CN114638405A - Energy demand forecasting method, device and computer equipment - Google Patents

Abstract

The invention relates to an energy demand prediction method, an energy demand prediction device and computer equipment. The invention provides an energy demand prediction model based on principal component analysis and multiple linear regression analysis, and compared with a method for predicting future energy demand according to historical energy demand change narrative, the model provided by the invention is more in line with energy demand prediction influenced by actual policies and economy.

Description

Energy demand forecasting method, device and computer equipment

technical field

The invention belongs to the technical field of electricity market, and in particular relates to a method, device and computer equipment for predicting energy demand.

Background technique

With the development of the resource industry, the proportion of new energy in energy consumption will continue to increase, and energy demand will change from single to diversified. Accurate and effective energy demand forecast can better study and judge the energy demand structure and energy demand level at each stage, and effectively manage the use and development of energy. At present, there is a lack of accurate and effective energy demand forecasting methods, which makes it impossible to make targeted adjustments and layouts to the existing energy development routes. There are many influencing factors and complex models in energy demand forecasting.

Among the related technologies, energy forecasting methods mainly include energy demand forecasting based on BP neural network model, autoregressive moving average model (ARIMA) and grey forecasting model. Affected by a variety of factors, economic growth is one of the important factors affecting the growth of China's energy consumption; in addition, the industrial structure is also an important factor affecting my country's energy demand, and the secondary industry is the main driving force for China's energy consumption growth; while the population The quantity and structure will also directly affect the total energy consumption. The urbanization process has a significant impact on energy consumption, and there is a positive correlation between the urbanization rate and energy consumption. Therefore, when forecasting future energy demand, not only the historical development trend of energy consumption itself, but also the coupling relationship between factors such as economic growth, industrial structure transformation and social development, and energy demand should be considered.

The existing energy demand forecasting method based on ARIMA model requires the given time series data to be stable, otherwise it cannot capture the changing law of energy demand. In addition, the autoregressive model describes the relationship between the current value and the historical value, that is, a prediction model based on a time perspective, which cannot reflect the relationship between my country's energy demand and economic, policy and other factors. In addition, the gray forecasting model does not use the original data series, but the generated data series, which makes it more suitable for medium and short-term forecasting and forecasting of exponential growth, and is not suitable for forecasting energy demand with a large time span under the dual-carbon target. Moreover, the existing energy demand forecast cannot take into account future environmental policies and economic development plans.

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to overcome the deficiencies of the prior art, and to provide an energy demand forecasting method, device and computer equipment, so as to solve the problem that the existing energy demand forecasting in the prior art cannot be suitable for forecasting under the dual carbon target. The problem of energy demand over a large time span.

To achieve the above purpose, the present invention adopts the following technical solutions: a method for predicting energy demand, comprising:

Obtain original independent variable data, establish an original independent variable matrix according to the original independent variable data and perform standardization processing to obtain a standardized matrix; wherein, the original independent variable data includes multiple categories of indicators related to the power industry and economic development, so The above indicators include industrial added value, output and price index;

According to the standardized matrix, a correlation coefficient matrix is obtained and the eigenvalues of the correlation coefficient matrix are obtained, corresponding eigenvectors are calculated according to the eigenvalues, and an independent variable matrix is constructed according to the eigenvectors;

Determine the contribution degree of the synthesized principal components according to the eigenvalues, screen the independent variable matrix according to the contribution degree, determine a preset number of principal components, and construct a prediction based on the preset number of principal components matrix;

Obtaining dependent variable data, and establishing a dependent variable matrix according to the dependent variable data; wherein, the dependent variable data includes historical energy demand data;

An energy demand prediction model is constructed according to the prediction matrix and the dependent variable matrix, and the energy demand prediction model is used to calculate future energy demand.

