Carbon intensity calculation method and device for carbon emission flow of power distribution network and equipment thereof

Abstract

The application discloses a power distribution network carbon emission flow carbon intensity calculation method and device and equipment thereof, relates to the technical field of low-carbon operation of power distribution networks, and the method is: obtaining enhanced features of a to-be-processed power distribution network containing reactive power and node type identification; extracting dynamic carbon flow distribution weights of each node in the to-be-processed power distribution network from the enhanced features based on a pre-trained carbon flow mapping model, the carbon flow mapping model is obtained by offline training based on labeled data generated by an alternating current optimal power flow program containing reactive power and loss solving, and the dynamic carbon flow distribution weights are used to represent the power transmission and carbon flow coupling relationship under the topology of the power distribution network; weighting and aggregating the carbon intensity of the power supply side in the enhanced features based on the dynamic carbon flow distribution weights, to obtain a standardized scale carbon intensity prediction value; and inversely restoring the carbon intensity prediction value to obtain real-time carbon intensity of all nodes including actual physical dimensions of the power distribution network. The carbon emission flow calculation of the power distribution network is adapted to the real-time scheduling demand.

Description

Technical Field

[0001] This application relates to the field of low-carbon operation technology of power distribution networks, and in particular to methods, devices and equipment for calculating the carbon intensity of carbon emission flows in power distribution networks. Background Technology

[0002] In distribution networks with a large number of distributed power sources, the carbon emission of electricity consumption at each node is quantified based on carbon emission flow, which has become the basis for low-carbon coordinated dispatch of distribution networks.

[0003] Currently, the carbon intensity calculation of carbon emission flows in distribution networks typically employs a method based on power flow calculations combined with the inversion of large-scale coefficient matrices. However, the computational complexity of this method increases significantly with the size of distribution network nodes, making real-time calculations difficult. Furthermore, this method only utilizes active power for calculations, failing to accurately capture carbon flow changes caused by AC grid voltage and network losses, resulting in low calculation accuracy. Consequently, the carbon emission flow calculations for distribution networks are ill-suited to real-time dispatching requirements.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a method, apparatus and equipment for calculating the carbon intensity of carbon emission flows in a distribution network, which aims to solve the technical problem that the calculation of carbon emission flows in a distribution network cannot adapt to the real-time scheduling requirements.

[0006] To achieve the above objectives, this application proposes a method for calculating the carbon intensity of carbon emission flows in a power distribution network, the method comprising: Obtain enhanced features of the distribution network to be processed, including reactive power and node type identifiers; The pre-trained carbon flow mapping model extracts the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features. The carbon flow mapping model is obtained by offline training based on labeled data generated by an AC optimal power flow program that includes reactive power and network loss solutions. The dynamic carbon flow allocation weights are used to characterize the power transmission and carbon flow coupling relationship under the distribution network topology. Based on the dynamic carbon flow allocation weights, the power supply-side carbon intensity in the enhanced features is weighted and aggregated to obtain a standardized carbon intensity prediction value. The predicted carbon intensity is reversed to obtain the real-time carbon intensity of all network nodes, including the actual physical dimensions.

[0007] In one embodiment, the step of extracting the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features based on the pre-trained carbon flow mapping model includes: The enhanced features are input into the fully connected layer of the pre-trained carbon flow mapping model to generate a latent feature matrix; Based on the MHA (Multi-Head Attention) in the carbon flow mapping model, two independent linear transformations are performed on the latent feature matrix to generate a query matrix representing the electricity demand of nodes and a key matrix representing the topological features of nodes, respectively. The association strength operation is performed sequentially on the transpose of the query matrix and the key matrix, and the result is converted into a dynamic carbon flow allocation weight that characterizes the proportion of carbon emission contribution of power nodes to load nodes.

[0008] In one embodiment, after performing association strength calculations on the transpose matrices of the query matrix and the key matrix sequentially, and converting the calculation results into dynamic carbon flow allocation weights characterizing the proportion of carbon emission contributions of power nodes to load nodes, the method further includes: After determining the dynamic carbon flow allocation weight as a physical characteristic of the dynamic power transfer distribution factor of the distribution network, the next step is to perform weighted aggregation of the power source side carbon intensity in the enhanced characteristics based on the dynamic carbon flow allocation weight to obtain a carbon intensity prediction value on a standardized scale. The dynamic power transfer distribution factor is used to characterize the nonlinear correction effect of reactive power flow and network loss changes on the spatial distribution of carbon flow.

[0009] In one embodiment, the step of obtaining enhanced characteristics of the distribution network to be processed, including reactive power and node type identifiers, includes: The reactive power and physical attributes of each node in the distribution network to be processed are obtained. The physical attributes include balancing nodes used to anchor the carbon intensity input of the main grid side of the distribution network, distributed power nodes used to carry the carbon intensity of local zero-carbon power sources, ordinary load nodes that characterize carbon responsibility traceability, and interconnection and transmission nodes that characterize the carbon flow channel in the feeder transfer scenario. The physical attributes are augmented using one-hot encoding to generate the node type identifier. The enhanced feature is constructed by integrating the reactive power and the node type identifier.

