Calculation method of regional carbon emissions based on power big data and time series

Abstract

The present invention discloses a method for calculating regional carbon emissions based on electric power big data and time series. The method comprises obtaining regional electric power big data and extracting the electricity consumption of each industry in the region; performing time series analysis on the electricity consumption of each industry in the region, and dividing the electricity consumption of each industry into long-term trend factors, seasonal period factors, cyclic factors, and random factors; calculating the joint confidence interval of the electricity consumption of each industry using the variance of the random factors of the electricity consumption of each industry, and obtaining the joint confidence interval of the regional electricity consumption; and performing weighted calculation based on the joint confidence interval of the regional electricity consumption and the carbon emission coefficient of the industry to obtain the regional carbon emission interval. The present invention combines electric power big data and time series analysis, and uses the time series analysis method and the joint confidence interval method to calculate the confidence interval of carbon emissions, thereby introducing uncertainty in the calculation of carbon emissions and enhancing the overall grasp of regional carbon emissions.

Description

Technical Field

[0001] The present invention relates to the field of carbon emission calculation, and in particular to a method for calculating regional carbon emissions based on power big data and time series. Background Art

[0002] At present, there are mainly the following methods for measuring carbon emissions: actual measurement method, input-output method, input coefficient method, and prediction methods using various machine learning models. The actual measurement method is to calculate carbon emissions based on measurement data obtained through on-site monitoring. However, for an entire region, the complete monitoring conditions that take into account various factors are very harsh, not to mention the huge consumption of manpower, material and financial resources. Therefore, it is not applicable to the calculation of carbon emissions in a region; the input-output method is to calculate the carbon emissions of the entire production process through the input-output model in economic statistics, but the update time of the input-output table is slow and the calculation is more complicated; and the prediction method through various machine learning models uses some economic variables and industry-related variables to train the model for existing or calculated carbon emissions. Since the acquisition of some economic variables is not timely and the simulated carbon emissions are prone to large errors, it is relatively unfeasible to use this method to accurately and real-time calculate carbon emissions. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for calculating regional carbon emissions based on power big data and time series, which can solve the problems of large errors and uncertainty in regional carbon emissions calculation in the existing technology.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] The present invention provides a method for calculating regional carbon emissions based on power big data and time series, comprising the following steps:

[0006] Obtain regional power big data and extract the power consumption of various industries in the region;

[0007] Conduct time series analysis on electricity consumption of various industries in the region, and divide the electricity consumption of each industry into long-term trend factors, seasonal cycle factors, cyclic factors and random factors;

[0008] The random factor variance of electricity consumption in each industry is used to calculate the joint confidence interval of electricity consumption in each industry, and the joint confidence interval of regional electricity consumption is obtained;

[0009] The regional carbon emission range is obtained by weighted calculation based on the joint confidence interval of regional electricity consumption and the industry carbon emission coefficient.

[0010] Furthermore, the time series analysis of the electricity consumption of each industry in the region includes performing additive decomposition of the time series of the electricity consumption of each industry in the region or performing multiplicative decomposition of the time series of the electricity consumption of each industry in the region, and the decomposition formula includes:

[0011] Additive Model for Time Series Decomposition: E i =T i +S i +C i +I i

[0012] Multiplicative Model for Time Series Decomposition: E i =T i ×S i ×C i ×I i

[0013] Among them, E i is the electricity consumption of the i-th industry; T i is the long-term trend factor decomposed from the electricity consumption of the i-th industry; S i is the seasonal cycle factor decomposed from the electricity consumption of the i-th industry; C i is the cycle factor decomposed from the electricity consumption of the i-th industry; I i It is the random factor decomposed from the electricity consumption of the i-th industry.

[0014] Furthermore, the formula for the joint confidence interval of the industry electricity consumption is as follows:

[0015] E i,min =f(T i ,S i ,C i ,I min ), E i,max =f(T i ,S i ,C i ,I max )

[0016] Among them, I max represents the maximum value of the random factor of electricity consumption in the i-th industry; I max Represents the minimum value of the interval of random factors of electricity consumption in the i-th industry.