Further, establishing an original independent variable matrix according to the original independent variable data and performing standardization processing to obtain a standardized matrix, including:

Assuming that the actual result of the j-th indicator of a certain industry in the historical period n is X _j =[x _1j ,x _2j ,...,x _nj ] ^T , the original independent variable matrix of the macro economy of the industry in the historical period n is:

Standardize the original independent variable matrix to obtain the standardized data x _ij in the i-th row and the j-th column;

表示历史时段内针对该行业的第j项指标的平均值，即

var(x_ij)表示第j类指标的标准差，即

(j＝1,2,…,k)。in,

Represents the average value of the jth indicator for the industry in the historical period, namely

var(x _ij ) represents the standard deviation of the jth index, namely

(j=1,2,...,k).

Further, obtain the eigenvalues of the correlation coefficient matrix, and calculate the corresponding eigenvectors according to the eigenvalues, including:

The correlation coefficient matrix is:

Calculate the eigenvalues λ ₁ , λ ₂ , . . . , λ _k of the correlation coefficient matrix according to the correlation coefficient matrix;

According to the eigenvalues, the eigenvectors corresponding to the eigenvalues are calculated in the following manner;

A ^T RA=diag(λ ₁ ,λ ₂ ,...,λ _k )

Among them, X ^T is the inverted matrix of the original independent variable matrix X, A=(a _ij ) _k×k represents the canonical orthogonal eigenvector corresponding to the eigenvalue, and a _ij represents the element in the ith column and the jth row of the eigenvector A.

Further, the constructing an independent variable matrix according to the eigenvectors includes:

According to the original independent variable matrix and eigenvectors, the independent variable matrix is calculated as follows;

Y=A ^T X

Further, determining the contribution degree of the synthetic principal component according to the eigenvalue, screening the independent variable matrix according to the contribution degree, and determining a preset number of principal components, according to the preset number of principal components. Each principal component constructs a prediction matrix, including:

并根据预设范围进行取值，以筛选各主成分，得到的k项经济运行指标合成的主成分为：The eigenvalues λ ₁ , λ ₂ , ..., λ _k represent the contribution rate of the synthetic principal components, and the eigenvectors of each principal component are arranged in descending order. If it is assumed that the cumulative value of the first p eigenvalues is close to 1, where p<k, Indicates that the above p principal components have a high degree of information retention to the original index set, then the cumulative contribution of the principal components is defined as

And select the value according to the preset range to filter the principal components, and the obtained principal components of the k-item economic operation indicators are as follows:

Further, for each principal component after synthesis, the covariance calculation is performed to obtain:

The variance of the principal component y _j is equal to its corresponding eigenvalue λ _i , and the covariance between any two different principal components is 0, that is, there is no correlation between the principal components.

Further, the energy demand forecasting model is a multiple linear regression analysis model, including:

Z=β ₀ +β ₁ Y′ ₁ +β ₂ Y′ ₂ +…+β _k Y′ _k (8)

Among them, it is assumed that there is a linear relationship between the explained variable Z and multiple explanatory variables Y′, Y′ ₂ ,…Y′ _k , which are multivariate linear functions of the explanatory variables, β ₀ , β ₁ , …, β _k are regression coefficients , the explanatory variable Z is the energy demand, and the explanatory variable Y is the prediction matrix formed by the processing of economic development planning, environmental policy objectives, industrial structure and other factors.

Further, the preset range is 85% to 95%.

The embodiment of the present application provides an energy demand forecasting device, including:

The acquisition module is used for acquiring original independent variable data, establishing an original independent variable matrix according to the original independent variable data and performing standardization processing to obtain a standardized matrix; wherein, the original independent variable data includes a plurality of power industry and economic development related categories of indicators including industrial value added, production and price indices;

a construction module, configured to obtain a correlation coefficient matrix and obtain eigenvalues of the correlation coefficient matrix according to the standardized matrix, calculate corresponding eigenvectors according to the eigenvalues, and construct an independent variable matrix according to the eigenvectors;

The determining module is configured to determine the contribution degree of the synthetic principal component according to the eigenvalue, screen the independent variable matrix according to the contribution degree, and determine a preset number of each principal component, according to the preset number of Each principal component constructs a prediction matrix;

a building module for acquiring dependent variable data, and establishing a dependent variable matrix according to the dependent variable data; wherein the dependent variable data includes historical energy demand data;

a building module for building an energy demand forecasting model according to the forecasting matrix and the dependent variable matrix.