[0010] In one embodiment, the step of weighted aggregation of the power-side carbon intensity in the enhanced features based on the dynamic carbon flow allocation weights to obtain a standardized carbon intensity prediction value includes: Based on the dynamic carbon flow allocation weights, the power supply-side carbon intensity in the enhanced features is weighted and aggregated to obtain initial carbon intensity data; Based on the pre-stored mean and standard deviation parameters, the initial carbon intensity data is standardized to obtain a carbon intensity prediction value on a standardized scale. The mean and standard deviation parameters are pre-calculated and stored based on the power supply side carbon intensity samples in the offline training dataset.

[0011] In one embodiment, before the step of performing a weighted aggregation operation on the power-side carbon intensity in the enhanced feature based on the dynamic carbon flow allocation weight to obtain initial carbon intensity data, the method further includes: Based on the preset node attribute identifier, determine whether the target node corresponding to the carbon intensity on the power supply side is a distributed power node; When the target node is determined to be a distributed power node, the carbon intensity value of the power supply side is locked to zero according to the preset zero-carbon energy mapping rule. The step of performing a weighted aggregation operation on the power-side carbon intensity in the enhanced feature based on the dynamic carbon flow allocation weights to obtain initial carbon intensity data further includes: Based on the dynamic carbon flow allocation weights, a weighted aggregation operation is performed on the locked power supply side carbon intensity values ​​to obtain initial carbon intensity data.

[0012] In one embodiment, the training steps of the carbon flow mapping model include: The AC optimal power flow program, which includes reactive power and network loss calculations, is invoked to perform simulation calculations on the preset operating condition samples of the distribution network, and obtain the node power distribution data corresponding to each sample. Based on the preset proportional sharing principle, the power distribution data of each node is analyzed to generate the true value of node carbon intensity as labeled data; The reactive power and node type identifier in the operating condition sample are used as input, and the actual value of node carbon intensity is used as the target label to train the predicted carbon flow mapping initial model until convergence, thus obtaining the carbon flow mapping model.

[0013] Furthermore, to achieve the above objectives, this application also proposes a carbon intensity calculation device for carbon emission flows in a power distribution network, the carbon intensity calculation device comprising: The acquisition module is used to acquire enhanced features of the distribution network to be processed, including reactive power and node type identifiers; The extraction module is used to extract the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features based on the pre-trained carbon flow mapping model. The carbon flow mapping model is obtained by offline training based on the labeled data generated by the AC optimal power flow program that includes reactive power and network loss solutions. The dynamic carbon flow allocation weights are used to characterize the power transmission and carbon flow coupling relationship under the distribution network topology. The weighting module is used to perform weighted aggregation of the power source-side carbon intensity in the enhanced features based on the dynamic carbon flow allocation weights to obtain a carbon intensity prediction value on a standardized scale. The reduction module is used to reverse the carbon intensity prediction value to obtain the real-time carbon intensity of all network nodes, including the actual physical dimensions.

[0014] In addition, to achieve the above objectives, this application also proposes a carbon intensity calculation device for carbon emission flows in a power distribution network. The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The computer program is configured to implement the steps of the carbon intensity calculation method for carbon emission flows in a power distribution network as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the carbon intensity calculation method for carbon emission flows of the distribution network as described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: Since reactive power is a direct representation of voltage fluctuations and network loss changes in AC power grids, and node type identification clarifies the physical functions of balancing nodes, distributed power generation nodes, etc., in carbon flow transmission, the enhanced features, which include reactive power and node type identification, can simultaneously encompass the electrical characteristics and topological attributes of the AC system. Furthermore, a carbon flow mapping model pre-trained based on physical simulation labeled data extracts dynamic carbon flow allocation weights from the enhanced features, representing the coupling relationship between power transmission and carbon flow in the distribution network topology. This avoids the high-complexity computation required for inverting large-scale coefficient matrices. Then, based on the dynamic carbon flow allocation weights, the carbon intensity on the power source side is weighted, aggregated, and inversely restored. The carbon flow distribution is calculated from the perspective of the entire distribution network coupling, avoiding physical distortion caused by using only active power. This reduces computation time and can meet real-time dispatch requirements while accurately capturing the nonlinear correction effect of reactive power flow and network loss changes on the spatial distribution of carbon flow. This enables the carbon emission flow calculation of the distribution network to adapt to real-time dispatch requirements. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating the carbon intensity calculation method for carbon emission flows in the distribution network presented in this application; Figure 2 A schematic diagram of the overall implementation process for the carbon intensity calculation method of carbon emission flow in the distribution network of this application; Figure 3 A flowchart illustrating the second embodiment of the carbon intensity calculation method for carbon emission flows in the distribution network of this application; Figure 4 A flowchart illustrating the carbon intensity calculation method for carbon emission flows in the distribution network presented in this application (Example 3). Figure 5 This is a schematic diagram of the module structure of the carbon intensity calculation device for carbon emission flow in the power distribution network according to an embodiment of this application; Figure 6 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the carbon intensity calculation method of carbon emission flow in the distribution network in the embodiments of this application.

