single region. However, when faced with the complex coupling relationship between multiple regions and the volatility of prediction results, it is difficult to fully capture the dynamic interactions between regions. Especially when faced with frequent changes in cross-regional transmission capacity and policy support, it is often impossible to effectively reflect the interactions between regions, thus limiting the accuracy of new energy installed capacity prediction and its application capabilities in multi-regional coordinated development.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The technical problem to be solved by the present invention is: to provide a method, system, device and storage medium for joint prediction of multi-regional new energy installed capacity taking into account regional coupling factors, introduce the proportion of new energy installed capacity in previous years to characterize the relationship between regional coupling factors, and establish a PLSR-GARCH joint prediction model to solve the volatility problem in new energy installed capacity, thereby improving the accuracy of new energy installed capacity prediction.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: a method for joint prediction of multi-regional new energy installed capacity considering regional coupling factors, comprising the following steps:

[0007] Collect and organize relevant historical data on the installed capacity of new energy in various regions and form a relevant database, and analyze the relevant historical data to obtain relevant data on factors affecting the development of new energy;

[0008] Preprocessing the data related to the factors affecting the development of new energy, and constructing a set of factors affecting the installed capacity of new energy that considers regional coupling factors based on the preprocessed data;

[0009] Constructing a PLSR-GARCH joint prediction model for renewable energy installed capacity in multiple regions, and training the joint prediction model based on the influencing factor set to obtain a revised prediction model;

[0010] Based on different new energy development scenarios, determine the relevant multi-scenario influencing factor data sets;

[0011] The modified prediction model is used to predict the installed capacity of new energy under multiple scenario influencing factor data sets.

[0012] Furthermore, the data preprocessing of the data related to the factors affecting the development of new energy includes:

[0013] Retrieving the natural resource data of each region in the relevant historical data, and converting the natural resource data of the same type into a unified format;

[0014] Obtain the thermal power grid-connected electricity price and renewable energy grid-connected electricity price in each region, and determine the renewable energy grid-connected electricity price data for each region based on the difference between the thermal power grid-connected electricity price and renewable energy grid-connected electricity price;

[0015] All relevant data are scaled to the range of [0,1] according to the proportional relationship and normalized uniformly;

[0016] The missing year data for each region are supplemented.

[0017] Furthermore, the set of influencing factors includes intra-regional influencing factors and inter-regional coupling factors;

[0018] The influencing factors within the region include: social electricity consumption within the region Regional GDP Development Level Cumulative length of UHV lines UHV cumulative transformer capacity New energy subsidy amount Carbon market trading volume

[0019] Inter-regional coupling factors include the overall proportion of new energy installed capacity in previous years It is expressed by the following formula:

[0020]

[0021] Where i = 1, 2, 3, ..., n, i represents the year, n is the total number of years; j = 1, 2, 3, ..., p, j represents the region number, p is the total number of regions; yi-1,j It represents the installed capacity of new energy in region j in year i-1.

[0022] Furthermore, the construction of the PLSR-GARCH joint prediction model for new energy installed capacity in multiple regions, and training the joint prediction model based on the influencing factor set to obtain a revised prediction model, includes the following steps:

[0023] Determine the influencing factor set as the independent variable, the installed capacity of new energy in each region as the dependent variable, construct matrices X and Y corresponding to the independent variable and the dependent variable respectively, and obtain the independent variable observation matrix A and the dependent variable observation matrix B respectively after standardization;

[0024] Extract the first principal components t1 and u1 of the independent variable matrix X and the dependent variable matrix Y respectively, calculate the scores of the first principal components according to the independent variable observation matrix A and the dependent variable observation matrix B respectively, and use the PLSR model to establish the residual matrix and regression equation corresponding to the first principal component;

[0025] If the difference between the absolute value of the element in the residual matrix and 0 is within the threshold range, it is determined that the regression equation corresponding to the first principal component meets the accuracy requirement, and the extraction of the next set of principal components is stopped. Otherwise, the above steps are repeated until the final regression equation is obtained;

[0026] Substitute the independent variable matrix into the regression equation to obtain the predicted value And calculate the residual sequence;

[0027] Use the residual sequence ε as input to build a GARCH model and determine the variance formula corresponding to the model;

[0028] The trained GARCH model is used to predict the standard deviation, and the predicted standard deviation is added as the independent variable X8 into the independent variable matrix. The PLSR model is updated to obtain the PLSR model corrected by the GARCH model as the final corrected prediction model.

