Method and system for tracing errors in carbon emission flow metering of power systems

Abstract

The present application relates to the technical field of electric carbon emission, in particular to a power system carbon emission flow metering error traceability method and system, the method comprises: based on the topological relation of power system, constructing the carbon emission factor calculation model of each node, and determining the upstream and downstream dependence relation among nodes; according to the error source of distributed metering system, the relative uncertainty of source side metering equipment error and data transmission error among nodes is quantified; according to the error transmission rule, the relative uncertainty of carbon emission factor of each node from source side to user side is calculated step by step, the global error propagation path is generated, and the overall relative uncertainty of user side carbon emission is calculated; through the disturbance analysis of power grid power data by probability statistical simulation method, the abnormal node is identified. The purpose is to realize the error traceability of carbon emission metering, and avoid the decrease of carbon emission metering accuracy caused by node error in power system.

Description

Technical Field

[0001] This invention relates to the field of carbon emission technology, specifically to a method and system for tracing the source of measurement errors in carbon emission flows of power systems. Background Technology

[0002] To accurately calculate carbon emissions from the power system and allocate carbon emission responsibility based on each user's electricity consumption, power system carbon emission metering has become a key focus in research on the low-carbon operation of power systems. Currently, various methods exist for power system carbon emission metering, but most are centralized methods. The computational load and time required increase dramatically with the expansion of the power grid, leading to a decline in real-time performance. To address this issue, distributed carbon emission flow iterative algorithms are gradually developing. However, as a crucial component of carbon emission metering, its error analysis and source tracing methods need to be clearly defined to ensure reliable operation in actual power grids.

[0003] Current distributed metering methods for carbon emission flows are mainly developed in laboratory environments, and have shortcomings in both metering error tracing and error analysis. In terms of error tracing, existing technologies do not clearly identify the errors in each stage of distributed metering, and cannot locate the source of errors in distributed iteration in a real operating environment, leading to error accumulation that affects metering accuracy. In terms of error analysis, there is a lack of error analysis combined with carbon flow metering hardware, resulting in a lack of support when theoretical algorithms are applied to actual hardware. Corresponding error analysis techniques need to be developed for actual hardware. Summary of the Invention

[0004] To achieve error tracing and online calibration in distributed carbon emission metering and improve the accuracy of carbon emission metering, this invention provides a method and system for error tracing in power system carbon emission flow metering. The specific technical solution adopted is as follows:

[0005] The first aspect of the present invention provides a method for tracing the source of measurement errors in carbon emission streams of power systems, the method comprising:

[0006] Based on the topology of the power system, a carbon emission factor calculation model for each node is constructed, and the upstream and downstream dependencies between nodes are determined.

[0007] Based on the error sources of the distributed metering system, the relative uncertainty of the source-side metering equipment error and the data transmission error between nodes is quantified;

[0008] According to the error propagation rule, the relative uncertainty of carbon emission factors at each node from the source side to the user side is calculated step by step to generate a global error propagation path, and the overall relative uncertainty of carbon emissions on the user side is calculated.

[0009] Pertactical simulation methods are used to analyze disturbances in power grid data and identify anomalous nodes.

[0010] Furthermore, based on the power system topology, a carbon emission factor calculation model for each node is constructed, and the upstream and downstream dependencies between nodes are determined, including:

[0011] The carbon emission factor of a node is defined as the indirect carbon emission corresponding to a unit of electricity consumption on the load side, and the carbon flow density of a branch is the carbon emission corresponding to a unit of electricity consumption on the branch.

[0012] Based on the power flow distribution, the carbon emission factor transmission relationship from the upper-level node to the lower-level node is determined by recursive calculation, where the transmission weight is the proportion of active power of each branch.

[0013] Furthermore, based on the error sources of the distributed metering system, the relative uncertainties of the source-side metering equipment errors and inter-node data transmission errors are quantified, including:

[0014] Calculate the relative uncertainty of the source-side carbon emission factor based on the standard uncertainty parameters of the metering equipment;

[0015] Calculate the relative uncertainty of data transmission error between nodes based on the communication truncation error in data transmission between nodes.

[0016] Furthermore, based on the standard uncertainty parameters of the metering equipment, the relative uncertainty of the source-side carbon emission factor is calculated, including:

[0017] Metering equipment is used to obtain source-side carbon dioxide emissions and active power generation, and the source-side carbon emission factor is calculated.

[0018] Based on the standard uncertainty of the measuring equipment, the relative uncertainties of carbon dioxide and active power are calculated respectively.

[0019] According to the law of error propagation, the relative uncertainties of carbon dioxide and active power are combined into the relative uncertainty of the source-side carbon emission factor.

[0020] Furthermore, based on the communication truncation error in inter-node data transmission, the relative uncertainty of inter-node data transmission error is calculated, including:

[0021] Calculate communication error based on the data truncation function of the communication algorithm;

[0022] The relative uncertainty of communication error between nodes is calculated based on the fluctuation range characteristics of the transmitted data.

