Regional energy economy prediction method and device, and storage medium

Abstract

The invention relates to the technical field of energy economy prediction, in particular to a regional energy economy prediction method, which can obtain influence factors of a to-be-predicted region and perform preprocessing; performing feature engineering on the influence factor data, evaluating the correlation between the influence factor data and energy economy, and selecting feature indexes; according to the characteristic indexes, constructing a total economic development prediction model, an industrial structure transformation prediction model and an energy structure transformation prediction model, and outputting corresponding results; evaluating the prediction precision of each model; and analyzing prediction results under different scenes according to the prediction precision to obtain weights of the prediction results in energy economy, and constructing a multiple regression regional energy economy prediction model to perform regional energy economy prediction. According to the technical scheme provided by the invention, the accuracy and reliability of prediction are improved by fusing multi-dimensional data and an advanced machine learning technology.

Description

Technical Field

[0001] The present invention relates to the technical field of energy economic forecasting, and in particular to a method, device and storage medium for regional energy economic forecasting. Background Art

[0002] With the development of the economy and society, the total energy consumption continues to grow. At the same time, with the increasingly severe environmental problems, in order to achieve sustainable economic and social development, policymakers are faced with the dual tasks of energy structure adjustment and high-quality economic development. Therefore, maintaining a dynamic balance between energy structure and economic development is one of the most important tasks in the current field of energy economics.

[0003] At present, the consumption of fossil energy has declined, but in the face of the global population surge and increasingly harsh environment, the task of optimizing and adjusting the energy structure is still urgent. Therefore, accurate prediction of regional energy economic development trends is crucial for resource allocation, industrial layout and macro decision-making.

[0004] Although current energy economic forecasting methods can reveal the relationship between energy consumption and economic development to a certain extent, they still have significant shortcomings. Although statistical analysis and econometric models are based on historical data, they often ignore the time series characteristics and nonlinear relationships of data and are difficult to adapt to market dynamics. Although machine learning and deep learning technologies have advantages in processing complex data patterns, they are limited by large data requirements, poor model interpretability, and insufficient rapid response to real-time data. In addition, existing technologies rarely comprehensively consider regional-specific economic, industrial, energy and other multi-dimensional factors, resulting in the prediction results may not accurately reflect the actual trend of regional energy consumption.

[0005] Therefore, the existing energy economic forecasting methods lack forecasting accuracy and practicality. Summary of the invention

[0006] In view of this, the purpose of the present invention is to provide a regional energy economic forecasting method, device and storage medium to solve the problem of insufficient prediction accuracy and practicality of the energy economic forecasting method in the prior art.

[0007] According to a first aspect of an embodiment of the present invention, a method for regional energy economic forecasting is provided, comprising: Obtain the influencing factor data of the economic aggregate development model, the influencing factor data of the industrial structure transformation model and the influencing factor data of the energy structure transformation model of the region to be predicted; Preprocessing the influencing factor data; Perform feature engineering on the preprocessed influencing factor data, and then evaluate the correlation between the influencing factor data and energy economy in turn, and select the influencing factor data with a correlation greater than a preset value as a feature indicator; According to the characteristic indicators, an economic aggregate development forecasting model is constructed based on a multivariate regression model, and the economic aggregate forecasting results are output; According to the characteristic indicators, an industrial structure transformation prediction model is constructed based on time series analysis and structural equation model, and the industrial structure prediction results are output; According to the characteristic indicators, an energy structure transformation prediction model is constructed based on multiple regression analysis and system dynamics model, and the energy structure prediction results are output; Based on historical data and the forecast results of each model, the forecast accuracy of each model is evaluated through residual analysis; Analyze the forecast results under different scenarios using scenario analysis based on the forecast accuracy to obtain the results of economic volume, industrial structure and energy structure; The weights of the total economic volume results, industrial structure results and energy structure results in energy economics are analyzed in turn, a multivariate regression regional energy economic forecasting model is constructed, and the regional energy economic forecast is carried out using the multivariate regression regional energy economic forecasting model.

[0008] Preferably, the method further comprises: The influencing factor data of the economic aggregate development model include historical GDP data, fixed asset investment, scientific and technological research and development investment, and industrial added value; the influencing factor data of the industrial structure transformation model include the added value of the primary industry, the added value of the secondary industry, the added value of the tertiary industry, the number of employees, and fixed asset investment; the influencing factor data of the energy structure transformation model include the consumption and proportion of various types of energy.