[0029] Furthermore, the independent variable matrix X and the dependent variable matrix Y are expressed by the following formula:

[0030]

[0031] The independent variable observation matrix A and the dependent variable observation matrix B are expressed by the following formula:

[0032]

[0033] in, and are the means of the independent variable matrix X and the dependent variable matrix Y, respectively, S X and SY are the standard deviations of the independent variable matrix X and the dependent variable matrix Y respectively;

[0034] The first principal components t1 and u1 of the independent variable matrix X and the dependent variable matrix Y are expressed as follows:

[0035] t1=w 11 x1+w 12 x2+…+w 1m x m

[0036] u1=c 11 y1+c 12 y2+…+c 1p y p

[0037] The scores of the first principal components t1 and u1 are expressed as follows:

[0038]

[0039] Among them, ω1 is the weight coefficient of the first principal component t1 of the independent variable matrix X, and c1 is the weight coefficient of the first principal component u1 of the dependent variable matrix Y;

[0040] The regression equations corresponding to the first principal component are expressed as follows:

[0041]

[0042] Among them, α1=[α 11 ,α 12 ,…,α 1m ] T , β1=[β 11 ,β 12 ,…,β 1p ] T Represent the parameter vectors in the independent variable matrix A and the dependent variable matrix regression model, E l and F l They represent the residual matrices respectively;

[0043] The residual sequence ε is calculated by the following formula:

[0044]

[0045] The variance formula corresponding to the GARCH model is expressed as follows:

[0046]

[0047] Where: α0 is a constant term, α i is the weight of the past residual square term, β jis the weight of the past conditional variance term, m is the order of the lagged residual square term, and n is the order of the lagged variance term.

[0048] Furthermore, the step of determining relevant multi-scenario influencing factor data sets based on different new energy development scenarios includes the following steps:

[0049] Referring to historical economic development statistics, considering different GDP development levels and electricity demand growth scenarios, we set the annual growth rates of GDP and electricity demand under three economic development scenarios: low, medium and high.

[0050] Refer to the historical data of renewable energy grid-connected electricity price subsidies, consider different scenarios of renewable energy subsidies and carbon market trading volume growth, and set the annual growth rates of renewable energy subsidies and carbon market trading volume under three policy support scenarios: low, medium and high.

[0051] Referring to the historical development trend of the UHV industry and considering the UHV line growth scenario, the annual growth rate of the UHV cumulative line length and UHV cumulative transformer capacity under three transmission capacity development scenarios of low, medium and high are set;

[0052] Based on the annual growth rates of GDP and electricity demand under the three set economic development scenarios of low, medium and high, the annual growth rates of new energy subsidies and carbon market trading volume under the three policy support scenarios of low, medium and high, and the annual growth rates of UHV cumulative line length and UHV cumulative transformer capacity under the three transmission capacity development scenarios of low, medium and high, a unified normalization processing was performed to obtain a multi-scenario influencing factor data set.

[0053] Furthermore, the use of the modified prediction model to predict the installed capacity of new energy under multiple scenario influencing factor data sets includes the following steps:

[0054] According to the obtained revised prediction model, the standard deviation value of the installed capacity of new energy is obtained;

[0055] According to the standard deviation value and the influencing factor data selected from the multi-scenario influencing factor data set in sequence, a modified prediction model is used to predict the new energy installed capacity in the scenarios corresponding to the influencing factor data in different regions to obtain the predicted value of the new energy installed capacity.

[0056] The present invention also provides a multi-regional new energy installed capacity joint prediction system considering regional coupling factors, comprising:

[0057] Data collection module, used to collect and organize relevant historical data of renewable energy installed capacity in various regions and form a relevant database;

[0058] A data analysis module, used to analyze the relevant historical data to obtain data related to factors affecting the development of new energy;

[0059] A data processing module, used to perform data preprocessing on the data related to the influencing factors of the new energy development, and construct an influencing factor set of the new energy installed capacity considering the regional coupling factor based on the preprocessed data;

[0060] A prediction model module is used to construct a PLSR-GARCH joint prediction model for new energy installed capacity in multiple regions, and to train the joint prediction model based on the influencing factor set to obtain a revised prediction model;

[0061] The scenario setting module is used to determine the relevant multi-scenario influencing factor data sets based on different new energy development scenarios;

[0062] The capacity prediction module is used to use the modified prediction model to predict the installed capacity of new energy under multiple scenario influencing factor data sets.

[0063] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for jointly predicting multi-regional new energy installed capacity taking into account regional coupling factors as described above are implemented.

[0064] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for jointly predicting multi-regional new energy installed capacity taking into account regional coupling factors as described above are implemented.

[0065] The beneficial effects of the present invention are as follows: the present invention obtains key influencing factors through in-depth analysis of relevant historical data of various regions, so that the prediction model results can comprehensively and accurately cover different development situations under multiple scenarios; at the same time, regional coupling factors are considered to form a joint prediction of multiple regions, which is conducive to achieving cross-regional coordinated development and improving the accuracy of new energy installed capacity prediction results. The constructed PLSR-GARCH joint prediction model can effectively solve the problems of multicollinearity and time series volatility in the prediction process, and has better adaptability and flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0067] Figure 1 It is a flow chart of a method for joint prediction of multi-regional renewable energy installed capacity taking into account regional coupling factors in an embodiment of the present invention;

[0068] Figure 2 A schematic diagram of a process for constructing a modified prediction model in an embodiment of the present invention;

[0069] Figure 3 Schematic diagram of the process of confirming a multi-scenario influencing factor data set in an embodiment of the present invention;

[0070] Figure 4 A schematic diagram of a process for predicting installed capacity of new energy in an embodiment of the present invention;

[0071] Figure 5 Schematic diagram of photovoltaic installation error level in Northwest China in an embodiment of the present invention;

[0072] Figure 6 Schematic diagram of the error level of wind power installed capacity in the northwest region in an embodiment of the present invention;