[0023] Furthermore, based on the error propagation rule, the relative uncertainty of the carbon emission factor at each node from the source side to the user side is calculated level by level, including:

[0024] Calculate the carbon emission factors of each node based on the power and carbon emission factor data of the upper-level node;

[0025] Based on the relative uncertainties of the active power and carbon emission factors of each branch and the correlation coefficient, the relative uncertainty of the carbon emission factors of the lower-level nodes without considering communication errors is calculated.

[0026] Furthermore, a global error propagation path is generated, and the overall relative uncertainty of user-side carbon emissions is calculated, including:

[0027] The relative uncertainties of carbon emission factor and active power are corrected based on the relative uncertainty of data transmission error between nodes;

[0028] The total relative uncertainty of user-side carbon emissions is calculated based on the corrected carbon emission factor and the relative uncertainty of active power.

[0029] Furthermore, perturbation analysis of power grid data is performed using probabilistic statistical simulation methods to identify anomalous nodes, including:

[0030] The power data of each node in the power grid topology are randomly sampled using a normal distribution;

[0031] Carbon emission factor is iteratively calculated from the sampled data to generate a carbon emission factor distribution;

[0032] The dispersion of carbon emission factors at each node is statistically analyzed, the actual uncertainty is calculated and compared with the total relative uncertainty, and abnormal nodes are marked according to the deviation threshold.

[0033] Furthermore, the method also includes:

[0034] Based on the error impact weight of abnormal nodes and the power system hierarchy, a calibration priority is generated;

[0035] Calibration coefficients are generated for abnormal nodes, and the calibration coefficients are dynamically adjusted using a smooth transition algorithm.

[0036] The second aspect of the present invention provides a power system carbon emission flow metering error tracing system, employing the power system carbon emission flow metering error tracing method described in the first aspect of the present invention. The system includes:

[0037] The carbon emission flow metering network model is configured to be based on the power system topology, constructing a carbon emission factor calculation model for each node and determining the upstream and downstream dependencies between nodes;

[0038] The relative uncertainty module is configured to quantify the relative uncertainty of source-side metering equipment errors and inter-node data transmission errors based on the error sources of the distributed metering system.

[0039] The error propagation module is configured to calculate the relative uncertainty of carbon emission factors at each node from the source side to the user side according to the error propagation rules, generate a global error propagation path, and calculate the overall relative uncertainty of carbon emissions on the user side.

[0040] The error detection module is configured to perform disturbance analysis on power grid data using probabilistic statistical simulation methods to identify abnormal nodes.

[0041] The calibration strategy generation module is configured to generate calibration strategies based on abnormal nodes.

[0042] The calibration execution module is configured to perform calibration operations according to the calibration strategy.

[0043] The present invention has the following beneficial effects:

[0044] The present invention provides a method and system for tracing errors in power system carbon emission flow metering. By constructing a node carbon emission factor model and upstream / downstream dependencies, it identifies and quantifies the relative uncertainty of data transmission errors between source-side metering equipment and nodes, establishing a core framework for error tracing. Furthermore, based on error propagation rules, it generates a global error propagation path level by level, achieving full-network tracking of carbon emission factor uncertainty and accurate assessment of overall user-side errors, effectively preventing the gradual propagation of single-node errors to lower-level nodes. Combined with probabilistic statistical simulation methods, it performs disturbance analysis on grid power data, dynamically identifying abnormal nodes and forming a closed-loop error location and calibration mechanism. This method, through full-process error tracing and quantitative analysis, significantly improves the reliability and accuracy of carbon emission metering, providing highly reliable data support for the low-carbon operation of power systems. Attached Figure Description

[0045] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart of a method for tracing the source of carbon emission flow measurement errors in a power system, provided in an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram of the communication layer architecture of a distributed carbon metering network provided in one embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of the structure of a power system carbon emission flow metering error tracing system provided in an embodiment of the present invention. Detailed Implementation

[0049] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method and system for tracing carbon emission flow measurement errors in a power system according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0051] This invention considers a power system carbon emission metering scenario using a distributed carbon emission flow iterative algorithm, addressing the error tracing problem in carbon emission flow metering in conjunction with metering equipment. A carbon emission flow is defined as a virtual network flow that exists in relation to power flow and is used to characterize the carbon emissions maintaining any branch of the power flow in the power system. The carbon emission flow adds a carbon emission attribute to all power flow in the power system. Following the power flow, the carbon emission flow originates from the generation side and flows into the power system, passing through various nodes on the grid side before finally entering the load nodes on the user side. Therefore, the actual carbon dioxide emissions emitted into the atmosphere by the generation side can be distributed to each load node, ultimately calculating the carbon emissions consumed by the power user.

[0052] According to the fundamental theory of power grid carbon metering, carbon emission flows are dependent on the power flow in the power grid. The power flow in the power system is distributed across the power grid from the generation side to the consumption side, and the power grid consists of a series of nodes and branches. Therefore, calculating carbon emission flows requires first determining the power flow distribution of the power system. However, for a large power grid, calculating its power flow distribution requires solving massive matrix operations. Therefore, the algorithm needs to be simplified to accommodate the lower computational capabilities of distributed metering systems.