[0017] Furthermore, the calculation formula of the joint confidence interval is as follows:

[0018]

[0019] Among them, μ I represents the mean of the random factor; Represents the mean vector of random factors decomposed from the time series; n represents the number of time points in the time series; S represents the covariance matrix of the decomposed random factors.

[0020] Furthermore, the calculation formula for the regional carbon emission range is as follows:

[0021]

[0022] Among them, C tota,lmin Indicates the minimum value of the interval of regional carbon emissions; C tota,lmax Indicates the maximum value of the interval of regional carbon emissions; C i,min represents the minimum value of the carbon emissions of the ith industry; C i,max represents the maximum value of the carbon emissions of the i-th industry; a i represents the carbon emission factor of the i-th industry; E i,min represents the minimum value of the electricity consumption of the i-th industry; E i,max Indicates the maximum value of the electricity consumption of the i-th industry.

[0023] Beneficial effects of the present invention:

[0024] 1. The present invention has the ability to calculate regional carbon emissions and the method is simple and efficient: based on power big data, the electricity consumption of each industry and residents in a region can be easily obtained, and the carbon emissions of each unit of electricity consumption in the corresponding industry can be obtained through existing data. At the same time, for methods that can calculate the carbon emissions of the entire region, the carbon emissions calculation method based on power big data is the most efficient while ensuring that the error is not large, and for the emission coefficient method for a single industry, the carbon emissions calculation method based on power big data can calculate the carbon emission capacity of the entire region.

[0025] 2. The present invention takes into account the uncertainty of energy consumption: the industry electricity consumption obtained from the power big data is regarded as time series data. The time series data contains four factors, namely long-term trend factors, seasonal cycle factors, cyclic factors, and random factors. The uncertainty of carbon emissions is introduced by the random factors of electricity consumption, so as to calculate the confidence interval of carbon emissions and ensure the accuracy and reliability of the carbon emissions calculation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 Schematic diagram of the steps for calculating regional carbon emissions based on electricity big data and time series. DETAILED DESCRIPTION

[0028] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0029] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0030] See also Figure 1 The present invention provides a method for calculating regional carbon emissions based on power big data and time series, comprising the following steps:

[0031] Step S1: Obtain regional power big data and extract the power consumption of each industry in the region;

[0032] Electricity big data is obtained from the front-end database, screened and processed accordingly, and then the electricity consumption of various related industries in the entire regional industry is extracted.

[0033] Step S2: Conduct a time series analysis on the electricity consumption of each industry in the region, and divide the electricity consumption of each industry into long-term trend factors, seasonal cycle factors, cyclic factors, and random factors;

[0034] Step S3: Calculate the joint confidence interval of electricity consumption of each industry using the variance of random factors of electricity consumption of each industry, and obtain the joint confidence interval of regional electricity consumption;

[0035] The 100(1-α)% joint confidence interval of the random factors is combined with the long-term trend factors, seasonal period factors, and cyclic factors obtained from the original time series to obtain the confidence interval of industry electricity consumption. The final regional carbon emission range is then obtained based on the following carbon emission coefficient calculation formula.

[0036] Step S4: Perform weighted calculation based on the joint confidence interval of regional electricity consumption and the industry carbon emission coefficient to obtain the regional carbon emission interval.

[0037] Specifically, the time series analysis of the electricity consumption of each industry in the region includes performing additive decomposition of the time series of the electricity consumption of each industry in the region or performing multiplicative decomposition of the time series of the electricity consumption of each industry in the region, and the decomposition formula includes:

[0038] Additive Model for Time Series Decomposition: E i =T i +S i +C i +I i

[0039] Multiplicative Model for Time Series Decomposition: E i =T i ×S i ×C i ×I i

[0040] Among them, E i is the electricity consumption of the i-th industry; T i is the long-term trend factor decomposed from the electricity consumption of the i-th industry; S i is the seasonal cycle factor decomposed from the electricity consumption of the i-th industry; C i is the cycle factor decomposed from the electricity consumption of the i-th industry; I i It is the random factor decomposed from the electricity consumption of the i-th industry.