[0073] Figure 7 This is a schematic diagram of photovoltaic installed capacity prediction results in a high-speed scenario in the northwest region according to an embodiment of the present invention;

[0074] Figure 8 This is a schematic diagram of photovoltaic installed capacity prediction results under a medium-speed scenario in the northwest region in an embodiment of the present invention;

[0075] Fig. 9 This is a schematic diagram of photovoltaic installed capacity prediction results in a low-speed scenario in the northwest region according to an embodiment of the present invention;

[0076] Fig.10 It is a schematic diagram of wind power installed capacity prediction results in a high-speed scenario in the northwest region according to an embodiment of the present invention;

[0077] Fig.11 It is a schematic diagram of wind power installed capacity prediction results in a medium-speed scenario in the northwest region according to an embodiment of the present invention;

[0078] Fig.12 It is a schematic diagram of wind power installed capacity prediction results in a low-speed scenario in the northwest region according to an embodiment of the present invention;

[0079] Fig.13 This is a schematic diagram of the prediction results of new energy installed capacity in the northwest region in an embodiment of the present invention;

[0080] Fig.14 Schematic diagram of the structure of a multi-regional new energy installed capacity joint prediction system considering regional coupling factors in an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0082] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation method.

[0083] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0084] like Figures 1 to 13 The multi-regional new energy installed capacity joint prediction method considering regional coupling factors includes the following steps:

[0085] Collect and organize relevant historical data on the installed capacity of new energy in various regions and form a relevant database, and analyze the relevant historical data to obtain relevant data on factors affecting the development of new energy;

[0086] Preprocess the data related to the factors affecting the development of new energy, and construct a set of factors affecting the installed capacity of new energy by considering regional coupling factors based on the preprocessed data;

[0087] Construct a PLSR-GARCH joint prediction model for renewable energy installed capacity in multiple regions, and train the joint prediction model based on the influencing factor set to obtain a revised prediction model;

[0088] Based on different new energy development scenarios, determine the relevant multi-scenario influencing factor data sets;

[0089] The modified prediction model is used to predict the installed capacity of new energy under multiple scenario influencing factor data sets.

[0090] The present invention obtains key influencing factors through in-depth analysis of relevant historical data of each region, so that the prediction model results can comprehensively and accurately cover different development situations in multiple scenarios; at the same time, regional coupling factors are considered to form a joint prediction of multiple regions, which is conducive to achieving cross-regional coordinated development and improving the accuracy of new energy installed capacity prediction results. The constructed PLSR-GARCH joint prediction model can effectively solve the problems of multicollinearity and time series volatility in the prediction process, and has better adaptability and flexibility.

[0091] By collecting and organizing historical data on new energy installed capacity in various regions, basic data support and basis are provided for subsequent analysis and prediction. Through the analysis of historical data, key factors affecting the development of new energy are identified to provide a basis for building a prediction model. Data preprocessing is performed on data related to factors affecting the development of new energy to ensure data quality and availability, laying the foundation for building an accurate prediction model. By constructing a joint prediction model of new energy installed capacity in multiple regions of PLSR-GARCH, the advantages of partial least squares regression and generalized autoregressive conditional heteroskedasticity model are combined to effectively handle the volatility and nonlinear relationship of data. The joint prediction model is trained using a set of influencing factors, and the model is corrected according to the training results to improve the prediction performance of the model. By constructing and verifying a multi-scenario influencing factor data set, the prediction results are closer to actual situations that may occur, which improves the practicality of the prediction and decision-making support capabilities, so that more reasonable planning and decisions can be made.

[0092] Based on the above embodiment, data preprocessing of data related to factors affecting the development of new energy includes:

[0093] Retrieve the natural resource data of each region in the relevant historical data and convert the natural resource data of the same type into a unified format; ensure that data from different sources and in different formats can be processed and analyzed uniformly, avoid analysis errors caused by inconsistent data formats, and improve the efficiency and accuracy of data processing;

[0094] Obtain the thermal power grid-connected electricity price and renewable energy grid-connected electricity price of each region, and determine the renewable energy grid-connected electricity price data of each region based on the difference between the thermal power grid-connected electricity price and the renewable energy grid-connected electricity price; This helps determine the economic feasibility and market competitiveness of renewable energy in different regions, more accurately evaluate the economic feasibility of renewable energy projects, and provide important economic basis for decision-making on renewable energy installed capacity;

[0095] All relevant data are scaled to the range of [0,1] according to the proportional relationship and normalized uniformly; this helps to eliminate the impact of different data dimensions and magnitudes on model training, and avoid certain features occupying too much weight in model training due to their large numerical range, thereby improving the generalization ability and prediction accuracy of the model;

[0096] The missing year data for each region are supplemented; this ensures the integrity of the data, avoids analysis bias caused by missing data, and helps improve the stability and predictive ability of the model.

[0097] Based on the above embodiment, the influencing factor set includes influencing factors within the region and coupling factors between regions;

[0098] Influencing factors within the region include: social electricity consumption within the region Regional GDP Development Level Cumulative length of UHV lines UHV cumulative transformer capacity New energy subsidy amount Carbon market trading volume

[0099] Inter-regional coupling factors include the overall proportion of new energy installed capacity in previous years It is expressed by the following formula:

[0100]

[0101] Where i = 1, 2, 3, ..., n, i represents the year, n is the total number of years; j = 1, 2, 3, ..., p, j represents the region number, p is the total number of regions; y i-1,j It represents the installed capacity of new energy in region j in year i-1.

