Economic cycle and power consumption cycle dynamic interaction influence analysis method and product

Abstract

The invention discloses a dynamic interaction influence analysis method and product for an economic cycle and a power consumption cycle, and relates to the technical field of power consumption, and the method comprises the steps: obtaining the economic cycle and the power consumption cycle according to the representative index data of the economic cycle and the representative index data of the power consumption cycle; constructing a TVP-VAR model by taking the economic cycle and the power consumption cycle as variables; setting prior distribution of model parameters based on a Bayesian inference method, and obtaining posterior distribution of the model parameters by combining economic cycle and power consumption cycle data samples; on the basis of an MCMC method, model parameters are extracted from posterior distribution through an iteration process, final model parameters are obtained, and a final evaluation model is obtained and used for predicting and analyzing the time-varying interaction relation between the economic cycle and the power consumption cycle. The long-term dynamic relationship between economic development and power demand is disclosed, and power supply can be adjusted in real time according to the trend of the economic cycle, so that the power demands of different areas and different industries are met.

Description

Technical Field

[0001] The present invention relates to the technical field of power consumption, and in particular to a method and product for analyzing the dynamic interaction impact of an economic cycle and a power consumption cycle. Background Art

[0002] As the foundation of the national economy, the supply and demand of electricity is directly related to all aspects of economic activities. With the transformation of the economic structure of various regions and the change of economic growth mode, the relationship between economy and electricity will also change accordingly. The growth rate of electricity demand is not always synchronized with the economic growth rate. When the economic growth rate is high, the demand for electricity increases. Due to the relatively stable power supply, there may be a power shortage in the short term. When the economic growth rate is low, the demand for electricity decreases, and there is a possibility of waste of electricity. Moreover, the relationship between electricity demand and its added value in different industries varies greatly. The high-energy consumption industry has a large demand for electricity, while the service industry has a low demand for electricity. It is necessary to sort out the long-term interaction between the economic cycle and the electricity consumption cycle, which is of great significance for the rational planning of power infrastructure construction and the optimization of resource allocation.

[0003] Therefore, how to provide an economic cycle and electricity consumption cycle analysis method and product that can reveal the long-term dynamic relationship between economic development and electricity demand is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0004] In view of the unclear interaction between economic cycles and electricity consumption cycles in the prior art, the present invention proposes a dynamic interaction impact analysis method and related products of economic cycles and electricity consumption cycles to reveal the long-term dynamic relationship between economic development and electricity demand, so as to help power companies adjust electricity supply in real time according to the trend of economic cycles, so as to meet the electricity demand of different regions and industries.

[0005] The present invention provides a method for analyzing the dynamic interaction impact of economic cycles and power consumption cycles, comprising the following steps:

[0006] S1: Obtain representative indicator data of the economic cycle and the representative indicator data of the power consumption cycle and perform HP filtering to obtain long-term trend components and short-term fluctuation components, which are used as long-term and short-term economic cycles and power consumption cycles respectively;

[0007] S2: Taking the economic cycle and power consumption cycle corresponding to the short-term fluctuation components as variables, a time-varying parameter vector autoregressive random fluctuation model is constructed;

[0008] S3: The prior distribution of model parameters is set based on the Bayesian inference method, and the posterior distribution of model parameters is obtained by combining the given economic cycle and electricity consumption cycle data samples; based on the Gibbs sampling method in Markov chain Monte Carlo simulation, the model parameters are extracted from the posterior distribution through an iterative process until the iteration ends, and the final model parameters are obtained;

[0009] S4: A final evaluation model is derived based on the final model parameters, and is used to predict and analyze the time-varying interaction between the economic cycle and the electricity consumption cycle.

[0010] Preferably, in S1, the economic cycle and the power consumption cycle of the specified industry in the specified area are obtained by using the HP filtering method based on the representative indicator data of the economic cycle and the representative indicator data of the power consumption cycle of the specified industry in the specified area.

[0011] Preferably, in S1, the representative indicator data of the economic cycle include one or more of the following: PMI index, consumer confidence index, total investment in newly started projects, gross domestic product, added value of industrial enterprises above designated size, railway freight volume, consumer price index, corporate profits, and unemployment rate;

[0012] Representative indicator data of the electricity consumption cycle include one or more of the following: total social electricity consumption, electricity consumption of the primary, secondary and tertiary industries, and residents' daily electricity consumption.