[0053] When starting from a decentralized approach, solving large sparse matrices can be simplified using recursive methods. Considering that carbon emission flows in a power system are coupled to power flow, and that power flow is influenced by nodes and branches, two parameters can be derived: the node carbon emission factor and the branch carbon flow density. The node carbon emission factor describes the carbon emissions per unit of electricity consumed at a node, while the branch carbon flow density describes the carbon emissions per unit of electricity flowing through a branch. Both correspond to the indirect carbon emissions associated with consumed electricity; the difference is that the node carbon emission factor indicates the electricity consumed by the load, while the branch carbon flow density indicates line losses.

[0054] Based on this, and according to the principle of fairness in carbon emission flows, the carbon flow density of all outflow branches connected to a node is numerically equal to the carbon emission factor of that node. Therefore, calculating carbon emission flows should begin with calculating the carbon emission factors of each node.

[0055] The following description, in conjunction with the accompanying drawings, details the specific scheme of the method and system for tracing the measurement error of carbon emission flow in a power system provided by the present invention.

[0056] Please see Figure 1 The diagram illustrates a flowchart of a method for tracing the source of carbon emission flow measurement errors in a power system according to an embodiment of the present invention. The method includes:

[0057] Step S100: Based on the power system topology, construct a carbon emission factor calculation model for each node and determine the upstream and downstream dependencies between nodes;

[0058] Step S100 specifically includes:

[0059] Step S110: Define the carbon emission factor of a node as the indirect carbon emission per unit of electricity consumption on the load side, and the carbon flow density of a branch as the carbon emission per unit of electricity consumption on the branch. Specifically, in this embodiment, the carbon emission factor of a node is defined as the indirect carbon emission per unit of electricity consumption on the load side; simultaneously, the carbon flow density of a branch is defined as the carbon emission per unit of electricity consumption on the branch, used to measure the carbon emissions generated during electricity transmission on the branch. The branch carbon flow density can reflect the carbon emissions caused by factors such as line losses.

[0060] Step S120: Based on the power flow distribution, the carbon emission factor transfer relationship from the upper-level node to the lower-level node is determined through recursive calculation, where the transfer weight is the proportion of active power in each branch. Specifically, power flow distribution data of the power system is obtained, which describes the distribution of electrical quantities such as voltage and power in each node of the power system. The power flow distribution is obtained through real-time monitoring data, historical operating data, or calculations based on the power system model. Based on the power flow distribution, the carbon emission factor transfer relationship between nodes is determined through recursive calculation. Starting from the generation-side node, for each lower-level node, the active power injected by its upper-level node through each branch is analyzed. Since the magnitude of the active power in each branch reflects the contribution of the upper-level node to the power supply of the lower-level node, the proportion of active power in each branch is used as the weight of the carbon emission factor transfer. Through this recursive calculation method, the carbon emission factor transfer relationship between nodes is determined step by step from the generation side to the user side, thereby constructing a complete carbon emission factor calculation model that clearly reflects the flow and transfer pattern of carbon emissions between nodes in the power system.

[0061] Step S200: Based on the error sources of the distributed metering system, quantify the relative uncertainty of the source-side metering equipment error and the data transmission error between nodes. Specifically, in a distributed metering system, all metering nodes require the calculation results of the upper-level nodes as input. Therefore, when any node in the metering network generates an error, this error will propagate to the lower-level nodes level by level through the iterative network, causing all lower-level nodes to be affected by the single-point failure of that point, resulting in a decrease in metering accuracy. The purpose of this invention is to quantitatively calculate the magnitude of the impact of an error generated by any node in a distributed metering system on the accuracy of lower-level nodes, and this impact is expressed as relative uncertainty.

[0062] Step S200 specifically includes:

[0063] Step S210: Calculate the relative uncertainty of the source-side carbon emission factor based on the standard uncertainty parameter of the metering equipment. Specifically, use the metering equipment to obtain the source-side carbon dioxide emissions and active power generation, and calculate the source-side carbon emission factor. Since the carbon flow is transmitted from the source side to the user side, the initial value of the carbon emission factor iteration is the actual carbon emission factor of the source-side power plant. On the source side, the carbon dioxide emissions over a certain period are measured by sensors placed at the carbon dioxide emission outlet of the generator unit. Simultaneously, the active power generation during this period is collected using the power plant's gate meter. Since there are no upstream nodes on the source side, the source-side carbon emission factor can be expressed as:

[0064]

[0065] In the formula, e gnt Indicates the source-side carbon emission factor; C gnt E represents source-side carbon dioxide emissions. gnt This represents the active power generated at the source; among which, for carbon dioxide emissions C gnt The uncertainty in the measurement mainly stems from the carbon dioxide sensor. The standard uncertainty can be obtained by consulting the technical specifications of the carbon dioxide sensor. This parameter reflects the dispersion of the carbon dioxide sensor's measurement results during normal operation, indicating the range of deviation of the measured value from the true value; for active power generation E... gnt The measurement is performed using an electricity meter, whose error mainly comes from internal ADC conversion error and sensor error. Refer to the electricity meter's technical specifications to obtain the standard uncertainty of the measurement value during normal operation, without considering internal faults in the electricity meter.