[0102] Among them, the social electricity consumption in the region and regional GDP development level The factors affecting electricity demand are the social electricity consumption in the region. It reflects the level of regional economic activity and electricity consumption demand, as well as the regional GDP development level. Measures the level of regional economic development; Cumulative length of UHV lines and UHV cumulative transformer capacity The factors affecting cross-regional transmission capacity are the cumulative length of UHV lines. Indicates the scale and transmission capacity of the power grid, and the cumulative UHV transformer capacity Displays the grid's ability to transmit large amounts of electricity; the amount of new energy subsidies and carbon market trading volume The factors affecting policy support include the amount of new energy subsidies. Reflects the government's support for the new energy industry; carbon market trading volume Reflects the impact of carbon emission policies on new energy investment; the overall proportion of new energy installed capacity in previous years As the inter-regional coupling factor, the overall proportion of new energy installed capacity in different regions is taken into account, reflecting the mutual influence and synergy between regions.

[0103] The set of influencing factors provides a comprehensive perspective for the prediction model, which enables the model to not only capture the various influencing factors within a single region, but also take into account the interactions and impacts between different regions, which helps to improve the accuracy and applicability of the prediction model and enable it to better adapt to the actual conditions and future development trends of different regions.

[0104] On the basis of the above embodiment, a PLSR-GARCH joint prediction model of multi-regional new energy installed capacity is constructed, and the joint prediction model is trained based on the influencing factor set to obtain a revised prediction model, including the following steps:

[0105] Determine the influencing factor set as the independent variable, and the installed capacity of new energy in each region as the dependent variable. Construct the matrices X and Y corresponding to the independent variable and the dependent variable respectively. After standardization, obtain the independent variable observation matrix A and the dependent variable observation matrix B respectively. Standardize these matrices to eliminate the influence of different dimensions and magnitudes, ensure that the data is on the same scale, and provide a basis for subsequent analysis.

[0106] The first principal components t1 and u1 of the independent variable matrix X and the dependent variable matrix Y are extracted respectively, and the scores of the first principal components are calculated according to the independent variable observation matrix A and the dependent variable observation matrix B respectively. The residual matrix and regression equation corresponding to the first principal component are established using the PLSR model; the component that best represents the variability of the data is extracted by dimensionality reduction, and the relationship between the independent variable and the dependent variable is maintained.

[0107] If the difference between the absolute value of the element in the residual matrix and 0 is within the threshold range, it is determined that the regression equation corresponding to the first principal component meets the accuracy requirement, and the extraction of the next set of principal components is stopped. Otherwise, the above steps are repeated until the final regression equation is obtained. This process is carried out in an iterative manner to gradually extract the principal component that contributes the most to the data variability until the regression equation reaches a satisfactory accuracy.

[0108] Substitute the independent variable matrix into the regression equation to obtain the predicted value And calculate the residual sequence; it helps to understand the prediction error of the model;

[0109] Using the residual sequence ε as input, a GARCH model is established, and the variance formula corresponding to the model is determined; the GARCH model can capture the volatility of data, which is particularly important for predicting the volatility of installed capacity of new energy;

[0110] The trained GARCH model is used to predict the standard deviation predicted value, and the predicted standard deviation is added as the independent variable X8 into the independent variable matrix. The PLSR model is updated to obtain the PLSR model corrected by the GARCH model as the final corrected prediction model. By introducing the volatility-related information predicted by the GARCH model, the PLSR model's ability to handle the volatility of new energy installed capacity is enhanced, and the final corrected prediction model is obtained.

[0111] Specifically, the independent variable matrix X and the dependent variable matrix Y are expressed as follows:

[0112]

[0113] The independent variable observation matrix A and the dependent variable observation matrix B are expressed as follows:

[0114]

[0115]

[0116] in, and are the means of the independent variable matrix X and the dependent variable matrix Y, respectively, S X and S Y are the standard deviations of the independent variable matrix X and the dependent variable matrix Y respectively;

[0117] The first principal components t1 and u1 of the independent variable matrix X and the dependent variable matrix Y are the linear combinations of the independent variable matrix X and the dependent variable matrix Y, respectively, and are expressed by the following formulas:

[0118] t1=w 11 x1+w 12 x2+…+w 1m x m

[0119] u1=c 11 y1+c 12 y2+…+c 1p y p

[0120] The scores of the first principal components t1 and u1 can be calculated from the corresponding independent variable observation matrix A and dependent variable observation matrix B, as expressed by the following formula:

[0121]

[0122] Among them, ω1 is the weight coefficient of the first principal component t1 of the independent variable matrix X, and c1 is the weight coefficient of the first principal component u1 of the dependent variable matrix Y;

[0123] The PLSR method combines principal component analysis and linear regression, which requires maximizing the correlation between t1 and u1 and maximizing the variance of t1 and u1, as expressed by the following formula:

[0124]

[0125] The problem becomes to find the unit vectors w1 and c1 so that Maximize, and use the Lagrange multiplier method to solve, and we get