[0009] Preferably, after preprocessing the influencing factor data, the method further includes: The influencing factor data is smoothed using a Savitzky-Golay filter to eliminate noise and anomalies in the influencing factor data.

[0010] Preferably, the correlation between the influencing factor data and energy economics is evaluated, including: Obtain historical data on the economic aggregate development model, industrial structure transformation model, and energy structure transformation model; converting the influencing factor data and the historical data into grades; Calculate the level difference between the influencing factor data and the historical data of the model to be predicted; The Spearman rank correlation coefficient was calculated based on the rank differences.

[0011] Preferably, the evaluation of the prediction accuracy of each model by residual analysis includes: The residual value is calculated through residual analysis; The residual sum of squares is calculated according to the following formula:

[0012] in, S is the residual sum of squares, is the residual value, n for quantity; The root mean square error is calculated according to the following formula:

[0013] in, is the root mean square error.

[0014] Preferably, after evaluating the prediction accuracy of each model by residual analysis, the method further includes: A residual histogram is drawn based on the calculated residual values, and the distribution of the residuals is evaluated based on the residual histogram to determine whether the residual distribution satisfies a normal distribution and obtain a determination result.

[0015] According to a second aspect of an embodiment of the present invention, there is provided a regional energy economic forecasting device, comprising: A main controller, and a memory connected to the main controller; The memory stores program instructions; The main controller is used to execute program instructions stored in the memory to perform any of the above methods.

[0016] According to a third aspect of an embodiment of the present invention, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the method described in any one of the above is implemented.

[0017] The technical solution provided by the embodiments of the present invention may have the following beneficial effects: It is understandable that the technical solution shown in the present invention can obtain the influencing factors of the area to be predicted and perform preprocessing; perform feature engineering on the influencing factor data, evaluate its correlation with energy economy, and select characteristic indicators; construct an economic total development forecast model, an industrial structure transformation forecast model, and an energy structure transformation forecast model based on the characteristic indicators, and output the corresponding results; evaluate the prediction accuracy of each model; analyze the prediction results under different scenarios based on the prediction accuracy, derive its weight in energy economy, construct a multivariate regression regional energy economy forecast model, and perform regional energy economy forecasting. The technical solution shown in the present invention improves the accuracy and reliability of prediction by integrating multidimensional data and advanced machine learning technology.

[0018] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0020] Figure 1 is a schematic diagram of steps of a method for regional energy economic forecasting according to an exemplary embodiment; Figure 2 The present invention is a flowchart of a method for regional energy economic forecasting according to an exemplary embodiment. DETAILED DESCRIPTION

[0021] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0022] In one embodiment, Figure 1 is a schematic diagram of the steps of a regional energy economic forecasting method according to an exemplary embodiment. Figure 1 and Figure 2 , providing a regional energy economic forecasting method, including: Step S11, obtaining the influencing factor data of the economic aggregate development model, the influencing factor data of the industrial structure transformation model and the influencing factor data of the energy structure transformation model of the region to be predicted.

[0023] It should be noted that the influencing factor data of the economic aggregate development model include historical GDP data, fixed asset investment, scientific and technological research and development investment, and industrial added value.

[0024] The influencing factor data of the industrial structure transformation model include the added value of the primary industry, the added value of the secondary industry, the added value of the tertiary industry, the number of employees and the fixed asset investment.

[0025] The influencing factor data of the energy structure transformation model include the consumption and proportion of various energy types, and the various energy types include at least coal, oil, natural gas, electricity, renewable energy, etc.

[0026] Step S12: pre-process the influencing factor data to ensure its consistency and integrity.

[0027] It should be noted that, after the influencing factor data is preprocessed, the method further includes: using a Savitzky-Golay filter to smooth the influencing factor data to eliminate noise and anomalies in the influencing factor data.

[0028] The Savitzky-Golay filter formula is as follows:

[0029] in, is the weight determined by the polynomial coefficients and the window position, is half the filter window size.

[0030] Step S13: perform feature engineering on the preprocessed influencing factor data, and then evaluate the correlation between the influencing factor data and energy economy in turn, and select the influencing factor data with a correlation greater than a preset value as a feature indicator.

[0031] In specific practice, since there are many variables in the prediction process, Ridge regression is used to select features. The specific formula is as follows:

[0032] in, is the vector parameter of the model, For the The target variable has observations, For the The feature vector of observations, is the sample size, is the number of features, is the regularization parameter, which is used to control the strength of regularization.

[0033] It should be noted that the Spearman rank correlation coefficient can be used to evaluate the correlation between the influencing factor data and energy economics, including the following steps: Step S131, obtaining historical data of the economic aggregate development model, the industrial structure transformation model and the energy structure transformation model.