[0066] Based on the standard uncertainty of the measuring equipment, the relative uncertainties of carbon dioxide and active power are calculated separately. Specifically, to eliminate the influence of dimensions and more intuitively reflect the relative dispersion of carbon dioxide emission measurements and the relative uncertainty of active power measurements, the relative standard uncertainty of carbon dioxide emissions is calculated. It can be represented as Relative standard uncertainty of active power generation It can be represented as

[0067] According to the law of error propagation, the relative uncertainties of carbon dioxide and active power can be combined into the relative uncertainty of the source-side carbon emission factor, which can be expressed as:

[0068]

[0069] In the formula, u rel_egnt Indicates the source-side carbon emission factor e gnt The relative uncertainty is used to quantify the amount of carbon dioxide emissions C. gnt and active power generation E gnt The uncertainty of carbon emission factors is caused by the combined uncertainty of these factors. This represents the partial derivative of the source-side carbon emission factor with respect to carbon dioxide emissions. This represents the partial derivative of the source-side carbon emission factor with respect to active power generation; The standard uncertainty of a carbon dioxide sensor; This represents the standard uncertainty of the power sensor. Based on actual measurement data and the characteristics of metering equipment, this embodiment systematically analyzes and integrates the measurement uncertainties of source-side carbon dioxide emissions and active power generation, providing source-end data support for tracing the source of carbon emission flow metering errors in the power system. By calculating the relative uncertainty of the source-side carbon emission factor, the accuracy of carbon emission metering on the generation side can be understood. This helps power system operation managers and researchers to fully grasp the source characteristics of carbon emission metering errors, thereby enabling targeted measures to be taken in subsequent error calibration and system optimization processes to improve the overall accuracy and reliability of power system carbon emission flow metering.

[0070] Step S220: Calculate the relative uncertainty of data transmission error between nodes based on the communication truncation error in data transmission between nodes; see details below. Figure 2As shown, in the distributed carbon emission metering system, smart meters serve as carbon metering nodes. Carbon emission flow is metered through distributed communication between smart meters. Nodes communicate using the national standard 698.45 protocol. During transmission, all floating-point data is transmitted in fixed-point form, which introduces some error. Therefore, it is necessary to conduct source analysis of this communication error. During data transmission, a propagation strategy is used for floating-point numbers; that is, when the floating-point number is represented as ±ab, the actual transmitted data is ab×10. n′ , where n′ is the number of decimal places, which is determined according to the required floating-point precision and numerical range, so that floating-point data can be transmitted in fixed-point form;

[0071] Specifically, this embodiment only considers the error caused by the truncation of floating-point numbers in the communication protocol, excluding the influence of factors such as communication network failures and packet loss. The Monte Carlo method is used to calculate the floating-point arithmetic error of the digitized value. This method simulates floating-point arithmetic operations through a large number of random samples, statistically analyzes the error distribution, and then obtains the uncertainty. According to the communication protocol, when the number of decimal places of the fixed-point number is n, the communication error can be expressed as:

[0072] e t =T RC (X t )-X t ′

[0073] Among them, T RC (X t ) represents the data received by the receiving end, and the data after being truncated by the communication algorithm; X t′ This represents the floating-point data of the carbon emission factor sent by the sending end, i.e., the original data to be transmitted; T RC This represents the data truncation function of the communication algorithm. According to the communication protocol, this function retains only n′ decimal places and discards any excess. t This indicates a communication error, reflecting the difference between the received data and the original transmitted data caused by the truncation operation of the communication protocol;

[0074] Based on the fluctuation range characteristics of the transmitted data, the relative uncertainty of the communication error between nodes is calculated; specifically, due to the characteristics of the truncation function, the transmitted data X... t The actual value will be in the range [T] RC (X t ),T RC (X t )+X min Fluctuating between ], where X min The least significant bit, as defined by the communication protocol, determines the minimum possible change in error after data truncation; based on this, the relative uncertainty of the communication error can be expressed as:

[0075]

[0076] In the formula, u t This represents the relative uncertainty of communication errors, used to quantify the dispersion of errors introduced by data truncation during communication. A smaller value indicates more stable communication errors and smaller fluctuations. It should be noted that, considering the transmission process from the source to the user, nodes may use different communication methods in different regions, and the communication protocol may also change accordingly. Therefore, the uncertainty of communication errors should be adjusted according to the specific circumstances and the corresponding X value. min Perform calculations;

[0077] This embodiment employs the Monte Carlo method for error simulation and analysis, accurately quantifying the uncertainty of communication errors. It not only clarifies the impact of communication on the accuracy of carbon emission flow measurement but also provides crucial data support for subsequent error tracing and calibration. In practical applications, this helps power system operators and researchers clearly understand the error characteristics during data transmission, enabling them to optimize communication protocols and improve data transmission strategies, reducing the interference of communication errors on carbon emission measurement results and significantly improving the overall accuracy and reliability of power system carbon emission flow measurement.