[0013] Preferably, the economic cycle and the electricity consumption cycle both include one or more of the following: length, expansion period, contraction period, peak, trough, and fluctuation amplitude.

[0014] Preferably, the disturbance term of the time-varying parameter vector autoregressive stochastic fluctuation model conforms to a normal distribution with variable variance.

[0015] Preferably, in S3, the prior distribution of the model parameters is set to include: time-varying parameters, initial volatility and initial values ​​of the lower triangular matrix elements; wherein the time-varying parameters include the time-varying influence coefficient of the economic cycle on the electricity consumption cycle, and the lower triangular matrix elements are used to characterize the causal relationship matrix between the economic cycle and the electricity consumption cycle.

[0016] Preferably, the S1 further comprises the following steps: identifying the economic cycle and the electricity consumption cycle corresponding to the long-term trend component, and performing the following calculations:

[0017] Calculate the correlation coefficient between the economic cycle and the electricity consumption cycle;

[0018] Calculating the degree of coordination between the economic cycle and the electricity consumption cycle;

[0019] Cointegration is used to test whether there is a long-term stable relationship between the economic cycle and the electricity consumption cycle.

[0020] 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 the memory is used to store representative indicator data of an economic cycle and representative indicator data of a power consumption cycle; when the processor executes the program, the steps of the method for analyzing the dynamic interaction impact of the economic cycle and the power consumption cycle are implemented.

[0021] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program and a time-varying parameter vector autoregressive stochastic fluctuation model are stored. When the computer program is executed by a processor, the steps of the dynamic interactive impact analysis method of the economic cycle and the electricity consumption cycle are implemented.

[0022] Compared with the prior art, it has the following beneficial effects:

[0023] The present invention proposes a dynamic interactive impact analysis method of the economic cycle and the electricity consumption cycle, constructs a time-varying parameter vector autoregressive random fluctuation model with the economic cycle and the electricity consumption cycle as variables, sets the initial value of the model parameter, estimates the parameter according to the Monte Carlo simulation method, and finally obtains an evaluation model. The final evaluation model can determine the interactive impact mechanism between the economic cycle and the electricity consumption cycle, that is, evaluate how economic factors affect electricity demand at different times and regions, and how great the impact is. The disturbance term of the model conforms to the normal distribution with variable variance to capture the impact of other factors such as temperature and climate on the electricity consumption cycle in addition to the economic cycle, and fully considers how the relationship between the economic cycle and the electricity consumption cycle will change when factors such as temperature, climate, and industrial structure change, and can comprehensively and systematically analyze the complex relationship between the economic cycle and the electricity consumption cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] 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 prior art descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without creative work.

[0025] Figure 1 A flow chart of a method for analyzing the dynamic interaction impact of economic cycles and power consumption cycles provided by an embodiment of the present invention;

[0026] Figure 2 A schematic diagram of a time-varying parameter vector autoregressive random fluctuation model provided by an embodiment of the present invention;

[0027] Figure 3A flow chart of a time-varying parameter vector autoregressive random fluctuation model using the economic cycle and the power consumption cycle as variables provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] like Figure 1 As shown, the dynamic interactive impact analysis method of the economic cycle and the power consumption cycle disclosed in the embodiment of the present invention includes the following steps:

[0030] S1: Obtain representative indicator data of the economic cycle and the representative indicator data of the power consumption cycle and perform HP filtering to obtain long-term trend components and short-term fluctuation components, which are used as long-term and short-term economic cycles and power consumption cycles respectively;

[0031] S2: Taking the economic cycle and power consumption cycle corresponding to the short-term fluctuation components as variables, a time-varying parameter vector autoregressive random fluctuation model is constructed;

[0032] S3: The prior distribution of model parameters is set based on the Bayesian inference method, and the posterior distribution of model parameters is obtained by combining the given economic cycle and electricity consumption cycle data samples; based on the Gibbs sampling method in Markov chain Monte Carlo simulation, the model parameters are extracted from the posterior distribution through an iterative process until the iteration ends, and the final model parameters are obtained;

[0033] S4: The final evaluation model is derived based on the final model parameters and used to predict and analyze the time-varying interaction between the economic cycle and the electricity consumption cycle.