[0034] Step S132: convert the influencing factor data and the historical data into levels, which can be properly processed during the calculation process.

[0035] Step S133: Calculate the level difference between the influencing factor data and the historical data of the model to be predicted, and the formula is as follows:

[0036] in, and Represent the levels of the two variables.

[0037] Step S134: Calculate the Spearman rank correlation coefficient according to the rank difference, and the formula is as follows:

[0038] in, is the total number of observations. Spearman rank coefficient The value range is -1 to 1, the stronger the positive correlation The closer it is to 1, the stronger the negative correlation. The closer it is to -1.

[0039] Additionally, the sum of squares of rank differences can be calculated: .

[0040] Step S14: construct an economic aggregate development forecasting model based on the multivariate regression model according to the characteristic indicators, and output the economic aggregate forecasting results.

[0041] In specific practice, multiple linear regression is often used as a mathematical model for the relationship between multiple influencing factors and the object to be predicted. Multiple linear regression is expressed as:

[0042] in, is the total economic output, , , …, is an influencing factor. , , …, It is a model parameter, which indicates the influence of the factors on the predicted object. and represent the intercept and error respectively.

[0043] In the multivariate linear regression prediction model, the model parameters , , …, It is usually estimated by minimizing the sum of squares of the error terms. This method is also called ordinary least squares method. Its principle is to find parameters that minimize the error between the model prediction value and the observed value. The formula is:

[0044] in, is the sample size, For the Observations, For the The model predicted value.

[0045] Step S15: According to the characteristic indicators, an industrial structure transformation prediction model is constructed based on time series analysis and structural equation model, and the industrial structure prediction result is output.

[0046] In specific practice, time series analysis and forecasting, as a method of predicting future trends based on historical data, is used to predict the baseline trend of value-added in various industries. Structural equation modeling, as a statistical method for studying the complex relationship between variables, is used to analyze the importance of different influencing factors to the transformation of industrial structure.

[0047] The autoregressive integrated moving average model is selected for time series analysis and prediction. The specific implementation process of this model is as follows: The autoregressive integrated moving average model consists of three parts, usually expressed as:

[0048] in, is the order of the autoregressive term, is the order of the difference, is the order of the sliding average term.

[0049] (1) Autoregressive part:

[0050] in, is a time series at a point in time The value of is a constant term, , , …, is the autoregressive coefficient, is the error term.

[0051] (2) Differential part:

[0052] in, is the symbol of the difference operation, is the order of the difference.

[0053] (3) Sliding average part:

[0054] in, is the error term, , , …, is the sliding average coefficient, For white noise.

[0055] The structural equation model consists of two parts: measurement model and structural model. The construction process is as follows: The measurement model defines the relationship between the observed variables and the latent variables, usually assuming that the observed variables are linear functions of the latent variables plus an error term.

[0056]

[0057] in, is the vector of observed variables, is the factor loading matrix, which contains the relationship between the observed variables and the latent variables. is the vector of latent variables, is the vector of measurement errors.

[0058] Structural models, which define the causal relationships between latent variables, can include multiple equations, each describing one relationship between latent variables.

[0059]

[0060] in, is the vector of endogenous latent variables, is the relationship matrix between endogenous latent variables, is the influence matrix of exogenous latent variables on endogenous latent variables, is a vector of exogenous observed variables, is the error term of the endogenous latent variable.

[0061] Step S16: According to the characteristic indicators, an energy structure transformation prediction model is constructed based on multiple regression analysis and system dynamics model, and the energy structure prediction result is output.

[0062] In this step, multivariate regression analysis is suitable for analyzing the relationship between the proportion of various energy consumption and variables such as economic growth and industrial structure; the system dynamics model comprehensively considers multiple influencing factors by simulating the dynamic changes of the energy system.

[0063] The construction of the multivariate regression analysis prediction model is the same as the model construction in step S14.

[0064] The system dynamics model is based on feedback control theory. According to the integrity and hierarchy of the system, the system is divided into several interrelated subsystems based on the decomposition principle, namely:

[0065] in, Represents the entire system, Represents a subsystem, .

[0066] The model includes four types of variables: state variables, rate variables, auxiliary variables and constants. State variables represent variables of the cumulative effect of the system; rate variables reflect the rate of change of state variables; auxiliary variables are independent of each other and are represented by auxiliary equations; constants do not change over time.