[0078] Step S300: Based on the error propagation rule, calculate the relative uncertainty of carbon emission factors at each node from the source side to the user side, generate a global error propagation path, and calculate the overall relative uncertainty of carbon emissions on the user side. Specifically, based on the fundamental theory of power grid carbon metering, carbon emission flow is dependent on the power flow in the power grid, and the power system consists of a series of nodes and branches. To address the complexity of power flow calculation in large power grids, a recursive method is adopted to simplify the solution of large sparse matrices, starting from a decentralized approach, and introducing two key parameters: "node carbon emission factor" and "branch carbon flow density." The "node carbon emission factor" reflects the carbon emissions corresponding to a unit of electricity consumption, while the "branch carbon flow density" reflects the carbon emissions corresponding to a unit of electricity flowing through a branch. Based on this, according to the principle of fairness in carbon emission flow, the carbon flow density of all outgoing branches connected to a node is equal to the carbon emission factor of that node, and the carbon flow density of any branch injected into a node is equal to the carbon emission factor of the node it is injected into, providing a theoretical basis for subsequent calculations.

[0079] For node i, according to the proportional allocation principle, the carbon emission factor calculation formula can be expressed as:

[0080]

[0081] In the formula, e i Represents the carbon emission factor of node i; This represents the set of branches that inject active power into node i; P represents the set of generator sets connected to node i; b ρ represents the active power of branch b; b G represents the carbon flux density of branch b; s e represents the active power of generator set s; s Let represent the carbon emission factor of generator set s. Based on this formula, and combined with the known information of the upper-level nodes (power and carbon emission factor), starting from the source side, the carbon emission factor of each level of node is calculated step by step using a recursive method.

[0082] Step S300 specifically includes:

[0083] Step S310: Calculate the carbon emission factors of each level of node based on the power and carbon emission factor data of the upper-level node;

[0084] Specifically, for node n+1, its carbon emission factor can be expressed as:

[0085]

[0086] In the formula, e n+1 This represents the carbon emission factor calculated at the (n+1)th level node, where n represents the level number of the node, which reflects the hierarchical relationship of the nodes in the system; P i Represents the active power transmitted in the i-th branch; e n_i represents the carbon emission factor of the i-th branch; m represents the number of branches;

[0087] Step S320: Based on the relative uncertainties of the active power and carbon emission factors of each branch, and the correlation coefficient, calculate the relative uncertainty of the carbon emission factors of the lower-level nodes without considering communication errors; since the active power of each branch is independent and the carbon emission factors are correlated, the technical formula for its relative uncertainty can be expressed as:

[0088]

[0089] In the formula, This represents the relative uncertainty of the carbon emission factor at the (n+1)th level node, used to measure the degree of uncertainty in the calculation results of the carbon emission factor at that node; This represents the standard uncertainty of the power of the j-th branch, reflecting the uncertainty of the active power measurement of the j-th branch; The standard uncertainty of the carbon emission factor for the j-th branch is represented by r(e). n_k ,e n_j ) represents the correlation coefficient between the carbon emission factors of the k-th and j-th branches, which is used to describe the degree of correlation between the carbon emission factors of the two branches, and the value ranges from [-1,1]. This represents the relative uncertainty of the active power of the j-th branch; This represents the relative uncertainty of the carbon emission factor for the j-th branch;

[0090] Substituting the partial derivatives and simplifying, we get:

[0091]

[0092] The above formula is a quantitative calculation of the impact of error on the next node during the propagation of error from the previous node to the next node; this formula is derived by e n+1 Regarding P h and e n_j Calculate the partial derivatives, and combine the uncertainties of the active power and carbon emission factors of each branch with the correlation coefficient r(e) between them. n_k ,e n_h This quantifies the error impact of the node caused by the uncertainty of the parameters of the upper-level node.

[0093] Step S330: Correct the relative uncertainties of carbon emission factor and active power based on the relative uncertainty of data transmission errors between nodes; the overall uncertainty of the distributed carbon emission flow metering system; specifically, for each level of node, messages are transmitted through communication lines, generating communication errors. These errors are independent of other errors and are used in the calculation of carbon emission factor and active power from all superior nodes. and The standard uncertainty after considering communication errors should be respectively

[0094]

[0095] in, This represents the relative uncertainty of the carbon emission factor of the j-th branch after considering communication errors; u rel_meter_P The standard uncertainty of the electricity meter is expressed and calculated based on the data in the electricity meter's technical specifications. This represents the relative uncertainty of the active power of the j-th branch after considering communication errors;

[0096] Step S340: Calculate the total relative uncertainty of user-side carbon emissions based on the corrected carbon emission factor and the relative uncertainty of active power;

[0097] On the final load side, when calculating the total carbon emissions from user consumption, it is necessary to combine the carbon emission factor considering communication errors and the relative uncertainty of active power to calculate the final relative uncertainty of carbon emissions:

[0098]

[0099] In the formula, u rel_CThis represents the relative uncertainty of the total carbon emissions consumed by the final load-side users, comprehensively reflecting the degree of influence of various errors throughout the calculation process on the final carbon emission calculation result; u rel_e This represents the relative uncertainty of the carbon emission factor after considering communication errors; u rel_P This indicates the relative uncertainty of active power after considering communication errors.

