CN-122022846-A - Electric carbon coupling metering credibility assessment method

CN122022846ACN 122022846 ACN122022846 ACN 122022846ACN-122022846-A

Abstract

The invention discloses an electric carbon coupling measurement credibility assessment method, and relates to the technical field of electric carbon measurement. The method comprises the steps of firstly obtaining and preprocessing a high-frequency time sequence data sequence, including electric energy consumption, carbon emission and new energy output proportion sequence, further estimating dynamic marginal coupling strength of an electric-carbon relationship through local regression slope in a steady mode, calculating a stability index value based on the strength by utilizing a multi-scale coupling stability index algorithm, simultaneously combining a conditional direction consistency deviation algorithm, calculating a direction consistency deviation value according to an electric energy change direction and the new energy output proportion, and finally normalizing and fusing the two to obtain a local time-varying credibility score. The invention realizes the accurate depiction of the dynamic evolution of the electric carbon coupling strength, constructs a multi-scale stability evaluation and physical logic consistency discrimination mechanism, can effectively distinguish the decoupling of normal new energy and the deviation of abnormal metering, and improves the comprehensiveness and applicability of the electric carbon metering credibility evaluation.

Inventors

CHEN YUANYUAN
CHU JINGHUI
LI DAN
ZHANG DAO
WANG XIAOLI
ZHU YONG
Shao Fengdong
WANG YONGQING
LIU YONGAN
WU CHANGHAI
LI HONGXIN
XU YAN
ZHAO JIAXING

Assignees

烟台东方威思顿电气有限公司

Dates

Publication Date: 20260512
Application Date: 20260204

Claims (10)

1. An electrical carbon coupling metering reliability assessment method, comprising: S1, acquiring a high-frequency time sequence data sequence and performing data preprocessing to obtain a preprocessed high-frequency time sequence data sequence; S2, based on the preprocessed high-frequency time sequence data sequence, the dynamic marginal coupling strength of the electric-carbon relationship is estimated in a robust mode through a local regression slope; Step S3, calculating a multi-scale coupling stability index value through a multi-scale coupling stability index algorithm based on the dynamic marginal coupling strength; s4, based on the preprocessed high-frequency time sequence data sequence and the dynamic marginal coupling strength, introducing a conditional direction consistency deviation algorithm to calculate and obtain a conditional direction consistency deviation value; And S5, respectively carrying out extremum normalization processing on the multi-scale coupling stability index value and the conditional direction consistency deviation value to obtain a normalized multi-scale coupling stability index value and a normalized conditional direction consistency deviation value, and then obtaining a local time-varying credibility score based on the normalized multi-scale coupling stability index value and the normalized conditional direction consistency deviation value.
2. The method for evaluating reliability of electric-carbon coupling measurement according to claim 1, wherein in step S1, the obtained high-frequency time series data sequence comprises a consumption-side and/or regional-level electric energy consumption sequence, a corresponding carbon emission sequence and a new energy output ratio sequence, and the preprocessed high-frequency time series data sequence comprises a standardized electric energy consumption sequence, a standardized carbon emission sequence and a standardized new energy output ratio sequence.
3. The method for evaluating reliability of electric-carbon coupling measurement according to claim 1, wherein in step S2, the dynamic marginal coupling strength of the electric-carbon relationship is robustly evaluated by using the normalized electric energy consumption sequence and the normalized carbon emission sequence in the preprocessed high-frequency time series data sequence obtained in step S1 through the local regression slope.
4. The method for evaluating reliability of electrical-carbon coupling measurement of claim 3, wherein in step S2, the current time is selected A sliding window as the center In sliding window Respectively calculating the average value of the standardized electric energy consumption sequence and the standardized carbon emission sequence, and obtaining a regression slope by the ratio of the covariance to the electric energy variance, namely dividing the sum of products of the electric energy deviation and the carbon emission deviation in the sliding window by the square sum of the electric energy deviation, adding an extremely small positive number to prevent the denominator from being zero, thereby obtaining the current moment Dynamic marginal coupling strength of (a) 。
5. The method for evaluating reliability of electric-carbon coupling measurement according to claim 1, wherein in step S3, for each time instant Respectively select A sliding window of different length, each window length being noted as Wherein In the sliding window of each scale, firstly calculating the average value of all dynamic marginal coupling strengths in the window as a central reference under the scale, and the moment Is the first of (2) The average value of the dynamic marginal coupling strength of each sliding window is recorded as Then each dynamic marginal coupling strength is calculated point by point in each sliding window Average value of The square of the difference value obtained in each sliding window is averaged respectively and square root is taken to obtain the standard deviation form variation degree corresponding to each scale, and finally the variation degrees of different time scales are weighted and summed to obtain the current moment Multi-scale coupling stability index value of (2) 。
6. The method for evaluating the reliability of the electric carbon coupling measurement according to claim 1, wherein in the step S4, based on the standardized electric energy consumption sequence, the standardized carbon emission sequence and the standardized new energy output proportion sequence in the preprocessed high-frequency time sequence obtained in the step S1 and the dynamic marginal coupling strength calculated in the step S2, a conditional direction consistency deviation algorithm is introduced, and the conditional direction consistency deviation algorithm performs conditional branch judgment on the electric energy change direction at each moment and combines the new energy output proportion at the current moment to realize differential evaluation on the electric carbon coupling trend, so as to quantify whether the direction trend accords with the physical expectation and the new energy situation.
7. The method for evaluating reliability of electric-carbon coupling measurement of claim 6, wherein in step S4, a sliding window for evaluation is first determined In sliding window Traversing time by time, for each time Is a standardized power consumption increment of (2) Make positive and negative judgment if I.e. the power consumption increases between adjacent moments, a forward increment penalty branch is entered if I.e. the power consumption is reduced or basically unchanged, entering a negative increment punishment branch; entire sliding window The penalty values at all times in the time are accumulated and averaged to obtain the current time Conditional directional consistency deviation value of (2) 。
8. The method for evaluating reliability of electric carbon coupling measurement according to claim 7, wherein in said forward increment penalty branch, the output ratio is calculated according to new energy Weighting punishment is carried out; when the new energy proportion is low, the dynamic marginal coupling strength is inspected If the sign of (1) If the result is negative, the physical rule is considered to be seriously violated, a large penalty is performed, and the penalty amplitude is equal to that of Proportional to the ratio; When the new energy proportion is higher, punishment in an exponential decay form is introduced, and the exponential decay mechanism uses the fluctuation scale in a low new energy scene as a reference standard.
9. The method for evaluating reliability of electric carbon coupling measurement according to claim 7, wherein in said negative incremental penalty branch, according to the new energy output ratio Weighting punishment is carried out; When the new energy proportion is low, calculating the dynamic marginal coupling strength With low new energy standard Degree of deviation of (2) The larger the deviation, the larger the penalty; When the new energy proportion is higher, the wide tolerance dynamic marginal coupling strength Low case, punishment Too high.
10. The method for evaluating reliability of electric carbon coupling measurement according to any one of claims 1 to 9, wherein in step S5, the index value of multi-scale coupling stability calculated in step S3 is calculated And the conditional direction consistency deviation value calculated in the step S4 Respectively making extremum normalization treatment based on sliding time window, then taking arithmetic average value of two normalized index values, subtracting said arithmetic average value from 1 so as to obtain local time-varying reliability score 。