[0067] The state variables are:

[0068] in, is a state variable exist The value of the moment, is the initial value, is the rate of change of the state variable, and Represent the input and output flows of state variables respectively.

[0069] The rate equation is:

[0070] in, For auxiliary variables The value of the moment, Is a constant.

[0071] The auxiliary equation is:

[0072] in, For auxiliary variables other than the auxiliary variable to be determined, The value of the moment.

[0073] The relationship between the variables is:

[0074] in, is the relationship matrix, the reaction rate variable and state variables Between and auxiliary variables The nonlinear relationship itself at the same time.

[0075] Step S17: Based on the historical data and the prediction results of each model, the prediction accuracy of each model is evaluated through residual analysis.

[0076] It should be noted that the above steps are specifically as follows: The residual value is calculated through residual analysis, and the formula is as follows:

[0077] in, is the observed value, is the model prediction value.

[0078] The residual sum of squares is calculated according to the following formula:

[0079] in, S is the residual sum of squares, is the residual value, n for quantity; The root mean square error is calculated according to the following formula:

[0080] in, is the root mean square error. The smaller the root mean square error value, the smaller the difference between the predicted value and the observed value, indicating that the prediction accuracy of the model is higher; the larger the root mean square error, the greater the difference between the predicted value and the observed value, indicating that the prediction accuracy of the model is lower.

[0081] It should be noted that, based on the residual analysis, a residual distribution test can also be performed: According to the calculated residual value, a residual histogram is drawn, and the distribution of the residual is evaluated according to the residual histogram to determine whether the residual distribution satisfies the normal distribution and obtain a determination result. This step is very important for correlation testing.

[0082] Furthermore, a residual autocorrelation test can be performed: check whether the model fully captures the time series characteristics of the data, and check whether the residuals show time series autocorrelation based on the residual distribution test.

[0083] Step S18: Analyze the forecast results under different scenarios using scenario analysis according to the forecast accuracy to obtain the total economic volume, industrial structure and energy structure results.

[0084] After the high-precision prediction model is built, the scenario analysis method is introduced to simulate the impact of policies on energy economic development.

[0085] Scenario analysis is a method of considering different future scenarios. By setting different policies and economic environments, it predicts the changes of the object under test under different scenarios. Common scenarios include baseline scenarios, optimistic scenarios and pessimistic scenarios, as follows: The first part is the scenario definition: Baseline scenario: A reference scenario with no policy changes, usually based on projections of current trends and historical data.

[0086] Optimistic scenario: Set a scenario in which both policy and economic conditions are relatively favorable, such as accelerated technological progress and enhanced policy support.

[0087] Pessimistic scenario: Set a scenario in which both policy and economic conditions are relatively unfavorable, such as slow technological progress and weakened policy support.

[0088] The second part is parameter setting: Policy variables: Based on the scenario definition, the weights of different policies are set from the perspectives of taxation, prices, subsidies, etc.

[0089] Economic environment: Adjust the economic environment parameters, take any two of the forecast results such as economic aggregate development, industrial structure transformation, and energy structure transformation as environmental parameters, and input them into the scenario analysis forecasting model as environmental factors, so as to predict the changing trend of the remaining variable under the influence of different policies.

[0090] The third part is model simulation: Under different policy scenarios, the changes in economic volume, industrial structure and energy structure are predicted based on the results of different forecasting models.

[0091] Step S19, analyzing the weights of the total economic volume results, industrial structure results and energy structure results in energy economics in turn, constructing a multivariate regression regional energy economic forecasting model, and using the multivariate regression regional energy economic forecasting model to perform regional energy economic forecasting.

[0092] In specific practice, data normalization is first performed: In order to eliminate the impact of different dimensions of economic aggregate, industrial structure, and energy structure on fusion calculation, the data is normalized. The formula is as follows:

[0093] in, For the data to be processed, is the minimum value in the data set, is the maximum value in the data set.

[0094] In specific practice, regarding the construction of multivariate regression regional energy economic forecasting model: This model is a mathematical model used to predict the relationship between economic aggregate, industrial structure, energy structure and energy economic development. The multiple linear regression is expressed as:

[0095] in, It is the development of energy economy. , , Represent the normalized data of economic aggregate, industrial structure and energy structure respectively, , , It is a model parameter, which indicates the influence of the factors on the predicted object. and represent the intercept and error respectively.

[0096] The parameter estimates are as follows: In the multivariate regression regional energy economic forecasting model, the model parameters , , It is estimated by minimizing the sum of squares of the error terms. This method is also called ordinary least squares method. Its principle is to find parameters that minimize the error between the model prediction value and the observed value. The formula is:

[0097] in, is the sample size, For the Observations, For the The model predicted value.