[0126]

[0127] Establish y1,y2,…,y p Regression on t1, and x1,x2,…,x mFor the regression of t1, the regression equation corresponding to the first principal component is expressed as follows:

[0128]

[0129] Among them, α1=[α 11 ,α 12 ,…,α 1m ] T , β1=[β 11 ,β 12 ,…,β 1p ] T Represent the parameter vectors in the independent variable matrix A and the dependent variable matrix regression model, E l and F l They represent the residual matrices respectively;

[0130] The least squares estimates of the regression parameter vectors α1 and β1 are:

[0131]

[0132] If the absolute value of the elements in the residual matrix is ​​close to 0, that is, the difference between the absolute value of the elements in the residual matrix and 0 is within the threshold range, it can be considered that the regression equation corresponding to the extracted first principal component meets the accuracy requirement, so the extraction of the next set of principal components can be stopped, otherwise, return to the previous step to extract the next set of principal components again to obtain the corresponding regression equation;

[0133] Replace A and B with E1 and F1 respectively, and continue the above iterative regression steps.

[0134] Let r be the rank of the matrix X,Y, then there are r components t1,t2,t3,…,t r .

[0135]

[0136] t k =w k1 x1+w k2 x2+…+w km x m (k=1,2,…,r), substitute into Y=t1β1 T +t2β2 T +…+t r β r T That is, the partial least squares regression equation of the dependent variable matrix Y and the independent variable matrix X is obtained, which is expressed as follows:

[0137] y j =h j1 x1+h j2 x2+…+h jmx m , j=1,2,…,p

[0138] Among them, h is the weight parameter in the PLSR regression model;

[0139] Substitute the dependent variable X into the regression equation to get the predicted value The residual sequence ε is calculated by the following formula:

[0140]

[0141] Using the residual sequence ε as input, a GARCH model is established to capture the volatility characteristics in the residual. It is assumed that the residual sequence ε follows a normal distribution with a mean of zero and a variance of ε. 2 , then the GARCH (p, q) model is used to describe the conditional heteroskedasticity characteristics. The variance formula corresponding to the GARCH model is expressed as follows:

[0142]

[0143] Where: α0 is a constant term, α i is the weight of the past residual square term, β j is the weight of the past conditional variance term, m is the order of the lagged residual square term, and n is the order of the lagged variance term;

[0144] Use MLE, the maximum likelihood estimation method, to estimate the model parameters α0, α i , β j Estimation is performed to optimize the description of the conditional variance, and the model parameters are adjusted during the fitting process so that the volatility of the predicted residual sequence best matches the characteristics of the actual observed data;

[0145] The trained GARCH model is used to predict the standard deviation, and the predicted standard deviation is added as the independent variable X8 into the independent variable matrix. The PLSR model is updated to obtain the PLSR model corrected by the GARCH model as the final corrected prediction model.

[0146] On the basis of the above embodiment, based on different new energy development scenarios, relevant multi-scenario influencing factor data sets are determined respectively, including the following steps:

[0147] Referring to historical economic development statistics, considering different GDP development levels and electricity demand growth scenarios, the annual growth rates of GDP and electricity demand under three economic development scenarios of low, medium and high are set; by setting different economic growth rates, the impact of economic growth rate on the demand for new energy installed capacity can be simulated, thereby providing diversified forecast scenarios for new energy development. The data source can be obtained through China's economic development research reports such as "Revitalizing Growth: China's Economic Development Outlook 2023-2027";

[0148] Refer to the historical data of renewable energy grid-connected electricity price subsidies, consider different scenarios of renewable energy subsidies and carbon market transaction volume growth, and set the annual growth rates of renewable energy subsidies and carbon market transaction volume under three policy support scenarios: low, medium, and high. Policy support is a key factor affecting the development of renewable energy. By setting different policy support levels, the potential impact of policy changes on renewable energy installed capacity can be evaluated. The data source can be obtained through new energy policy research reports such as "China's Carbon Market Construction Results and Prospects (2024)";

[0149] Referring to the historical development trend of the UHV industry, considering the UHV line growth scenario, the UHV cumulative line length and the annual growth rate of the UHV cumulative transformer capacity under the three transmission capacity development scenarios of low, medium and high are set; the improvement of transmission capacity is crucial for the long-distance transmission of new energy electricity. By setting different transmission capacity growth scenarios, the impact of power grid construction on the installed capacity of new energy can be evaluated; the data source can be obtained from the UHV development research reports such as "Analysis of the Development Status and Trends of China's UHV Industry in 2023";

[0150] According to the annual growth rates of GDP and electricity demand under the three economic development scenarios of low, medium and high, the annual growth rates of new energy subsidies and carbon market trading volume under the three policy support scenarios of low, medium and high, and the annual growth rates of UHV cumulative line length and UHV cumulative transformer capacity under the three transmission capacity development scenarios of low, medium and high, a unified normalization process is performed to obtain a multi-scenario influencing factor data set; the normalization process ensures the comparability between different indicators, so that the model can fairly evaluate the impact of various factors on the installed capacity of new energy. It helps to improve the accuracy and comprehensiveness of the prediction model.