[0034] In one embodiment, the HP filter (Hodrick-Prescott Filter) is a time series analysis tool in the field of macroeconomic analysis, which is used to analyze the long-term trend and short-term fluctuations of economic growth and electricity consumption. The long-term trend and short-term fluctuations are respectively used as long-term and short-term economic cycles and electricity consumption cycles, wherein the short-term fluctuation component is used to analyze the interaction between the time-varying characteristics of the economic cycle and the electricity consumption cycle through the time-varying parameter vector autoregressive random fluctuation model, and the long-term trend component is used to analyze the interaction between stable characteristics. Specifically, the following analysis steps are also included for the long-term trend component as the corresponding economic cycle and electricity consumption cycle:

[0035] Calculate the correlation coefficient between the economic cycle and the electricity consumption cycle;

[0036] Calculate the degree of coordination between the economic cycle and the electricity consumption cycle;

[0037] Cointegration is used to test whether there is a long-term stable relationship between the economic cycle and the electricity consumption cycle.

[0038] The above calculations are used to achieve the interaction and impact analysis between the economic cycle and electricity consumption cycle corresponding to the long-term trend component, such as whether economic growth will lead to an increase in electricity consumption, or whether electricity consumption will affect economic growth.

[0039] In one embodiment, for short-term volatility components, this embodiment first constructs an initial model, which includes representative indicators of the economic cycle and the electricity consumption cycle, estimates model parameters, and sets the initial values ​​and distribution of the model parameters; iterates the initial values ​​according to the Monte Carlo simulation method until the parameter values ​​converge; and obtains the final evaluation model.

[0040] In one embodiment, in many cases, the data generation process of economic variables has drift coefficients and random volatility shocks. The time-varying parameter vector autoregressive random volatility model is used to analyze the time-varying relationship between different time series. It mainly includes the following three aspects:

[0041] First, the variables are representative of the economic cycle and the electricity consumption cycle, and the data meet the requirements of cointegration or stability of the same order;

[0042] Second, the parameters are time-varying, that is, the coefficients of the economic cycle on the electricity consumption cycle change over time, allowing temporary or permanent changes in the coefficients to capture the structural changes in the relationship between the economic cycle and the electricity consumption cycle;

[0043] Third, the model disturbance term conforms to the normal distribution with variable variance to capture the impact of factors other than the economic cycle, such as temperature and climate, on the electricity consumption cycle.

[0044] In one embodiment, representative indicators of the economic cycle include one or more of the following: PMI index, consumer confidence index, total investment in newly started projects, gross domestic product, value added of industrial enterprises above designated size, railway freight volume, consumer price index (CPI), corporate profits, unemployment rate, etc., and representative indicators of the electricity consumption cycle include one or more of the following: total social electricity consumption, electricity consumption of the primary, secondary and tertiary industries, and electricity consumption for residents' lives, etc.

[0045] In one embodiment, the process of constructing the initial model of the time-varying parameter vector autoregressive stochastic fluctuation model is as follows:

[0046] The vector autoregression model (VAR model) is as follows:

[0047] (1)

[0048] In the formula, is a k-dimensional observable variable, i.e., a variable of the economic cycle and the electricity consumption cycle, A, F1, ..., F S , is a K×K coefficient matrix, u t is the K×1-dimensional exogenous shock disturbance term, , Indicates u t Variance of the vector.

[0049]

[0050] In the formula, for u t The standard deviation of .

[0051] Assume A is a lower triangular matrix, , the lower triangular matrix (A matrix) is used to represent the causal relationship matrix between the economic cycle and the electricity consumption cycle, and its non-zero elements are used to represent the immediate impact of one variable on another variable. In the study of the relationship between the economic cycle and the electricity consumption cycle, if the electricity consumption cycle variable is directly affected by the economic cycle variable at a certain point in time, then there will be a non-zero value in the corresponding position in the A matrix to represent this immediate impact.