[0020] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or dispatch center capable of performing the above functions. The following description uses a dispatch center as an example to illustrate this embodiment and the subsequent embodiments.

[0024] Based on this, embodiments of this application provide a method for calculating the carbon intensity of carbon emission flows in a power distribution network, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the carbon intensity calculation method for carbon emission flows in the power distribution network according to this application.

[0025] In this embodiment, the method for calculating the carbon intensity of the carbon emission stream of the distribution network includes steps S10 to S40: Step S10: Obtain the enhanced features of the distribution network to be processed, including reactive power and node type identifiers; It should be noted that the distribution network to be processed is a power distribution network that requires carbon emission flow analysis and carbon intensity calculation. Reactive power is the power used to establish magnetic and electric fields in AC circuits but does not directly perform useful work. Node type identifiers are sparse attribute vectors generated by feature expansion of node physical attributes using one-hot encoding, used to clearly classify distribution network nodes into four categories: balancing nodes, distributed generation nodes, ordinary load nodes, and interconnection transmission nodes. Enhanced features are high-dimensional input feature matrices generated by concatenating the standardized numerical feature vectors with the one-hot encoded feature vectors of node attributes along the feature dimension. Enhanced features include, but are not limited to, active power injection, reactive power injection, real-time boundary carbon intensity of electricity purchased from the upstream main grid, carbon intensity of various distributed units, and node attribute labels.

[0026] It is understandable that by incorporating reactive power injection as an explicit input feature into the enhanced feature matrix, the carbon flow mapping model can automatically perceive the impact of reactive power injection at different nodes on the branch power flow distribution through the dot product operation in the attention mechanism during forward calculation. This implicitly compensates for the nonlinear correction effect of node voltage amplitude fluctuations and active power loss changes caused by reactive flow on the spatial distribution of carbon flow when assigning weights to the output carbon flow, ultimately making the model prediction results conform to the physical laws of AC power flow.

[0027] Understandably, by incorporating the node type identifier into the enhanced feature matrix, the carbon flow mapping model can apply differentiated feature extraction strategies to different nodes based on their type identifiers during training. This enables the model to accurately distinguish the carbon flow transmission direction between power supply nodes and load nodes when extracting carbon flow allocation weights, thereby improving the carbon flow mapping model's ability to learn the spatial distribution pattern of carbon flow and its physical consistency.

[0028] Specifically, step S10 also includes: The reactive power and physical attributes of each node in the distribution network to be processed are obtained. The physical attributes include balancing nodes used to anchor the carbon intensity input of the main grid side of the distribution network, distributed power nodes used to carry the carbon intensity of local zero-carbon power sources, ordinary load nodes that characterize carbon responsibility traceability, and interconnection and transmission nodes that characterize the carbon flow channel in the feeder transfer scenario. The physical attributes are augmented using one-hot encoding to generate the node type identifier. The enhanced feature is constructed by integrating the reactive power and the node type identifier.

[0029] It should be noted that physical attributes are the inherent characteristics of each node in the distribution network in terms of physical structure and function, including balancing nodes, distributed generation nodes, ordinary load nodes, and interconnection and transmission nodes. The node type identifier is a sparse attribute vector representing the physical role of heterogeneous nodes, generated after one-hot encoding.

[0030] It is understandable that mapping the physical attributes of nodes to four-dimensional sparse attribute vectors and then concatenating them into the enhanced feature matrix allows the fully connected mapping layer and attention mechanism in the carbon flow mapping model to adaptively adjust the weight allocation strategy based on the category information in the one-hot encoding when processing the features of each node. This enables the model to focus on extracting carbon emission output features for power nodes and carbon responsibility tracing features for load nodes, thereby improving the model's learning accuracy and physical consistency in understanding the spatial distribution of carbon flow.

[0031] Step S20: Based on the pre-trained carbon flow mapping model, extract the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features. The carbon flow mapping model is obtained by offline training based on labeled data generated by the AC optimal power flow program that includes reactive power and network loss solutions. The dynamic carbon flow allocation weights are used to characterize the power transmission and carbon flow coupling relationship under the distribution network topology. It should be noted that the carbon flow mapping model is an end-to-end neural network model built on a multi-head attention mechanism, used to realize a nonlinear mapping from the operating state of the distribution network to the spatial carbon flow distribution. The AC optimal power flow program is a high-precision physical simulation program that includes reactive power and network loss solutions, used to calculate accurate analytical solutions for the carbon intensity of all network nodes by combining the principle of carbon emission flow proportion sharing. The labeled data is a high-quality offline training dataset containing operating states for multiple time periods and their corresponding real label vectors, constructed through high-precision physical simulation steps. The dynamic carbon flow allocation weights are elements in the weight allocation matrix obtained by calculating through a scaled dot product attention mechanism and normalizing using the Softmax function, which are physically and logically equivalent to the dynamic power transfer distribution factor under AC power flow.