[0151] Based on the above embodiment, the modified prediction model is used to predict the installed capacity of new energy under multiple scenario influencing factor data sets, including the following steps:

[0152] According to the obtained revised forecast model, the standard deviation value of the installed capacity of new energy is obtained; the standard deviation value reflects the uncertainty or volatility of the forecast of the installed capacity of new energy. By obtaining this indicator, the potential risks and credibility under different forecast scenarios can be evaluated;

[0153] According to the standard deviation value and the influencing factor data selected from the multi-scenario influencing factor data set in sequence, the modified prediction model is used to predict the new energy installed capacity in the scenarios corresponding to the influencing factor data in different regions, and the predicted value of new energy installed capacity is obtained; it ensures that the prediction model can take into account the influencing factors under different economic development, policy support and transmission capacity development scenarios, thereby providing diversified data support for the prediction of new energy installed capacity, and more accurately predicting the new energy installed capacity while considering the specific influencing factors of different regions. It can provide regionalized prediction results and provide a basis for regional new energy planning and decision-making.

[0154] Specifically, in this embodiment, the northwest region is taken as the prediction area for installed capacity of new energy, and the installed capacity of the new energy region is predicted. First, the historical data related to the installed capacity of new energy in each region is collected, and the main influencing factors of the development of the installed capacity of new energy in the northwest region are analyzed. Due to the inherent intermittent and volatile characteristics of new energy power generation, its development is not only affected by policy factors, but also involves multiple factors such as economy, technology, environment and society. From the current situation, the natural resources of northwest wind and light are excellent, and will not become a bottleneck affecting the installed capacity of new energy in the northwest region. The main factors affecting the development of new energy installed capacity in the northwest region can be summarized as shown in Table 1:

[0155] Table 1: Influencing factors

[0156]

[0157]

[0158] The PLSR-GARCH model is constructed using historical electricity data from the China Electricity Industry Committee, economic data from the China Statistical Yearbook and related public reports. The performance of the model is tested by analyzing and selecting the factors affecting the installed capacity of new energy in Northwest China, combined with the historical data series of installed capacity of renewable energy in Northwest China. For photovoltaic installations, data from 2011-2020 are used as the training set, and data from 2021-2023 are used as the validation set. Wind power generation installations are divided into two groups for prediction. Qinghai and Shaanxi use data from 2011-2020 as the training set and data from 2021-2022 as the validation set, respectively. For Ningxia, Xinjiang, and Gansu, data from 2006-2019 are used as the training set, and data from 2021-2022 are used as the validation set.

[0159] like Figure 5 and Figure 6 As shown, the test set exhibits a higher error level than the training set. However, the relative error to the actual value remains within 20%, which is within an acceptable range. This shows that the accuracy of new energy installed capacity prediction using the PLSR-GARCH model is within a reasonable range.

[0160] According to national and relevant industry plans, as well as industry reports released by research institutions, three development scenarios have been formulated according to different development speeds: low-speed, medium-speed, and high-speed development scenarios. The specific situation is shown in the following table:

[0161] Table 2: Multi-scenario influencing factor dataset

[0162]

[0163] According to the scenario settings and the historical data of new energy installed capacity in Northwest my country from 2006 to 2023, the PLSR-GARCH joint forecasting model was used as a tool to predict the new energy installed capacity under the low-speed development scenario, medium-speed development scenario and high-speed development scenario.

[0164] Through Figures 7 to 13 As shown in the forecast results, under the scenario of low-speed development, during the period 2024-2030, the growth trend of new energy installed capacity in Northwest China is obvious. By 2030, the installed capacity of new energy in Northwest China will reach 446.28 million kilowatts, with an average annual increase of 279.9589427 million kilowatts; under the scenario of medium-speed development, during the period 2024-2030, the growth trend of new energy installed capacity in Northwest China is obvious. By 2030, the installed capacity of new energy in Northwest China will reach 480.85 million kilowatts, with an average annual increase of 32.31 million kilowatts; under the scenario of high-speed development, during the period 2024-2030, the growth trend of new energy installed capacity in Northwest China will be more obvious. By 2030, the installed capacity of new energy in Northwest China will reach 518.55 million kilowatts, with an average annual increase of 37.03 million kilowatts.

[0165] like Fig.14 As shown, the present invention also provides a multi-regional new energy installed capacity joint prediction system considering regional coupling factors, comprising:

[0166] Data collection module, used to collect and organize relevant historical data of renewable energy installed capacity in various regions and form a relevant database;

[0167] Data analysis module, used to analyze relevant historical data and obtain data related to factors affecting the development of new energy;

[0168] A data processing module is used to preprocess the data related to the factors affecting the development of new energy, and construct a set of factors affecting the installed capacity of new energy that takes into account regional coupling factors based on the preprocessed data;

[0169] The prediction model module is used to build a joint prediction model of PLSR-GARCH multi-regional new energy installed capacity, and train the joint prediction model based on the influencing factor set to obtain a revised prediction model;

[0170] The scenario setting module is used to determine the relevant multi-scenario influencing factor data sets based on different new energy development scenarios;

[0171] The capacity prediction module is used to use the modified prediction model to predict the installed capacity of new energy under multiple scenario influencing factor data sets.