An embodiment of the present application provides a computer device, including: a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform any one of the foregoing energy demand predictions steps of the method.

Embodiments of the present application further provide a computer storage medium, which stores a computer program, and when the computer program is executed by a processor, causes the processor to execute any of the steps of the above-mentioned energy demand forecasting method.

The present invention adopts the above technical solutions, and the beneficial effects that can be achieved include:

The present invention provides an energy demand forecasting method, device and computer equipment. The present invention takes into account the economic growth that affects energy demand for forecasting, and also considers that the relevant indicators of economic development, such as output value, price index, etc., have strong correlation. , which can lead to "pseudo-regression" problems in multiple linear regression models. In this regard, this method uses the principal component analysis method to process the independent variable matrix, and obtains the factors in the new independent variable matrix that are not correlated with each other, which avoids the "pseudo regression" caused by excessive variable autocorrelation in the prediction of the multiple regression model. question.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a schematic diagram of the steps of the energy demand forecasting method of the present invention;

Fig. 2 is the schematic flow chart of the energy demand forecasting method of the present invention;

3 is a schematic structural diagram of an energy demand forecasting device of the present invention;

FIG. 4 is a schematic diagram of the hardware structure of the specific implementation of the energy demand forecasting method of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other implementations obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

A specific energy demand forecasting method, apparatus, and computer equipment provided in the embodiments of the present application will be described below with reference to the accompanying drawings.

As shown in FIG. 1 , the energy demand forecasting method provided in the embodiment of the present application includes:

S101: Obtain original independent variable data, establish an original independent variable matrix according to the original independent variable data, and perform standardization processing to obtain a standardized matrix; wherein, the original independent variable data includes multiple categories of indicators related to the power industry and economic development , the indicators include industrial value added, output and price indices;

As shown in FIG. 2 , the original independent variable data to be acquired in the present invention includes indicators related to national development of the industry or region, such as industrial added value, output, and price index, and the dependent variable data includes historical energy demand data. It can be understood that the original independent variable data may also include data of other dimensions, and the dependent variable data may also include data of other dimensions, which is not limited in this application.

After obtaining the original independent variable data, establish the original independent variable matrix X, as follows:

Assuming that the actual result of the j-th related indicators of a certain industry in the historical period n is X _j =[x _1j ,x _2j ,...,x _nj ] ^T , the original macroeconomic matrix of the industry in the historical period n can be expressed as:

Then, standardize the original independent variable matrix to obtain the standardized data x _ij in the i-th row and the j-th column; as follows:

It should be noted that due to the differences in the nature, dimension, order of magnitude, etc. of each index, it needs to be standardized. In this application, the standardization of the original independent variable matrix can be specifically processed by Z-score method and standard deviation. and other methods.

表示历史时段内针对该行业的第j项指标的平均值，即

var(x_ij)表示第j类指标的标准差，即

(j＝1,2,…,k)。In the formula,

Represents the average value of the jth indicator for the industry in the historical period, namely

var(x _ij ) represents the standard deviation of the jth index, namely

(j=1,2,...,k).

S102, according to the standardized matrix, obtain a correlation coefficient matrix and obtain eigenvalues of the correlation coefficient matrix, calculate corresponding eigenvectors according to the eigenvalues, and construct an independent variable matrix according to the eigenvectors;

Specifically, each macroeconomic index j after standardized processing has the conditions for homogenization processing. For various macroeconomic indicators within the time span n, the correlation coefficient matrix can be obtained:

Then, the eigenvalues λ ₁ , λ ₂ , . . . , λ _k of the correlation coefficient matrix are obtained by calculating according to the correlation coefficient matrix. The eigenvalues λ ₁ , λ ₂ , ..., λ _k have corresponding eigenvectors respectively, and the calculation methods are as follows:

A ^T RA=diag(λ ₁ ,λ ₂ ,...,λ _k ) (5)

Among them, X ^T is the inverted matrix of the original independent variable matrix X, A=(a _ij ) _k×k represents the canonical orthogonal eigenvector corresponding to the eigenvalue, and a _ij represents the element in the ith column and the jth row of the eigenvector A.