[0032] Understandably, since the carbon flow mapping model has already completed parameter learning in the offline training phase using high-precision labeled data generated by the AC optimal power flow program that includes reactive power and network loss solutions, the complex nonlinear physical solution process is compressed into the weight parameters of the neural network. Therefore, in the online inference phase, only deterministic matrix multiplication and basic algebra operations are required to output dynamic carbon flow allocation weights, completely eliminating the steps of Jacobian matrix iteration and high-dimensional sparse matrix inversion. This reduces the carbon flow calculation time to the millisecond level, meeting the real-time requirements of low-carbon economic dispatch of the distribution network.

[0033] Specifically, the training steps for the carbon flow mapping model include: The AC optimal power flow program, which includes reactive power and network loss calculations, is invoked to perform simulation calculations on the preset operating condition samples of the distribution network, and obtain the node power distribution data corresponding to each sample. Based on the preset proportional sharing principle, the power distribution data of each node is analyzed to generate the true value of node carbon intensity as labeled data; The reactive power and node type identifier in the operating condition sample are used as input, and the actual value of node carbon intensity is used as the target label to train the predicted carbon flow mapping initial model until convergence, thus obtaining the carbon flow mapping model.

[0034] It should be noted that the operating condition sample is a large-scale historical operating data set representing the distribution network under various typical days, covering scenarios such as high-proportion renewable energy output, low-proportion renewable energy output, load peak-valley alternation, and switching operations under different topologies. Node power distribution data is the electrical quantity data of each node obtained through an AC optimal power flow program. The true value of node carbon intensity is the accurate analytical solution of the entire network node carbon intensity calculated using a large-scale sparse matrix inversion analytical model, serving as the true label vector for the corresponding input data.

[0035] Understandably, simulations using an AC optimal power flow program that includes reactive power and network loss calculations can accurately calculate the voltage magnitude, voltage phase angle, active power, and reactive power distribution at each node. Furthermore, the generated node power distribution data includes nonlinear correction information for the carbon flow distribution caused by reactive flow and network loss changes, providing accurate physical data for generating high-quality labeled data.

[0036] In practical implementation, historical operating data of the distribution network can be collected. For each operating state, an AC optimal power flow program that includes reactive power and network loss calculations is adopted. Combined with the principle of carbon emission flow ratio sharing, an analytical model for inverting a large-scale sparse matrix is ​​used to calculate the accurate analytical solution of the carbon intensity of all nodes in the network. The carbon intensity of all nodes in the network is used as the true label vector of the corresponding input data to construct an offline training dataset. Mean square error is used as the basic empirical risk measure, and a regularization term is introduced to construct a comprehensive loss function. An adaptive moment estimation optimization algorithm is adopted, and the gradient of the loss function with respect to the parameter set is calculated based on the backpropagation mechanism. End-to-end weight updates are performed across the complex physical power flow equation solution process until the loss function converges. The optimal parameter set is saved as the final carbon flow mapping model.

[0037] Step S30: Based on the dynamic carbon flow allocation weight, the power supply side carbon intensity in the enhanced feature is weighted and aggregated to obtain a carbon intensity prediction value on a standardized scale. It should be noted that the carbon intensity on the power supply side includes the real-time boundary carbon intensity flowing from the upstream main grid into the distribution network, as well as the carbon intensity of various distributed generating units connected to the distribution network. Weighted aggregation is the process of independently inputting the feature matrix containing carbon flow allocation weight information, output by the multi-head attention mechanism, into a feedforward neural network containing nonlinear activation functions for feature aggregation and dimensionality reduction mapping, categorized by node dimension. The carbon intensity prediction value at the standardized scale is the node carbon intensity value after nonlinear aggregation and dimensionality reduction mapping by the feedforward neural network, but before physical dimension restoration.

[0038] It is understandable that by independently inputting the attention output matrix into a feedforward neural network containing two fully connected layers and nonlinear activation functions such as Gaussian error linear units, and performing layer-by-layer aggregation and dimensionality reduction mapping of high-dimensional coupled features through multi-layer nonlinear transformations, the carbon flow mapping model can accurately capture the nonlinear laws of carbon flow interaction between sources and loads. Finally, the aggregated features are compressed into a one-dimensional output to obtain the predicted carbon intensity value of each node under the standardized scale.

[0039] It is understandable that, since the enhanced features in step S10 are derived from the standardized data, and the mapping relationship learned by the carbon flow mapping model during training is based on the standardized data space, in this embodiment, the output of the feedforward neural network directly corresponds to the carbon intensity value in the standardized space, thereby avoiding numerical shifts or oscillations in the model output due to differences in the magnitude of the input data.

[0040] Step S40: Reversely restore the predicted carbon intensity value to obtain the real-time carbon intensity of all network nodes, including the actual physical dimensions.