[0041] In specific implementation, different models need to be used for different power consumption sequences. Assume that the random factors follow the normal distribution, that is, is the variance of the random factor.

[0042] Model decomposition methods include the moving average method, the ARIMA method, the X11 method, and the newly developed X12-ARIMA method. It is necessary to select the appropriate model and method based on the corresponding electricity consumption series to completely decompose the time series. The residual should conform to a normal distribution with a mean of 0.

[0043] For example, the moving average method uses a set of recent actual data values to average and smooth time series data. The moving average method can be divided into simple moving average and weighted moving average according to the weight of each element used. The formula of the moving average method is as follows:

[0044]

[0045] in, is the smoothed data of the jth period, ω j is the smoothing weight of each period, E j is the data before smoothing in period j. Based on the comparative analysis of the time series data before and after smoothing, the four factors of the time series are obtained.

[0046] The ARIMA model, also known as the autoregressive integrated moving average model (moving can also be called sliding), is one of the time series forecasting and analysis methods. For some non-stationary time series, after eliminating their local levels or trends, the remaining parts show a certain degree of homogeneity, that is, some parts of the series are very similar to other parts. This non-stationary time series can be converted into a stationary time series after being processed by the ARIMA method, and the random factors can be decomposed. The formula is:

[0047] E t =c+α1E t-1 +α2E t-2 +...+α p E t-p +ε t +θ1ε t-1 +θ2ε t-2 +...+θ q ε t-q

[0048] Among them, E t is the electricity consumption in period t, ε t is the random factor of period t, α,θ are the regression coefficients.

[0049] Specifically, the 100(1-α)% joint confidence interval of the random factors is combined with the long-term trend factors, seasonal period factors, and cyclic factors obtained from the original time series to obtain the confidence interval of industry electricity consumption. The formula for the joint confidence interval of industry electricity consumption is as follows:

[0050] E i,min =f(T i ,S i ,C i ,I min ), E i,max =f(T i ,S i ,C i ,I max )

[0051] Among them, I max represents the maximum value of the random factor of electricity consumption in the i-th industry; I max Represents the minimum value of the interval of random factors of electricity consumption in the i-th industry.

[0052] Specifically, the electricity consumption caused by random factors in each industry at different time points can be obtained. The random factors decomposed from the electricity consumption of the same industry can be used to estimate the mean of the random factors of the electricity consumption of that industry. If the random factors of all industries are regarded as vectors of a multivariate normal distribution, the covariance matrix of the multivariate normal distribution can be calculated using the electricity consumption of all industries at all times. After obtaining the mean vector and covariance matrix of the random factors of all industries, we can obtain the 100(1-α)% joint confidence interval of regional electricity consumption as follows. The calculation formula of the joint confidence interval is as follows:

[0053]

[0054] Among them, μ I represents the mean of the random factor; Represents the mean vector of random factors decomposed from the time series; n represents the number of time points in the time series; S represents the covariance matrix of the decomposed random factors.

[0055] Furthermore, the calculation formula for the regional carbon emission range is as follows:

[0056]

[0057] Among them, C tota,lmin Indicates the minimum value of the interval of regional carbon emissions; C tota,lmax Indicates the maximum value of the interval of regional carbon emissions; C i,min represents the minimum value of the carbon emissions of the ith industry; C i,max represents the maximum value of the carbon emissions of the i-th industry; a i represents the carbon emission factor of the i-th industry; E i,min represents the minimum value of the electricity consumption of the i-th industry; E i,max Indicates the maximum value of the electricity consumption of the i-th industry.