Description

Electric carbon coupling metering credibility assessment method Technical Field The invention relates to the field of electric carbon metering, in particular to an electric carbon coupling metering credibility evaluation method. Background Green low carbon conversion of energy systems has become an important direction of global interest driven by the "two carbon" target. As a basic guarantee to drive this transformation, accurate metering of carbon emissions is critical. Particularly, in the construction process of a novel power system, the power system and the carbon emission system are deeply fused to promote the concept of electric carbon coupling, and high integration and collaborative management of power data and carbon emission data are required. The core objective of the electric carbon coupling metering developed from the method is to realize accurate accounting and dynamic tracking of carbon emission by coupling the electric consumption data with the carbon emission factor, so that key support is provided for application scenes such as low-carbon scheduling, carbon asset management, carbon market transaction and the like. At present, the discussion of the reliability guarantee of an electric carbon metering system mainly takes over the thought of the traditional electric energy metering field, namely, the metering equipment or algorithm is calibrated and verified in a controlled environment by constructing a standard source and a tracing chain. For example, the Chinese patent of publication No. CN121008214A proposes a method for inspecting and tracing an electric power carbon emission metering system, which is characterized in that standard inputs such as power grid topology, unit output, carbon flow density and the like are preset in a laboratory or simulation environment, a tested electric carbon metering device is driven to operate, and an output result is compared with a standard electric carbon amount obtained by theoretical calculation, so that metering deviation of the device is evaluated. However, the conventional inspection method based on the controlled environment and the static standard input is difficult to cope with the complex dynamic characteristics of the actual grid operation and the carbon monitoring scene. The specific expression is as follows: First, there is a lack of sophisticated modeling and dynamic analysis capabilities for electrical carbon coupling relationships. The prior method can not describe the marginal response relation between the electric energy consumption and the carbon emission in real time based on the high-frequency time sequence data, and is difficult to capture and quantify the dynamic evolution process of the time-varying coupling strength. Second, the multiscale evaluation mechanism is missing, resulting in a coarser analysis of the coupling relationship fluctuations. The existing method usually evaluates on a single or fixed time scale, and cannot construct an effective multi-scale coupling stability index, so that local short-time unstable behaviors cannot be sensitively identified, and the robustness of long-term trends is difficult to evaluate from a macroscopic level, so that the perception capability on fluctuation risks is insufficient. Again, consideration of trend rationality and physical consistency is inadequate. The existing credibility assessment mainly focuses on the magnitude of the numerical value or fluctuation amplitude of the coupling strength, and neglects the logic consistency of the electric carbon change direction and the physical background such as new energy output. For example, in a thermal power dominated scenario, an increase in electricity usage generally expects a synchronous rise in carbon emissions, while a higher proportion of new energy is used for electricity usage and carbon emissions, an increase in electricity usage does not mean a comparable increase in carbon emissions. The existing method lacks an effective discrimination mechanism for the direction consistency, and is difficult to identify trend deviation caused by data abnormality or physical logic violation. Finally, under the background of new energy permeation, the decoupling trend lacks a credibility discrimination mechanism. The relation between the fluctuation and uncertainty of the new energy output cause the electric energy consumption and the carbon emission to present the non-linear even decoupling characteristic, and the traditional evaluation model based on the linear or static assumption is difficult to effectively distinguish whether the normal decoupling phenomenon is caused by the clean energy output or the abnormal decoupling caused by the metering deviation and the data abnormality, so that the reliability evaluation mechanism of the traditional method loses effectiveness under the high-proportion new energy scene. Therefore, the requirement of comprehensively and accurately evaluating the reliability of the electric carbon coupling measuremen