[0041] It should be noted that reverse restoration utilizes the mean and standard deviation of the carbon intensity characteristics of centralized nodes in the offline dataset saved during the data preprocessing stage to convert the predicted carbon intensity value under standardized scale into a value with actual physical dimensions. The actual physical dimensions refer to the actual physical meaning and unit of carbon intensity. The real-time carbon intensity of all network nodes is the weighted average carbon emission rate corresponding to the unit active power injection or outflow of all target nodes in the distribution network, used to characterize the carbon emission attributes of node electrical energy.

[0042] Understandably, since the output value of the feedforward neural network in the carbon flow mapping model is a value in a dimensionless space after standardization, this value cannot be directly used for carbon emission accounting or low-carbon dispatching decisions. Therefore, it is necessary to call the mean and standard deviation parameters of the node carbon intensity characteristics in the offline dataset saved in the data preprocessing stage of step S10 to perform inverse standardization transformation on the carbon intensity prediction value at the standardized scale. That is, multiply by the standard deviation and add the mean to map the prediction value from the standardized space back to the original physical dimension space, thereby obtaining a real-time carbon intensity prediction vector of all network nodes with actual physical units. This allows the calculation result to be directly embedded into the distribution network low-carbon economic dispatching model as a carbon flow constraint.

[0043] Understandably, through end-to-end rapid calculation, real-time carbon intensity with actual physical dimensions for each node in the entire network can be output within milliseconds. This allows the dispatch center to directly use the data to build a low-carbon economic dispatch model with carbon flow constraints, realize the coordinated optimization dispatch of power distribution network carbon, and at the same time provide load-side users with accurate carbon responsibility traceability data to support carbon quota management and carbon footprint accounting.

[0044] In the specific implementation, refer to Figure 2 In the offline training phase: Historical operating data for at least one year is collected from the distribution network's historical operating database, including electrical quantities such as active power, reactive power, voltage amplitude, and phase angle at each node, as well as environmental parameters such as carbon intensity at the upper-level main grid boundary and carbon intensity of distributed generation sources. In the data preprocessing phase, the physical attributes of nodes are one-hot encoded, mapping four types of physical attributes—balanced nodes, distributed generation nodes, ordinary load nodes, and interconnection transmission nodes—to 4-dimensional sparse vectors. These vectors are then concatenated with standardized power and carbon intensity data along the feature dimension to construct an enhanced feature input vector. In the offline label data generation phase, the AC Optimal Power Flow (AC-OPF) program, which includes reactive power and network loss calculations, is used to simulate and calculate the historical operating data. Combining the proportional sharing principle, the accurate carbon intensity values ​​of each node in the entire network are solved using a large-scale sparse matrix inversion analytical model. In the model building and training phase, an end-to-end neural network model with a multi-head attention mechanism (MHA) is constructed. This model consists of a fully connected layer, a multi-head attention layer, and a feedforward neural network. During training, the Adam (Adaptive Moment Estimation) optimization algorithm is used to update the parameters. Training is stopped when the validation set loss no longer decreases for 10 consecutive rounds. Finally, the optimal model parameters are saved.

[0045] In the online inference phase: During the real-time data acquisition stage, node electrical quantity data under the current operating state of the distribution network to be processed are collected in real time. In the data preprocessing stage, the mean and standard deviation vectors saved during the offline training stage are used to standardize the real-time data, and the node physical attributes are one-hot encoded and concatenated to the feature vector. In the online fast inference stage, the standardized enhanced feature input vector is imported into the trained carbon flow mapping model, and the dynamic carbon flow allocation weights and standardized carbon intensity prediction values ​​for each node are calculated through forward propagation. In the result output stage, the predicted values ​​are inversely restored using the carbon intensity mean and standard deviation parameters saved during the offline training stage, generating a dynamic carbon intensity prediction vector for all network nodes with actual physical dimensions. The results are then transmitted to the distribution network dispatch center or edge computing terminal for low-carbon economic dispatch and carbon emission monitoring.

[0046] This embodiment provides a method for calculating the carbon intensity of carbon emission flows in a distribution network. Since reactive power is a direct representation of voltage fluctuations and network loss changes in an AC power grid, and node type identifiers clarify the physical functions of balancing nodes, distributed power generation nodes, etc., in carbon flow transmission, the enhanced features, which include reactive power and node type identifiers, can simultaneously include the electrical characteristics and topological attributes of the AC system. Based on a carbon flow mapping model pre-trained with physical simulation labeled data, dynamic carbon flow allocation weights representing the coupling relationship between power transmission and carbon flow in the distribution network topology are extracted from the enhanced features. This avoids the high-complexity operations required for inverting large-scale coefficient matrices. Then, the carbon intensity on the power generation side is weighted, aggregated, and reversed according to the dynamic carbon flow allocation weights. The carbon flow distribution is calculated from the perspective of the entire distribution network coupling, avoiding physical distortion caused by using only active power. This reduces computation time and can meet the real-time scheduling requirements while accurately capturing the nonlinear correction effect of reactive power flow and network loss changes on the spatial distribution of carbon flow. This enables the calculation of carbon emission flows in the distribution network to adapt to real-time scheduling requirements.