[0172] The present invention obtains key influencing factors through in-depth analysis of relevant historical data of each region, so that the prediction model results can comprehensively and accurately cover different development situations in multiple scenarios; at the same time, regional coupling factors are considered to form a joint prediction of multiple regions, which is conducive to achieving cross-regional coordinated development and improving the accuracy of new energy installed capacity prediction results. The constructed PLSR-GARCH joint prediction model can effectively solve the problems of multicollinearity and time series volatility in the prediction process, and has better adaptability and flexibility.

[0173] The specific working process and technical effects of the above system are the same as those in the above embodiment and will not be described in detail here.

[0174] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the above-mentioned method for joint prediction of multi-regional new energy installed capacity taking into account regional coupling factors are implemented, including collecting and collating relevant historical data of new energy installed capacity in each region and forming a relevant database, and analyzing the relevant historical data to obtain relevant data on factors affecting new energy development; performing data preprocessing on the relevant data on factors affecting new energy development, and constructing a set of factors affecting new energy installed capacity taking into account regional coupling factors for the preprocessed data; constructing a PLSR-GARCH joint prediction model for new energy installed capacity in multiple regions, and training the joint prediction model based on the set of factors affecting new energy, to obtain a revised prediction model; based on different new energy development scenarios, respectively determining relevant multi-scenario influencing factor data sets; and using the revised prediction model to predict the new energy installed capacity under the multi-scenario influencing factor data sets.

[0175] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for jointly predicting multi-regional new energy installed capacity considering regional coupling factors are implemented, including collecting and collating relevant historical data of new energy installed capacity in each region and forming a relevant database, and analyzing the relevant historical data to obtain relevant data on factors affecting new energy development; performing data preprocessing on the relevant data on factors affecting new energy development, and constructing a set of factors affecting new energy installed capacity considering regional coupling factors for the preprocessed data; constructing a PLSR-GARCH joint prediction model for new energy installed capacity in multiple regions, and training the joint prediction model based on the set of factors affecting new energy to obtain a revised prediction model; based on different new energy development scenarios, respectively determining relevant multi-scenario influencing factor data sets; and using the revised prediction model to predict the new energy installed capacity under the multi-scenario influencing factor data sets.

[0176] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0177] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0178] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0179] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0180] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0181] Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for joint prediction of multi-regional renewable energy installed capacity considering regional coupling factors, characterized in that: The following steps are involved: Collect and organize relevant historical data on the installed capacity of new energy in various regions and form a relevant database, and analyze the relevant historical data to obtain relevant data on factors affecting the development of new energy; Preprocessing the data related to the factors affecting the development of new energy, and constructing a set of factors affecting the installed capacity of new energy that considers regional coupling factors based on the preprocessed data; Constructing a PLSR-GARCH joint prediction model for renewable energy installed capacity in multiple regions, and training the joint prediction model based on the influencing factor set to obtain a revised prediction model; Based on different new energy development scenarios, determine the relevant multi-scenario influencing factor data sets; The modified prediction model is used to predict the installed capacity of new energy under multiple scenario influencing factor data sets.

2. The method for joint prediction of multi-regional new energy installed capacity considering regional coupling factors according to claim 1 is characterized in that: The data preprocessing of the data related to the factors affecting the development of new energy includes: Retrieving the natural resource data of each region in the relevant historical data, and converting the natural resource data of the same type into a unified format; Obtain the thermal power grid-connected electricity price and renewable energy grid-connected electricity price in each region, and determine the renewable energy grid-connected electricity price data for each region based on the difference between the thermal power grid-connected electricity price and renewable energy grid-connected electricity price; All relevant data are scaled to the range of [0,1] according to the proportional relationship and normalized uniformly; The missing year data for each region are supplemented.

3. The method for joint prediction of multi-regional renewable energy installed capacity considering regional coupling factors according to claim 1, characterized in that: The influencing factor set includes intra-regional influencing factors and inter-regional coupling factors; The influencing factors within the region include: social electricity consumption within the region Regional GDP Development Level Cumulative length of UHV lines UHV cumulative transformer capacity New energy subsidy amount Carbon market trading volume Inter-regional coupling factors include the overall proportion of new energy installed capacity in previous years It is expressed by the following formula: Where i = 1, 2, 3, ..., n, i represents the year, n is the total number of years; j = 1, 2, 3, ..., p, j represents the region number, p is the total number of regions; y i-1,j It represents the installed capacity of new energy in region j in year i-1.

4. The method for joint prediction of multi-regional new energy installed capacity considering regional coupling factors according to claim 3 is characterized in that: The constructing of the PLSR-GARCH joint prediction model of multi-regional new energy installed capacity, and training the joint prediction model based on the influencing factor set to obtain a revised prediction model, includes the following steps: Determine the influencing factor set as the independent variable, the installed capacity of new energy in each region as the dependent variable, construct matrices X and Y corresponding to the independent variable and the dependent variable respectively, and obtain the independent variable observation matrix A and the dependent variable observation matrix B respectively after standardization; Extract the first principal components t1 and u1 of the independent variable matrix X and the dependent variable matrix Y respectively, calculate the scores of the first principal components according to the independent variable observation matrix A and the dependent variable observation matrix B respectively, and use the PLSR model to establish the residual matrix and regression equation corresponding to the first principal component; If the difference between the absolute value of the element in the residual matrix and 0 is within the threshold range, it is determined that the regression equation corresponding to the first principal component meets the accuracy requirement, and the extraction of the next set of principal components is stopped. Otherwise, the above steps are repeated until the final regression equation is obtained; Substitute the independent variable matrix into the regression equation to obtain the predicted value And calculate the residual sequence; Use the residual sequence ε as input to build a GARCH model and determine the variance formula corresponding to the model; The trained GARCH model is used to predict the standard deviation, and the predicted standard deviation is added as the independent variable X8 into the independent variable matrix. The PLSR model is updated to obtain the PLSR model corrected by the GARCH model as the final corrected prediction model.