[0100] In summary, for a given topology, when calculating the theoretical error magnitude of each node, it is necessary to propagate the error down level by level from the source side, calculating the relative uncertainty of each node separately until reaching the load side, thus obtaining a quantitative result of the theoretical error of any load-side node. This embodiment achieves comprehensive source tracing and quantification of carbon emission flow metering errors in the power system through systematic error propagation calculation and analysis. Based on the power system topology and carbon emission flow characteristics, a recursive approach is used to efficiently calculate the carbon emission factors at each level of the node; considering the independence of active power in branches, the correlation of carbon emission factors, and communication errors, the relative uncertainty of each node is accurately quantified, and a global error propagation path is generated. This enables power system operators to clearly understand the propagation patterns and impact of carbon emission metering errors in the system, effectively improving the accuracy and reliability of the distributed carbon emission flow metering system.

[0101] Step S400: Disturbance analysis of power grid data is performed using probabilistic statistical simulation methods to identify abnormal nodes; specifically, this embodiment uses the Monte Carlo method to perform disturbance analysis of power grid data to identify abnormal nodes; the system mainly has two types of error sources, namely data transmission error and active power measurement error;

[0102] Step S400 specifically includes:

[0103] Step S410: Perform normal distribution random sampling of the power data of each node in the power grid topology; specifically, with a period of 6 hours, perform random sampling of the power data of all nodes in the power grid topology, and perform 10,000 simulation calculations in each period. The high frequency of sampling and the large number of simulations ensure that various possible error combinations can be covered; at the same time, during the sampling process, random errors conforming to normal distribution are introduced into the sampled data according to the accuracy level of the electricity meter to ensure the authenticity of the simulation.

[0104] Step S420: Perform iterative calculation of carbon emission factors on the sampled data to generate carbon emission factor distribution; specifically, the sampled data with random errors is used to perform iterative calculation of carbon emission flow through actual hardware equipment to simulate the operation scenario of a real power system, so that the calculation results are closer to the actual situation.

[0105] Step S430: Statistically analyze the dispersion of carbon emission factors at each node, calculate the actual uncertainty and compare it with the total relative uncertainty, and mark abnormal nodes according to the deviation threshold; specifically, the system collects the results of each iteration calculation to obtain the statistical distribution of carbon emission factors for all nodes. These data reflect the fluctuation range and probability distribution of carbon emission factors under the influence of simulation errors. For each node, the standard uncertainty of the carbon emission factor is calculated according to the following formula:

[0106]

[0107] Where, u(CF) i ) represents the actual uncertainty of the carbon emission factor at node i, CF i,j Let CF be the carbon emission factor value of the j-th sample. i,avg Here, n is the average value, and n″ is the number of samples. This formula is based on the principle of sample standard deviation calculation in statistics. By measuring the deviation of the sampled values ​​from the average value, it quantifies the uncertainty of the carbon emission factor at each node. The overall relative uncertainty calculated by the theoretical formula in steps S200 and S300 is compared with the actual uncertainty calculated experimentally in this step. The relative uncertainty between the theoretical true value and the actual calculated value of the carbon emission factor at each node is analyzed. When the deviation between the two exceeds a set threshold, the node is determined to be an abnormal node.

[0108] This embodiment uses the Monte Carlo method to analyze disturbances in power grid data, achieving practical detection of carbon emission flow measurement errors and accurate identification of abnormal nodes in the power system. Extensive random sampling and hardware-in-the-loop computation simulate errors in a real operating environment, making the calculation results more relevant and valuable. By comparing the experimental calculation uncertainty with the theoretical calculation results, nodes where measurement errors exceed expectations can be quickly located, effectively compensating for the limitations of theoretical analysis in comprehensively covering complex real-world situations. In practical applications, this helps to promptly identify and resolve potential problems in the carbon emission measurement process, improve the reliability of carbon emission data in the power system, provide strong data support for carbon emission management, carbon trading, and energy conservation and emission reduction policy formulation, and promote the development of a more refined and intelligent low-carbon operation model in the power industry.

[0109] Preferably, the method for tracing the source of carbon emission flow measurement errors in the power system includes: Step S500: performing graded calibration based on the impact degree of abnormal nodes; specifically including:

[0110] Step S510: Generate calibration priorities based on the error impact weights of abnormal nodes and the power system hierarchy. Specifically, firstly, select the first node on the generation side as the calibration reference point, and fix the calibration coefficient of this node at 1.00, serving as the basis for the entire calibration process and ensuring its consistency and accuracy. Continuously monitor the carbon emission factor uncertainty of each node. When the deviation of the carbon emission factor uncertainty of any node from the theoretical value exceeds ±0.5%, the calibration process is triggered. This threshold is set based on the requirements for system metering accuracy and the assessment of actual error conditions, allowing for the timely identification of nodes requiring calibration.