[0047] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 Step S20 also includes steps S01 to S03: Step S01: Input the enhanced features into the fully connected layer of the pre-trained carbon flow mapping model to generate a latent feature matrix; Step S02: Based on the MHA in the carbon flow mapping model, perform two independent linear transformations on the latent feature matrix to generate a query matrix representing the node's electricity demand and a key matrix representing the node's topological features. Step S03: Perform association strength calculations on the transpose of the query matrix and the key matrix in sequence, and convert the calculation results into dynamic carbon flow allocation weights that characterize the proportion of carbon emission contribution of power nodes to load nodes.

[0048] It should be noted that the latent feature matrix refers to the high-dimensional feature representation generated after the enhanced feature input vector is processed through a fully connected mapping layer and a nonlinear activation function. The query matrix represents the electricity demand of each load node and the direction of carbon responsibility tracing. The key matrix represents the carbon emission characteristics and grid topology characteristics of each power generation node or main grid interconnection node. The correlation strength calculation is a process of calculating the feature interaction correlation score between nodes using a scaled dot product attention mechanism and then normalizing it using the Softmax function.

[0049] Understandably, by using the learnable weight matrix of the fully connected mapping layer to linearly project and nonlinearly activate the enhanced features, the original features are mapped to a high-dimensional hidden layer space. This allows the dimensions of the mapped hidden feature matrix to capture the abstract interaction relationships between the original features, thus providing a feature representation basis for the generation of the query matrix and key matrix, and improving the recognition accuracy of the attention mechanism for the carbon flow interaction features between the source and load.

[0050] It is understandable that query and key linear transformations are performed independently in each attention head, allowing different attention heads to focus on carbon flow interaction features at different scales and under different modes. Finally, the output sub-matrices of multiple attention heads are concatenated along the column dimension and the information is synthesized through a linear output layer, thereby constructing a carbon flow mapping operator with physical interpretability and improving the ability of the carbon flow mapping model to represent complex carbon flow spatial coupling laws.

[0051] Understandably, the carbon flow correlation strength between each node pair is measured by calculating the dot product score of the query matrix and the transpose of the key matrix. Then, the score vector is row-normalized using the Softmax function, so that the sum of the elements in each row of the normalized weight allocation matrix is ​​one. This is highly consistent with the mathematical form of the carbon flow density weighted average in the proportional sharing principle, thereby ensuring that the carbon flow allocation weights output by the model conform to the basic physical laws of carbon emission flow.

[0052] Furthermore, after step S03, the method for calculating the carbon intensity of the distribution network carbon emission stream further includes: After determining the dynamic carbon flow allocation weight as a physical characteristic of the dynamic power transfer distribution factor of the distribution network, the next step is to perform weighted aggregation of the power source side carbon intensity in the enhanced characteristics based on the dynamic carbon flow allocation weight to obtain a carbon intensity prediction value on a standardized scale. The dynamic power transfer distribution factor is used to characterize the nonlinear correction effect of reactive power flow and network loss changes on the spatial distribution of carbon flow.

[0053] It should be noted that the dynamic power transfer distribution factor is a physical quantity used to characterize the degree of influence of node-injected power on branch power flow or node voltage in a power grid.

[0054] It is understandable that, since the labeled data used by the carbon flow mapping model in the offline training phase is generated by an AC optimal power flow program that includes solutions for reactive power and network losses, and by minimizing the prediction error during the end-to-end training process, the weight allocation matrix learns the correction law of reactive power flow and network loss changes on the carbon flow distribution, the weight allocation matrix output during online inference contains nonlinear correction information for reactive power and network losses. That is, the physical rationality and computational accuracy of the carbon flow allocation results in the AC environment can be guaranteed without explicitly solving the AC power flow equations.

[0055] Based on the first and second embodiments of this application, the same or similar content as the above embodiments in the third embodiment of this application can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 4 Step S30 also includes steps S1 to S2: Step S1: Perform a weighted aggregation operation on the power supply side carbon intensity in the enhanced feature based on the dynamic carbon flow allocation weight to obtain initial carbon intensity data; Step S2: Based on the pre-stored mean and standard deviation parameters, the initial carbon intensity data is standardized to obtain a carbon intensity prediction value on a standardized scale. The mean and standard deviation parameters are pre-calculated and stored based on the power supply side carbon intensity samples in the offline training dataset.

[0056] It should be noted that the mean parameter and standard deviation parameter are statistical parameters that are pre-calculated and stored based on the power supply side carbon intensity samples in the offline training dataset, and are used for standardization and its reverse reduction.