5. The method for joint prediction of multi-regional new energy installed capacity considering regional coupling factors according to claim 4 is characterized in that: The independent variable matrix X and the dependent variable matrix Y are expressed as follows: The independent variable observation matrix A and the dependent variable observation matrix B are expressed by the following formula: in, and are the means of the independent variable matrix X and the dependent variable matrix Y, respectively, S X and S Y are the standard deviations of the independent variable matrix X and the dependent variable matrix Y respectively; The first principal components t1 and u1 of the independent variable matrix X and the dependent variable matrix Y are expressed as follows: t1=w 11 x1+w 12 x2+…+w 1m x m u1=c 11 y1+c 12 y2+…+c 1p yes p The scores of the first principal components t1 and u1 are expressed as follows: Among them, ω1 is the weight coefficient of the first principal component t1 of the independent variable matrix X, and c1 is the weight coefficient of the first principal component u1 of the dependent variable matrix Y; The regression equations corresponding to the first principal component are expressed as follows: Among them, α1=[α 11 ,α 12 ,…,α 1m ] T , β1=[β 11 ,β 12 ,…,β 1p ] T Represent the parameter vectors in the independent variable matrix A and the dependent variable matrix regression model, E l and F l They represent the residual matrices respectively; The residual sequence ε is calculated by the following formula: The variance formula corresponding to the GARCH model is expressed as follows: Where: α0 is a constant term, α i is the weight of the past residual square term, β j is the weight of the past conditional variance term, m is the order of the lagged residual square term, and n is the order of the lagged variance term.

6. The method for joint prediction of multi-regional new energy installed capacity considering regional coupling factors according to claim 1, characterized in that: The method of determining relevant multi-scenario influencing factor data sets based on different new energy development scenarios includes the following steps: Referring to historical economic development statistics, considering different GDP development levels and electricity demand growth scenarios, we set the annual growth rates of GDP and electricity demand under three economic development scenarios: low, medium and high. Refer to the historical data of renewable energy grid-connected electricity price subsidies, consider different scenarios of renewable energy subsidies and carbon market trading volume growth, and set the annual growth rates of renewable energy subsidies and carbon market trading volume under three policy support scenarios: low, medium and high. Referring to the historical development trend of the UHV industry and considering the UHV line growth scenario, the annual growth rate of the UHV cumulative line length and UHV cumulative transformer capacity under three transmission capacity development scenarios of low, medium and high are set; Based on the annual growth rates of GDP and electricity demand under the three set economic development scenarios of low, medium and high, the annual growth rates of new energy subsidies and carbon market trading volume under the three policy support scenarios of low, medium and high, and the annual growth rates of UHV cumulative line length and UHV cumulative transformer capacity under the three transmission capacity development scenarios of low, medium and high, a unified normalization processing was performed to obtain a multi-scenario influencing factor data set.

7. The method for joint prediction of multi-regional new energy installed capacity considering regional coupling factors according to claim 6, characterized in that: The method of using the modified prediction model to predict the installed capacity of new energy under multiple scenario influencing factor data sets includes the following steps: According to the obtained revised prediction model, the standard deviation value of the installed capacity of new energy is obtained; According to the standard deviation value and the influencing factor data selected from the multi-scenario influencing factor data set in sequence, a modified prediction model is used to predict the new energy installed capacity in the scenarios corresponding to the influencing factor data in different regions to obtain the predicted value of the new energy installed capacity.

8. A multi-regional new energy installed capacity joint prediction system considering regional coupling factors, characterized in that: include: Data collection module, used to collect and organize relevant historical data of renewable energy installed capacity in various regions and form a relevant database; A data analysis module, used to analyze the relevant historical data to obtain data related to factors affecting the development of new energy; A data processing module, used to perform data preprocessing on the data related to the influencing factors of the new energy development, and construct an influencing factor set of the new energy installed capacity considering the regional coupling factor based on the preprocessed data; A prediction model module is used to construct a PLSR-GARCH joint prediction model for new energy installed capacity in multiple regions, and to train the joint prediction model based on the influencing factor set to obtain a revised prediction model; The scenario setting module is used to determine the relevant multi-scenario influencing factor data sets based on different new energy development scenarios; The capacity prediction module is used to use the modified prediction model to predict the installed capacity of new energy under multiple scenario influencing factor data sets.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for joint prediction of multi-regional new energy installed capacity taking into account regional coupling factors as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the multi-regional new energy installed capacity joint prediction method considering regional coupling factors as described in any one of claims 1 to 7.

Images