[0052] Formula (1) is multiplied by A on the left -1 have to:

[0053] ,

[0054] In the formula, ,i=1,...,s,B i The row elements are stacked to obtain dimensional vector β, is a disturbance term that follows a standard normal distribution, ,make , is the Kronecker product, is the identity matrix, for The above expression can be written as:

[0055] (2)

[0056] In the formula, β is The coefficient matrix of order.

[0057] If β, A, Satisfying the time-varying characteristics, the time-varying parameters are as follows: the coefficient matrix β t , parameter A tand the variance matrix , then formula (2) becomes the following TVP-VAR model, which is characterized by the fact that all coefficients are time-varying and can describe the structural changes of the economic system, that is, the time-varying parameter vector autoregressive stochastic fluctuation model (TVP-VAR):

[0058] (3)

[0059] Assume that the parameters in equation (3) obey a random walk process:

[0060]

[0061] In the formula, is the element of the transposed matrix of the lower triangular matrix A, , , , for The variance of

[0062] , , .

[0063] In one embodiment, Figure 2 As shown in the figure, first, the model is set, including the number of variables and the lagging indicator value of the model. Secondly, the MCMC method (Monte Carlo simulation and Markov chain) is used to estimate the parameters, including the time-varying parameters, initial volatility, and the prior distribution setting of the initial values ​​of the lower triangular matrix elements. The Gibbs sampling method is used for parameter sampling and the convergence standard is set. Finally, the final model parameters are obtained, and the final evaluation model can be derived from the final model parameters.

[0064] In the specific implementation, the model parameters to be estimated are β, a and h, and Bayesian inference is used to perform MCMC (Markov Chain Monte Carlo Sampling Method) simulation estimation. , , set the prior probability density , based on the given data Y, we can get the posterior distribution , and then the MCMC simulation method is used for iteration. The steps are as follows:

[0065] S31: Given β, a, h, The prior distribution of

[0066] S32: Based on the sample ;

[0067] S33: Based on the sample ;

[0068] S34: Based on the sample ;

[0069] S35: Based on the sample ;

[0070] S36: Based on the sample ;

[0071] S37: Based on the sample ;

[0072] S38: Return to S2 and iterate in sequence until a certain condition is met.

[0073] In one embodiment, in S3, setting the prior distribution of model parameters includes: time-varying parameters, initial volatility, and initial values ​​of lower triangular matrix elements.

[0074] In one embodiment, Figure 3 As shown, the short-term fluctuation components as the corresponding economic cycle and electricity consumption cycle include one or more of the following: length, expansion period, contraction period, peak, trough, and fluctuation amplitude.

[0075] In one embodiment, the interaction between the economic cycle and the electricity consumption cycle is analyzed based on the final model. The interaction specifically refers to: first, the short-term, medium-term and long-term impact of the economic cycle on the electricity consumption cycle; second, selecting representative time points, such as economic crises, epidemic outbreaks, economic structural transformation, etc., to analyze the impact of the economic cycle on the electricity consumption cycle.

[0076] In one embodiment, Figure 3 As shown, based on the representative indicator data of the economic cycle and the representative indicator data of the power consumption cycle of the specified industry in the specified area, the HP filtering method is used to obtain the economic cycle and the power consumption cycle of the specified industry in the specified area.

[0077] Based on the short-term fluctuation components obtained by HP filtering, the length, expansion period, contraction period, peak, trough and fluctuation amplitude of regional and industrial economic cycles and electricity consumption cycles are examined; the TVP-VAR model is used to examine the relationship between regional and industrial economic cycles and electricity consumption cycles.

[0078] The following is a specific example of analyzing the interaction between economic cycles and power consumption cycles using the method of the embodiment of the present invention:

[0079] S10: Based on economic theory, select leading indicators, synchronous indicators and lagging indicators of the economic cycle

[0080] Leading indicators: PMI index, consumer confidence index, total investment in newly started projects

[0081] Synchronous indicators: GDP, industrial added value above designated size, railway freight volume

[0082] Lagging indicators: Consumer Price Index (CPI), corporate profits, unemployment rate

[0083] S20: Analyze the interaction between the above indicators and electricity consumption based on the analysis method of the interaction between the economic cycle and the electricity consumption cycle

[0084] S30: Analyze the relationship between the above indicators and electricity consumption in different regions based on the analysis method of the interaction between economic cycle and electricity consumption cycle.