[0057] It is understandable that the initial carbon intensity data is standardized by using the mean and standard deviation parameters that are pre-calculated and stored in the offline training phase. This is to uniformly map the carbon intensity values ​​under different operating conditions to a standardized space with zero mean and unit variance, so that the carbon flow mapping model can perform inference calculations under a consistent data distribution under various operating conditions, thereby improving the generalization ability and prediction stability of the carbon flow mapping model in different scenarios.

[0058] Furthermore, prior to step S1, the method for calculating the carbon intensity of the distribution network carbon emission stream includes: Based on the preset node attribute identifier, determine whether the target node corresponding to the carbon intensity on the power supply side is a distributed power node; When the target node is determined to be a distributed power node, the carbon intensity value of the power supply side is locked to zero according to the preset zero-carbon energy mapping rule. The step of performing a weighted aggregation operation on the power-side carbon intensity in the enhanced feature based on the dynamic carbon flow allocation weights to obtain initial carbon intensity data further includes: Based on the dynamic carbon flow allocation weights, a weighted aggregation operation is performed on the locked power supply side carbon intensity values ​​to obtain initial carbon intensity data.

[0059] It should be noted that node attribute identifiers are used to determine whether a node is a specific type of flag or code. Zero-carbon energy mapping rules refer to pre-defined rules used to force the carbon intensity of renewable energy units such as wind power and photovoltaics in distributed power generation nodes to be zero.

[0060] Understandably, forcibly locking the carbon intensity values ​​of distributed power nodes before weighted aggregation can eliminate calculation biases caused by incorrect carbon intensity parameters from the source of data. This ensures that the carbon flow mapping model can accurately quantify the carbon reduction effect of local clean energy consumption when dealing with distribution networks with a high proportion of distributed energy, thereby improving the authenticity and reliability of the carbon intensity calculation results for the entire network.

[0061] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the carbon intensity calculation method of carbon emission flow of distribution network in this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0062] This application also provides a carbon intensity calculation device for carbon emission flows in a power distribution network. Please refer to [reference needed]. Figure 5 The carbon intensity calculation device for the carbon emission stream of the power distribution network includes: The acquisition module 10 is used to acquire enhanced features of the distribution network to be processed, including reactive power and node type identifiers; Extraction module 20 is used to extract the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features based on the pre-trained carbon flow mapping model. The carbon flow mapping model is obtained by offline training based on the labeled data generated by the AC optimal power flow program that includes reactive power and network loss solutions. The dynamic carbon flow allocation weights are used to characterize the power transmission and carbon flow coupling relationship under the distribution network topology. The weighting module 30 is used to perform weighted aggregation of the power supply side carbon intensity in the enhanced features based on the dynamic carbon flow allocation weights to obtain a carbon intensity prediction value on a standardized scale. The restoration module 40 is used to reverse restore the predicted carbon intensity value to obtain the real-time carbon intensity of all network nodes, including the actual physical dimensions.

[0063] The carbon intensity calculation device for distribution network carbon emission flows provided in this application employs the carbon intensity calculation method for distribution network carbon emission flows in the above embodiments, which can solve the technical problem that the calculation of carbon emission flows in distribution networks cannot adapt to real-time scheduling requirements. Compared with the prior art, the beneficial effects of the carbon intensity calculation device for distribution network carbon emission flows provided in this application are the same as the beneficial effects of the carbon intensity calculation method for distribution network carbon emission flows provided in the above embodiments, and other technical features in the carbon intensity calculation device for distribution network carbon emission flows are the same as the features disclosed in the methods of the above embodiments, and will not be repeated here.

[0064] This application provides a carbon intensity calculation device for carbon emission flows in a power distribution network. The carbon intensity calculation device for carbon emission flows in a power distribution network includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the carbon intensity calculation method for carbon emission flows in a power distribution network as described in Embodiment 1 above.

[0065] The following is for reference. Figure 6 The diagram illustrates a structural schematic of a carbon intensity calculation device suitable for implementing carbon emission flows in a power distribution network according to embodiments of this application. The carbon intensity calculation device for carbon emission flows in a power distribution network according to embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, personal digital assistants, tablets, portable multimedia players, and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 6 The carbon intensity calculation device for carbon emission flows in the power distribution network shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0066] like Figure 6As shown, the carbon intensity calculation device for carbon emission streams in a power distribution network may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the carbon intensity calculation device for carbon emission streams in the power distribution network. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. An I / O (input / output) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the carbon intensity calculation device for distribution network carbon emission flows to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a carbon intensity calculation device for distribution network carbon emission flows with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0067] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0068] The carbon intensity calculation device for distribution network carbon emission flows provided in this application, employing the carbon intensity calculation method for distribution network carbon emission flows in the above embodiments, can solve the technical problem that the calculation of carbon emission flows in distribution networks cannot adapt to real-time scheduling requirements. Compared with the prior art, the beneficial effects of the carbon intensity calculation device for distribution network carbon emission flows provided in this application are the same as the beneficial effects of the carbon intensity calculation method for distribution network carbon emission flows provided in the above embodiments, and other technical features in this carbon intensity calculation device for distribution network carbon emission flows are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0069] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0070] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0071] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the carbon intensity calculation method for carbon emission streams of the power distribution network in the above embodiments.