S103: Determine the contribution degree of the synthetic principal component according to the eigenvalue, screen the independent variable matrix according to the contribution degree, determine each principal component of a preset number, and determine each principal component according to the preset number of principal components Build a prediction matrix;

Different from the original independent variable matrix X, the independent variable matrix will be constructed according to the results in the historical period n of the industry: Y=A ^T X. get

Then, the independent variable matrix Y is screened to obtain the matrix Y' used to predict the dependent variable. The specific process is:

可根据实际情况取值(一般为85％至95％之间)，由此筛选各主成分。得到的k项经济运行指标合成的主成分为：The eigenvalues λ ₁ , λ ₂ , ..., λ _k express the contribution rate of the synthetic principal components, and the eigenvectors of each principal component are arranged in descending order. If it is assumed that the cumulative value of the first p (p<k) eigenvalues is very close to 1 , indicating that the above p principal components have a very high degree of information retention to the original index set, which is defined as the cumulative contribution of the principal components

The value can be selected according to the actual situation (usually between 85% and 95%), so as to screen each principal component. The principal components of the obtained k-item economic operation index synthesis are:

The independent variable Y matrices obtained after the principal component analysis method are not correlated with each other, so there will be no problem of pseudo-regression;

From the mathematical analysis, for each principal component after synthesis, the following conclusions can be obtained after covariance calculation:

The variance of the principal component _y'j is equal to its corresponding eigenvalue λ _i , and the covariance between any two different principal components is 0 (ie, irrelevant).

S104, obtaining dependent variable data, and establishing a dependent variable matrix according to the dependent variable data; wherein the dependent variable data includes historical energy demand data;

The dependent variable data includes historical energy demand data.

S105. Build an energy demand prediction model according to the prediction matrix and the dependent variable matrix, where the energy demand prediction model is used to calculate future energy demand.

Using the above obtained prediction matrix and the dependent variable matrix formed by the obtained dependent variable data, the multiple linear regression analysis model is established with the above independent variable matrix Y and dependent variable matrix Z, that is, the historical value of energy demand, and the energy-economic prediction model is obtained.

The multiple linear regression model is usually used to study the relationship between a dependent variable and multiple independent variables. The model is shown in the following formula:

Z=β ₀ +β ₁ Y′ ₁ +β ₂ Y′ ₂ +…+β _k Y′ _k (9)

Among them, it is assumed that there is a linear relationship between the explained variable Z and multiple explanatory variables Y′, Y′ ₂ ,…Y′ _k , which are multivariate linear functions of the explanatory variables, β ₀ , β ₁ , …, β _k are regression coefficients , where the explanatory variable Z is the energy demand, and the explanatory variable Y is the independent variable matrix formed by processing factors such as economic development planning, environmental policy goals, and industrial structure.

The present application can set the future independent variable matrix X, calculate the corresponding prediction matrix Y', and calculate the future energy demand matrix Z based on the established multiple linear regression model.

As shown in Figure 3, the present application provides an energy demand forecasting device, including:

The acquisition module 201 is configured to acquire original independent variable data, establish an original independent variable matrix according to the original independent variable data and perform standardization processing to obtain a standardized matrix; wherein, the original independent variable data includes multiple data related to the power industry and economic development. categories of indicators including industrial value added, production and price indices;

The construction module 202 is configured to obtain a correlation coefficient matrix and obtain eigenvalues of the correlation coefficient matrix according to the standardized matrix, calculate corresponding eigenvectors according to the eigenvalues, and construct an independent variable matrix according to the eigenvectors;

A determination module 203, configured to determine the contribution degree of the synthetic principal component according to the eigenvalue, screen the independent variable matrix according to the contribution degree, and determine a preset number of each principal component, according to the preset number Each principal component of , constructs a prediction matrix;

The establishment module 204 is used for acquiring dependent variable data, and establishing a dependent variable matrix according to the dependent variable data; wherein, the dependent variable data includes historical energy demand data;

The building module 205 is configured to build an energy demand prediction model according to the prediction matrix and the dependent variable matrix.