[0058] The focus of the present invention is to establish a data relationship between energy consumption variables and carbon emission results. The calculation method of regional carbon emissions based on power big data and time series can achieve this evaluation purpose. The calculation method of the present invention adopts the more mainstream emission coefficient method. The advantage of this method is that it requires fewer variables, the speed of obtaining results is relatively efficient, and the periodicity is short. Compared with the input-output method and machine learning algorithm, this method is more effective and can obtain the carbon emissions of the corresponding period in real time. At the same time, the energy consumption variables based on power big data in this embodiment are easy to obtain and cover a wide range of industries. It can easily obtain the electricity consumption of each industry and residents in a region, and can more comprehensively monitor all carbon emissions in the region, avoiding the inapplicability of the traditional emission coefficient method in the application scenario of multi-industry or even full-region carbon emission calculation. Finally, the time series analysis method is used to calculate the uncertainty in the energy consumption variable. The uncertainty of carbon emissions is introduced by decomposing the random factor of electricity consumption, thereby calculating the confidence interval of carbon emissions and ensuring the accuracy and reliability of the carbon emission calculation results. In summary, the calculation method of regional carbon emissions based on power big data and time series can introduce the uncertainty of carbon emissions for the calculation scenario of regional carbon emissions by using power big data and time series analysis, overcoming the problems of slow timeliness, cumbersome calculations, and incomplete coverage of traditional methods, thereby achieving accurate calculation of the carbon emissions range for the entire region.

[0059] The above is only for explaining the embodiments of the present invention and is not intended to limit the present invention. For those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present invention without creative work should be included in the scope of protection of the present invention.

Claims

1. A method for calculating regional carbon emissions based on power big data and time series, characterized by: The following steps are involved: Obtain regional power big data and extract the power consumption of various industries in the region; Conduct time series analysis on electricity consumption of various industries in the region, and divide the electricity consumption of each industry into long-term trend factors, seasonal cycle factors, cyclic factors and random factors; The random factor variance of electricity consumption in each industry is used to calculate the joint confidence interval of electricity consumption in each industry, and the joint confidence interval of regional electricity consumption is obtained; The regional carbon emission range is obtained by weighted calculation based on the joint confidence interval of regional electricity consumption and the industry carbon emission coefficient; The time series analysis of the electricity consumption of each industry in the region includes performing additive decomposition of the time series of the electricity consumption of each industry in the region or performing multiplicative decomposition of the time series of the electricity consumption of each industry in the region. The decomposition formula includes: Additive Model for Time Series Decomposition: E i =T i +S i +C i +I i Multiplicative Model for Time Series Decomposition: E i =T i ×S i ×C i ×I i Among them, E i is the electricity consumption of the i-th industry; T i is the long-term trend factor decomposed from the electricity consumption of the i-th industry; S i is the seasonal cycle factor decomposed from the electricity consumption of the i-th industry; C i is the cycle factor decomposed from the electricity consumption of the i-th industry; I i is the random factor decomposed from the electricity consumption of the i-th industry; The formula for the joint confidence interval of the industry electricity consumption is as follows: E i,min =f(T i ,S i ,C i ,I min ),E i,max =f(T i ,S i ,C i ,I max ) Among them, I max represents the maximum value of the random factor of electricity consumption in the i-th industry; I max represents the minimum value of the random factor of electricity consumption in the i-th industry; The calculation formula of the joint confidence interval of regional electricity consumption is as follows: Among them, μ I represents the mean of the random factor; Represents the mean vector of random factors decomposed from the time series; n represents the number of time points in the time series; S represents the covariance matrix of the decomposed random factors; The calculation formula for the regional carbon emission range is as follows: Among them, C total,min Indicates the minimum value of the interval of regional carbon emissions; C total,max Indicates the maximum value of the regional carbon emissions; Cmi i n represents the minimum value of the carbon emissions of the i-th industry; Cmax i represents the maximum value of the carbon emissions of the i-th industry; a i represents the carbon emission factor of the i-th industry; E i,min represents the minimum value of the electricity consumption of the i-th industry; E i,max Indicates the maximum value of the electricity consumption of the i-th industry.

Images