[0098] This model can be used to predict the impact of policies on energy economics.

[0099] It is understandable that the technical solution shown in this embodiment can obtain the influencing factors of the area to be predicted and perform preprocessing; perform feature engineering on the influencing factor data, evaluate its correlation with energy economy, and select characteristic indicators; construct an economic total development forecast model, an industrial structure transformation forecast model, and an energy structure transformation forecast model based on the characteristic indicators, and output the corresponding results; evaluate the prediction accuracy of each model; analyze the prediction results under different scenarios based on the prediction accuracy, derive its weight in energy economy, construct a multivariate regression regional energy economy forecast model, and perform regional energy economy forecasting. The technical solution shown in this embodiment improves the accuracy and reliability of prediction by integrating multidimensional data and advanced machine learning technology.

[0100] According to a second aspect of an embodiment of the present invention, there is provided a regional energy economic forecasting device, comprising: A main controller, and a memory connected to the main controller; The memory stores program instructions; The main controller is used to execute program instructions stored in the memory to perform any of the above methods.

[0101] According to a third aspect of an embodiment of the present invention, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the method described in any one of the above is implemented.

[0102] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0103] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0104] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0105] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0106] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0107] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0108] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0109] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0110] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A regional energy economic forecasting method, characterized in that: include: Obtain the influencing factor data of the economic aggregate development model, the influencing factor data of the industrial structure transformation model and the influencing factor data of the energy structure transformation model of the region to be predicted; Preprocessing the influencing factor data; Perform feature engineering on the preprocessed influencing factor data, and then evaluate the correlation between the influencing factor data and energy economy in turn, and select the influencing factor data with a correlation greater than a preset value as a feature indicator; According to the characteristic indicators, an economic aggregate development forecasting model is constructed based on a multivariate regression model, and the economic aggregate forecasting results are output; According to the characteristic indicators, an industrial structure transformation prediction model is constructed based on time series analysis and structural equation model, and the industrial structure prediction results are output; According to the characteristic indicators, an energy structure transformation prediction model is constructed based on multiple regression analysis and system dynamics model, and the energy structure prediction results are output; Based on historical data and the forecast results of each model, the forecast accuracy of each model is evaluated through residual analysis; Analyze the forecast results under different scenarios using scenario analysis based on the forecast accuracy to obtain the total economic volume, industrial structure and energy structure results; The weights of the total economic volume results, industrial structure results and energy structure results in energy economics are analyzed in turn, a multivariate regression regional energy economic forecasting model is constructed, and the regional energy economic forecast is carried out using the multivariate regression regional energy economic forecasting model.

2. The method according to claim 1, characterized in that Also includes: The influencing factor data of the economic aggregate development model include historical GDP data, fixed asset investment, scientific and technological research and development investment, and industrial added value; the influencing factor data of the industrial structure transformation model include the added value of the primary industry, the added value of the secondary industry, the added value of the tertiary industry, the number of employees, and fixed asset investment; the influencing factor data of the energy structure transformation model include the consumption and proportion of various types of energy.

3. The method according to claim 1, characterized in that After preprocessing the influencing factor data, the method further includes: The influencing factor data is smoothed using a Savitzky-Golay filter to eliminate noise and anomalies in the influencing factor data.

4. The method according to claim 1, characterized in that: Evaluate the relevance of influencing factor data to energy economics, including: Obtain historical data on the economic aggregate development model, industrial structure transformation model, and energy structure transformation model; converting the influencing factor data and the historical data into grades; Calculate the level difference between the influencing factor data and the historical data of the model to be predicted; The Spearman rank correlation coefficient was calculated based on the rank differences.

5. The method according to claim 1, characterized in that: The prediction accuracy of each model is evaluated by residual analysis, including: The residual value is calculated through residual analysis; The residual sum of squares is calculated according to the following formula: in, S is the residual sum of squares, is the residual value, n for quantity; The root mean square error is calculated according to the following formula: in, is the root mean square error.

6. The method according to claim 1, characterized in that After evaluating the prediction accuracy of each model through residual analysis, it also includes: A residual histogram is drawn based on the calculated residual values, and the distribution of the residuals is evaluated based on the residual histogram to determine whether the residual distribution satisfies a normal distribution and obtain a determination result.

7. A regional energy economic forecasting device, characterized in that: include: A main controller, and a memory connected to the main controller; The memory stores program instructions; The main controller is used to execute program instructions stored in the memory and execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images