[0111] Step S520: Generate calibration coefficients for abnormal nodes and dynamically adjust the calibration coefficients using a smooth transition algorithm; specifically, generate calibration coefficients for problem nodes that are detected and require calibration, and the calculation rules are as follows:

[0112] Calibration coefficient K = 1 + (actual measured value - theoretical calculated value) / theoretical calculated value

[0113] The actual measured value is the node carbon emission factor value obtained through actual measurement by the system; the theoretical calculated value is the node carbon emission factor value calculated based on the theoretical model established in steps S200 and S300; the formula calculates the deviation ratio between the actual measured value and the theoretical calculated value, and adds 1 to it to obtain the calibration coefficient used to adjust the measured value.

[0114] Then, calibration is carried out step by step in the order of "generation side → transmission side → distribution side → user side". The calibration of the upper-level node is completed first. The lower-level node is calibrated only after the calibration of the upper-level node is completed. This is because the carbon emission factor of the lower-level node may be affected by the upper-level node. Only when the upper-level node is accurately calibrated can the effectiveness of the calibration of the lower-level node be guaranteed. Each node is verified immediately after calibration to check whether the carbon emission factor of the calibrated node meets the expectations. If it does not meet the expectations, the calibration is repeated.

[0115] After completing the full network calibration, rerun the Monte Carlo error analysis in step S400 to obtain the standard uncertainty u before and after calibration. 校准前 and u 校准后 Calculate the calibration effect evaluation index η 校准 , can be represented as:

[0116] η 校准 =(u 校准前 -u 校准后 ) / u 校准前 ×100%

[0117] This indicator reflects the proportion of change in standard uncertainty before and after calibration, intuitively demonstrating the calibration effect. A calibration validity threshold is set, and when the evaluation index η... 校准If the value exceeds this threshold, the calibration is considered valid; otherwise, the calibration process needs to be rechecked or the calibration parameters adjusted.

[0118] Meanwhile, to ensure real-time performance during online calibration, this embodiment also designs a three-level calibration strategy to ensure that the calibration process does not affect the normal operation of the system, as follows:

[0119] Emergency calibration: When a node error is detected to exceed a preset threshold, emergency calibration is immediately triggered, prioritizing the calibration of that node to ensure system stability and measurement accuracy;

[0120] Routine calibration: The system performs a comprehensive calibration of all nodes according to a preset cycle (default 24 hours) to ensure the accuracy of the system's long-term operation;

[0121] Deep calibration: Perform full-system deep calibration during periods of low grid load (such as 2:00-4:00 AM). At this time, the grid is relatively stable, which minimizes interference with the calibration process and allows for more accurate calibration.

[0122] To avoid data jumps during the calibration process, this embodiment adopts a smooth transition method, which can be expressed as:

[0123] CF 输出 (t)=CF 原始 (t)×K 平滑 +CF 校准 (t-1)×(1-K 平滑 )

[0124] In the formula, CF 输出 (t) represents the smoothed carbon emission factor value output at time t; CF 原始 (t) represents the original carbon emission factor measurement at time t; CF 校准 (t-1) represents the calibrated carbon emission factor value at time t-1; K 平滑 This represents the smoothing coefficient, with a default value of 0.1, which can be adjusted according to the system calibration frequency. This formula achieves a smooth data transition during the calibration process by weighted averaging of the original measured value and the calibration value at the previous moment.

[0125] This embodiment, based on the source tracing and analysis results of metering errors in power system carbon emission flows, implements a dynamic and efficient calibration strategy. By establishing calibration benchmarks, real-time error detection, and parameter generation, accurate calibration can be performed on problematic nodes. A tiered calibration and closed-loop verification mechanism ensures the systematic nature and effectiveness of the calibration process, guaranteeing the accuracy and reliability of the calibration results. Emergency calibration, routine calibration, and in-depth calibration enable timely calibration and maintenance of the system without affecting its normal operation. Finally, a smooth transition method effectively avoids data jumps during the calibration process, improving the system's stability and reliability.

[0126] Please see Figure 3 This diagram illustrates a structural schematic of a power system carbon emission flow metering error tracing system according to an embodiment of the present invention. The system includes:

[0127] The carbon emission flow metering network model is configured to be based on the power system topology, constructing a carbon emission factor calculation model for each node and determining the upstream and downstream dependencies between nodes;

[0128] The relative uncertainty module is configured to quantify the relative uncertainty of source-side metering equipment errors and inter-node data transmission errors based on the error sources of the distributed metering system.

[0129] The error propagation module is configured to calculate the relative uncertainty of carbon emission factors at each node from the source side to the user side according to the error propagation rules, generate a global error propagation path, and calculate the overall relative uncertainty of carbon emissions on the user side.

[0130] The error detection module is configured to perform disturbance analysis on power grid data using probabilistic statistical simulation methods to identify abnormal nodes.

[0131] The calibration strategy generation module is configured to generate calibration strategies based on abnormal nodes.

[0132] The calibration execution module is configured to perform calibration operations according to the calibration strategy.