The working principle of the energy demand forecasting device provided by the present application is that the acquisition module 201 acquires original independent variable data, establishes an original independent variable matrix according to the original independent variable data, and performs standardization processing to obtain a standardized matrix; wherein, the original independent variable The data includes multiple categories of indicators related to the power industry and economic development, and the indicators include industrial added value, output and price index; the construction module 202 obtains a correlation coefficient matrix according to the standardized matrix and acquires the characteristics of the correlation coefficient matrix value, calculate the corresponding eigenvectors according to the eigenvalues, and construct an independent variable matrix according to the eigenvectors; the determining module 203 determines the contribution degree of the synthetic principal component according to the eigenvalues, and determines the contribution degree of the independent variable matrix according to the contribution degree to the independent variable matrix Perform screening, determine each principal component of a preset number, and construct a prediction matrix according to each principal component of the preset number; the establishment module 204 obtains dependent variable data, and establishes a dependent variable matrix according to the dependent variable data; wherein, the The dependent variable data includes historical energy demand data; the building module 205 constructs an energy demand prediction model according to the prediction matrix and the dependent variable matrix.

The present application provides a computer device, including: a memory and a processor, and may also include a network interface, the memory stores a computer program, and the memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and/or forms of non-volatile memory such as read only memory (ROM) or flash memory (flash RAM). The computer device stores an operating system, and the memory is an example of a computer-readable medium. When the computer program is executed by the processor, the processor executes the energy demand forecasting method, and the structure shown in FIG. The definition of a computer device to which it applies, a particular computer device may include more or fewer components than those shown in the figures, or combine certain components, or have a different arrangement of components.

In one embodiment, the energy demand forecasting method provided by the present application can be implemented in the form of a computer program, and the computer program can be executed on a computer device as shown in FIG. 4 .

In some embodiments, when the computer program is executed by the processor, the processor causes the processor to perform the following steps: acquiring original independent variable data, establishing an original independent variable matrix according to the original independent variable data, and performing normalization processing to obtain Standardized matrix; wherein, the original independent variable data includes multiple categories of indicators related to the power industry and economic development, and the indicators include industrial added value, output and price index; according to the standardized matrix, obtain a correlation coefficient matrix and obtain The eigenvalues of the correlation coefficient matrix, the corresponding eigenvectors are calculated according to the eigenvalues, and the independent variable matrix is constructed according to the eigenvectors; the contribution degree of the synthetic principal component is determined according to the eigenvalues, and the The independent variable matrix is screened to determine a preset number of principal components, and a prediction matrix is constructed according to the preset number of principal components; the dependent variable data is obtained, and a dependent variable matrix is established according to the dependent variable data; wherein, The dependent variable data includes historical data of energy demand; an energy demand prediction model is constructed according to the prediction matrix and the dependent variable matrix, and the energy demand prediction model is used to calculate future energy demand.

The present application also provides a computer storage medium. Examples of the computer storage medium include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Flash or other memory technology, Compact Disc Read Only Memory (CD-ROM), Digital Multimedia A functional optical disc (DVD) or other optical storage, magnetic cassette storage or other magnetic storage device or any other non-transmission medium can be used to store information that can be accessed by a computing device.