[0085] The above is a detailed introduction to the dynamic interactive impact analysis method of the economic cycle and the electricity consumption cycle provided by the present invention. This article uses specific examples to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0086] In this article, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

Claims

1. A dynamic interactive impact analysis method of economic cycle and power consumption cycle, characterized in that: The steps include: S1: Obtain representative indicator data of the economic cycle and the representative indicator data of the power consumption cycle and perform HP filtering to obtain long-term trend components and short-term fluctuation components, which are used as long-term and short-term economic cycles and power consumption cycles respectively; S2: Taking the economic cycle and power consumption cycle corresponding to the short-term fluctuation components as variables, a time-varying parameter vector autoregressive random fluctuation model is constructed; S3: The prior distribution of model parameters is set based on the Bayesian inference method, and the posterior distribution of model parameters is obtained by combining the given economic cycle and electricity consumption cycle data samples; based on the Gibbs sampling method in Markov chain Monte Carlo simulation, the model parameters are extracted from the posterior distribution through an iterative process until the iteration ends, and the final model parameters are obtained; S4: A final evaluation model is derived based on the final model parameters, and is used to predict and analyze the time-varying interaction between the economic cycle and the electricity consumption cycle.

2. The dynamic interactive impact analysis method of economic cycle and power consumption cycle according to claim 1 is characterized in that: In S1, the economic cycle and the power consumption cycle of the specified industry in the specified area are obtained by using the HP filtering method according to the representative indicator data of the economic cycle and the representative indicator data of the power consumption cycle of the specified industry in the specified area.

3. The dynamic interactive impact analysis method of economic cycle and power consumption cycle according to claim 1 is characterized in that: In S1, the representative indicator data of the economic cycle include one or more of the following: PMI index, consumer confidence index, total investment in newly started projects, gross domestic product, added value of industrial enterprises above designated size, railway freight volume, consumer price index, corporate profits, and unemployment rate; Representative indicator data of the electricity consumption cycle include one or more of the following: total social electricity consumption, electricity consumption of the primary, secondary and tertiary industries, and residents' daily electricity consumption.

4. The dynamic interactive impact analysis method of economic cycle and power consumption cycle according to claim 1 is characterized in that: The economic cycle and electricity consumption cycle both include one or more of the following: length, expansion period, contraction period, peak, trough, and fluctuation amplitude.

5. The dynamic interactive impact analysis method of economic cycle and power consumption cycle according to claim 1 is characterized in that: The disturbance term of the time-varying parameter vector autoregressive stochastic fluctuation model conforms to a normal distribution with variable variance.

6. The dynamic interactive impact analysis method of economic cycle and power consumption cycle according to claim 1 is characterized in that: In S3, the prior distribution of the model parameters is set to include: time-varying parameters, initial volatility and initial values ​​of the lower triangular matrix elements; wherein the time-varying parameters include the time-varying influence coefficient of the economic cycle on the electricity consumption cycle, and the lower triangular matrix elements are used to characterize the causal relationship matrix between the economic cycle and the electricity consumption cycle.

7. The dynamic interaction analysis method of economic cycle and power consumption cycle according to claim 1 is characterized in that: The S1 also includes the following steps: identifying the economic cycle and electricity consumption cycle of the long-term trend component, and performing the following calculations: Calculate the correlation coefficient between the economic cycle and the electricity consumption cycle; Calculating the degree of coordination between the economic cycle and the electricity consumption cycle; Cointegration is used to test whether there is a long-term stable relationship between the economic cycle and the electricity consumption cycle.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The memory is used to store representative indicator data of the economic cycle and representative indicator data of the electricity consumption cycle; when the processor executes the program, it implements the steps of the dynamic interactive impact analysis method of the economic cycle and the electricity consumption cycle as described in any one of claims 1-7.

9. A non-transitory computer-readable storage medium, characterized in that: A computer program and a time-varying parameter vector autoregressive random fluctuation model are stored thereon, and when the computer program is executed by a processor, the steps of the dynamic interactive impact analysis method of the economic cycle and the electricity consumption cycle as described in any one of claims 1-7 are implemented.

Images