[0133] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0134] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for tracing the source of measurement errors in carbon emission flows of power systems, characterized in that, The method includes: Based on the topology of the power system, a carbon emission factor calculation model for each node is constructed, and the upstream and downstream dependencies between nodes are determined. Based on the error sources of the distributed metering system, the relative uncertainty of the source-side metering equipment error and the data transmission error between nodes is quantified; According to the error propagation rule, the relative uncertainty of carbon emission factors at each node from the source side to the user side is calculated step by step to generate a global error propagation path, and the overall relative uncertainty of carbon emissions on the user side is calculated. Perturbation analysis of power grid data is performed using probabilistic statistical simulation methods to identify anomalous nodes; this includes randomly sampling the power data of each node in the power grid topology according to a normal distribution. Carbon emission factor is iteratively calculated from the sampled data to generate a carbon emission factor distribution; The dispersion of carbon emission factors at each node is statistically analyzed, the actual uncertainty is calculated and compared with the total relative uncertainty, and abnormal nodes are marked according to the deviation threshold. Based on the error impact weight of abnormal nodes and the power system hierarchy, a calibration priority is generated; Calibration coefficients are generated for abnormal nodes, and the calibration coefficients are dynamically adjusted using a smooth transition algorithm.

2. The method for tracing the source of measurement errors in carbon emission flows of a power system as described in claim 1, characterized in that, Based on the power system topology, a carbon emission factor calculation model for each node is constructed, and the upstream and downstream dependencies between nodes are determined, including: The carbon emission factor of a node is defined as the indirect carbon emission corresponding to a unit of electricity consumption on the load side, and the carbon flow density of a branch is the carbon emission corresponding to a unit of electricity consumption on the branch. Based on the power flow distribution, the carbon emission factor transmission relationship from the upper-level node to the lower-level node is determined by recursive calculation, where the transmission weight is the proportion of active power of each branch.

3. The method for tracing the source of measurement errors in carbon emission flows of a power system as described in claim 1, characterized in that, Based on the error sources of the distributed metering system, the relative uncertainty of the source-side metering equipment error and the data transmission error between nodes is quantified, including: Calculate the relative uncertainty of the source-side carbon emission factor based on the standard uncertainty parameters of the metering equipment; Calculate the relative uncertainty of data transmission error between nodes based on the communication truncation error in data transmission between nodes.

4. The method for tracing the source of measurement errors in carbon emission flows of a power system as described in claim 3, characterized in that, Based on the standard uncertainty parameters of the metering equipment, calculate the relative uncertainty of the source-side carbon emission factor, including: Metering equipment is used to obtain source-side carbon dioxide emissions and active power generation, and the source-side carbon emission factor is calculated. Based on the standard uncertainty of the measuring equipment, the relative uncertainties of carbon dioxide and active power are calculated respectively. According to the law of error propagation, the relative uncertainties of carbon dioxide and active power are combined into the relative uncertainty of the source-side carbon emission factor.

5. The method for tracing the source of measurement errors in carbon emission flows of a power system as described in claim 3, characterized in that, Based on the communication truncation error in inter-node data transmission, the relative uncertainty of the inter-node data transmission error is calculated, including: Calculate communication error based on the data truncation function of the communication algorithm; The relative uncertainty of communication error between nodes is calculated based on the fluctuation range characteristics of the transmitted data.

6. The method for tracing the source of measurement errors in carbon emission streams of a power system as described in any one of claims 1 to 5, characterized in that, According to the error propagation rule, the relative uncertainty of carbon emission factors at each node from the source side to the user side is calculated step by step, including: Calculate the carbon emission factors of each node based on the power and carbon emission factor data of the upper-level node; Based on the relative uncertainties of the active power and carbon emission factors of each branch and the correlation coefficient, the relative uncertainty of the carbon emission factors of the lower-level nodes without considering communication errors is calculated.

7. The method for tracing the source of measurement errors in carbon emission streams of a power system as described in claim 6, characterized in that, Generate a global error propagation path and calculate the overall relative uncertainty of user-side carbon emissions, including: The relative uncertainties of carbon emission factor and active power are corrected based on the relative uncertainty of data transmission error between nodes; The total relative uncertainty of user-side carbon emissions is calculated based on the corrected carbon emission factor and the relative uncertainty of active power.

8. A power system carbon emission flow metering error traceability system, characterized in that, The method for tracing the measurement error of carbon emission flows in a power system according to any one of claims 1 to 7, wherein the system comprises: The carbon emission flow metering network model is configured to be based on the power system topology, constructing a carbon emission factor calculation model for each node and determining the upstream and downstream dependencies between nodes; The relative uncertainty module is configured to quantify the relative uncertainty of source-side metering equipment errors and inter-node data transmission errors based on the error sources of the distributed metering system. The error propagation module is configured to calculate the relative uncertainty of carbon emission factors at each node from the source side to the user side according to the error propagation rules, generate a global error propagation path, and calculate the overall relative uncertainty of carbon emissions on the user side. The error detection module is configured to perform disturbance analysis on power grid data using probabilistic statistical simulation methods to identify abnormal nodes. The calibration strategy generation module is configured to generate calibration strategies based on abnormal nodes. The calibration execution module is configured to perform calibration operations according to the calibration strategy.

Images