In some embodiments, the present invention also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, obtains original argument data, and establishes an original argument according to the original argument data. matrix and standardize to obtain a standardized matrix; wherein, the original independent variable data includes multiple categories of indicators related to economic development in the power industry, and the indicators include industrial added value, output and price index; according to the standardized matrix , obtain the correlation coefficient matrix and obtain the eigenvalues of the correlation coefficient matrix, calculate the corresponding eigenvectors according to the eigenvalues, construct the independent variable matrix according to the eigenvectors; determine the contribution of the synthetic principal components according to the eigenvalues, Screen the independent variable matrix according to the contribution degree, determine each principal component of a preset number, and construct a prediction matrix according to each principal component of the preset number; A dependent variable matrix is established; wherein the dependent variable data includes historical energy demand data; an energy demand prediction model is constructed according to the prediction matrix and the dependent variable matrix, and the energy demand prediction model is used to calculate future energy demand.

In summary, the present invention provides an energy demand forecasting method, device and computer equipment, including acquiring original independent variable data, establishing an original independent variable matrix, standardizing the original independent variable matrix, and obtaining the correlation coefficient matrix of the original independent variable matrix. , find the eigenvalues of the correlation coefficient matrix and the eigenvectors corresponding to the eigenvalues, construct the independent variable matrix according to the eigenvectors, screen the independent variable matrix to obtain the prediction matrix for predicting the dependent variable, and combine the prediction matrix and the dependent variable matrix to obtain the energy demand forecasting model. Based on principal component analysis and multiple linear regression analysis, the present invention proposes an energy demand forecasting model. Compared with the method of predicting future energy demand based on historical energy demand change narratives, the model provided by the present application is more in line with actual policies and economic conditions. Impact on energy demand forecasts.

It can be understood that the method embodiments provided above correspond to the above-mentioned apparatus embodiments, and the corresponding specific contents can be referred to each other, which will not be repeated here.

As will be appreciated by those skilled in the art, the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein, including but not limited to disk storage, optical storage, and the like.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

The computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory result in an article of manufacture comprising the method of the instructions, the instructions A method implements the functions specified in the flow diagram or flow diagrams and/or the block diagram diagram block or blocks.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention. should be included within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (10)

1. A method for forecasting energy demand, comprising: Obtain original independent variable data, establish an original independent variable matrix according to the original independent variable data and perform standardization processing to obtain a standardized matrix; wherein, the original independent variable data includes multiple categories of indicators related to the power industry and economic development, so The above indicators include industrial added value, output and price index; According to the standardized matrix, a correlation coefficient matrix is obtained and the eigenvalues of the correlation coefficient matrix are obtained, corresponding eigenvectors are calculated according to the eigenvalues, and an independent variable matrix is constructed according to the eigenvectors; Determine the contribution degree of the synthesized principal components according to the eigenvalues, screen the independent variable matrix according to the contribution degree, determine a preset number of principal components, and construct a prediction based on the preset number of principal components matrix; Obtaining dependent variable data, and establishing a dependent variable matrix according to the dependent variable data; wherein, the dependent variable data includes historical energy demand data; An energy demand prediction model is constructed according to the prediction matrix and the dependent variable matrix, and the energy demand prediction model is used to calculate future energy demand. 2. The method according to claim 1, wherein, establishing an original independent variable matrix according to the original independent variable data and performing standardization processing to obtain a standardized matrix, comprising: Assuming that the actual result of the j-th indicator of a certain industry in the historical period n is X _j =[x _1j ,x _2j ,...,x _nj ] ^T , the original independent variable matrix of the macro economy of the industry in the historical period n is:

Standardize the original independent variable matrix to obtain the standardized data x _ij in the i-th row and the j-th column;

in,

Represents the average value of the jth indicator for the industry in the historical period, namely

var(x _ij ) represents the standard deviation of the jth index, namely

3. The method according to claim 1 or 2, wherein obtaining the eigenvalues of the correlation coefficient matrix, and calculating the corresponding eigenvectors according to the eigenvalues, comprising: The correlation coefficient matrix is:

Calculate the eigenvalues λ ₁ , λ ₂ , . . . , λ _k of the correlation coefficient matrix according to the correlation coefficient matrix; According to the eigenvalues, the eigenvectors corresponding to the eigenvalues are calculated in the following manner; A ^T RA=diag(λ ₁ ,λ ₂ ,...,λ _k ) Among them, X ^T is the inverted matrix of the original independent variable matrix X, A=(a _ij ) _k×k represents the canonical orthogonal eigenvector corresponding to the eigenvalue, and a _ij represents the element in the ith column and the jth row of the eigenvector A. 4. The method according to claim 1, wherein the constructing an independent variable matrix according to the eigenvectors comprises: According to the original independent variable matrix and eigenvectors, the independent variable matrix is calculated as follows; Y=A ^T X

5 . The method according to claim 1 , wherein the contribution degree of the synthetic principal component is determined according to the eigenvalue, the independent variable matrix is screened according to the contribution degree, and a preset number of For each principal component, a prediction matrix is constructed according to each principal component of the preset number, including: The eigenvalues λ ₁ , λ ₂ , ..., λ _k represent the contribution rate of the synthetic principal components, and the eigenvectors of each principal component are arranged in descending order. If it is assumed that the cumulative value of the first p eigenvalues is close to 1, where p<k, Indicates that the above p principal components have a high degree of information retention to the original index set, then the cumulative contribution of the principal components is defined as

And select the value according to the preset range to filter the principal components, and the obtained principal components of the k-item economic operation indicators are as follows: y′ ₁ =a ₁₁ X ₁ +a ₁₂ X ₂ +…+a _1k X _k y′ ₂ =a ₂₁ X ₁ +a ₂₂ X ₂ +…+a _2k X _k … y′ _p = a _p1 X ₁ +a _p2 X ₂ +...+a _pk X _k The principal components synthesized by k items of economic operation indicators constitute the prediction matrix Y'. 6. method according to claim 5, is characterized in that, for each principal component after synthesis, after doing covariance calculation, obtain:

The variance of the principal component y _j is equal to its corresponding eigenvalue λ _i , and the covariance between any two different principal components is 0, that is, there is no correlation between the principal components. 7. The method of claim 6, wherein The energy demand forecasting model is a multiple linear regression analysis model, including: Z=β ₀ +β ₁ Y ₁ ′+β ₂ Y′ ₂ +…+β _k Y′ _k Among them, it is assumed that there is a linear relationship between the explained variable Z and multiple explanatory variables Y′, Y′ ₂ ,…Y′ _k , which are multivariate linear functions of the explanatory variables, β ₀ , β ₁ , …, β _k are regression coefficients , the explanatory variable Z is the energy demand, and the explanatory variable Y is the prediction matrix formed by the processing of economic development planning, environmental policy objectives, industrial structure and other factors. 8. The method of claim 5, wherein The preset range is 85% to 95%. 9. An energy demand forecasting device, characterized in that it comprises: The acquisition module is used for acquiring original independent variable data, establishing an original independent variable matrix according to the original independent variable data and performing standardization processing to obtain a standardized matrix; wherein, the original independent variable data includes a plurality of power industry and economic development related categories of indicators including industrial value added, production and price indices; a construction module, configured to obtain a correlation coefficient matrix and obtain eigenvalues of the correlation coefficient matrix according to the standardized matrix, calculate corresponding eigenvectors according to the eigenvalues, and construct an independent variable matrix according to the eigenvectors; The determining module is configured to determine the contribution degree of the synthetic principal component according to the eigenvalue, screen the independent variable matrix according to the contribution degree, and determine a preset number of each principal component, according to the preset number of Each principal component constructs a prediction matrix; a building module for acquiring dependent variable data, and establishing a dependent variable matrix according to the dependent variable data; wherein the dependent variable data includes historical energy demand data; a building module for building an energy demand forecasting model according to the forecasting matrix and the dependent variable matrix. 10. A computer device, comprising: a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the execution of claims 1 to 8 The energy demand forecasting method described in any one of.

Images