[0072] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0073] The aforementioned computer-readable storage medium may be included in a carbon intensity calculation device for carbon emission streams in a power distribution network; or it may exist independently and not be incorporated into a carbon intensity calculation device for carbon emission streams in a power distribution network.

[0074] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the carbon intensity calculation device for carbon emission streams in the distribution network, enable the carbon intensity calculation device for carbon emission streams in the distribution network as described in the above embodiments to perform the carbon intensity calculation method for carbon emission streams in the distribution network.

[0075] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0076] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0077] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0078] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the carbon intensity calculation method for carbon emission flows in the distribution network described above. This solves the technical problem that the calculation of carbon emission flows in the distribution network cannot adapt to real-time scheduling requirements. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the carbon intensity calculation method for carbon emission flows in the distribution network provided in the above embodiments, and will not be repeated here.

[0079] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the carbon intensity calculation method for carbon emission flows in a power distribution network as described above.

[0080] The computer program product provided in this application can solve the technical problem that the calculation of carbon emission flows in distribution networks cannot adapt to the real-time scheduling requirements. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the carbon intensity calculation method for carbon emission flows in distribution networks provided in the above embodiments, and will not be repeated here.

[0081] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the scope of protection of this application.

Claims

1. A method for calculating the carbon intensity of carbon emission flows in a power distribution network, characterized in that, The method includes: Obtain enhanced features of the distribution network to be processed, including reactive power and node type identifiers; The pre-trained carbon flow mapping model extracts the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features. The carbon flow mapping model is obtained by offline training based on labeled data generated by an AC optimal power flow program that includes reactive power and network loss solutions. The dynamic carbon flow allocation weights are used to characterize the power transmission and carbon flow coupling relationship under the distribution network topology. Based on the dynamic carbon flow allocation weights, the power supply-side carbon intensity in the enhanced features is weighted and aggregated to obtain a standardized carbon intensity prediction value. The predicted carbon intensity is reversed to obtain the real-time carbon intensity of all network nodes, including the actual physical dimensions. The steps for extracting the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features based on the pre-trained carbon flow mapping model include: The enhanced features are input into the fully connected layer of the pre-trained carbon flow mapping model to generate a latent feature matrix; Based on the MHA in the carbon flow mapping model, two independent linear transformations are performed on the latent feature matrix to generate a query matrix representing the electricity demand of nodes and a key matrix representing the topological features of nodes, respectively. The association strength operation is performed sequentially on the transpose of the query matrix and the key matrix, and the result is converted into a dynamic carbon flow allocation weight that characterizes the proportion of carbon emission contribution of power nodes to load nodes.

2. The method for calculating the carbon intensity of carbon emission flows in a power distribution network as described in claim 1, characterized in that, After performing association strength calculations on the transpose matrices of the query matrix and the key matrix sequentially, and converting the results into dynamic carbon flow allocation weights characterizing the proportion of carbon emission contributions from power nodes to load nodes, the method further includes: After determining the dynamic carbon flow allocation weight as a physical characteristic of the dynamic power transfer distribution factor of the distribution network, the next step is to perform weighted aggregation of the power source side carbon intensity in the enhanced characteristics based on the dynamic carbon flow allocation weight to obtain a carbon intensity prediction value on a standardized scale. The dynamic power transfer distribution factor is used to characterize the nonlinear correction effect of reactive power flow and network loss changes on the spatial distribution of carbon flow.

3. The method for calculating the carbon intensity of carbon emission flows in a power distribution network as described in claim 1, characterized in that, The steps for obtaining enhanced characteristics of the distribution network to be processed, including reactive power and node type identifiers, include: The reactive power and physical attributes of each node in the distribution network to be processed are obtained. The physical attributes include balancing nodes used to anchor the carbon intensity input of the main grid side of the distribution network, distributed power nodes used to carry the carbon intensity of local zero-carbon power sources, ordinary load nodes that characterize carbon responsibility traceability, and interconnection and transmission nodes that characterize the carbon flow channel in the feeder transfer scenario. The physical attributes are augmented using one-hot encoding to generate the node type identifier. The enhanced feature is constructed by integrating the reactive power and the node type identifier.

4. The method for calculating the carbon intensity of carbon emission flows in a power distribution network as described in claim 1, characterized in that, The step of weighted aggregation of the power source-side carbon intensity in the enhanced features based on the dynamic carbon flow allocation weights to obtain a standardized carbon intensity prediction value includes: Based on the dynamic carbon flow allocation weights, the power supply-side carbon intensity in the enhanced features is weighted and aggregated to obtain initial carbon intensity data; Based on the pre-stored mean and standard deviation parameters, the initial carbon intensity data is standardized to obtain a carbon intensity prediction value on a standardized scale. The mean and standard deviation parameters are pre-calculated and stored based on the power supply side carbon intensity samples in the offline training dataset.

5. The method for calculating the carbon intensity of carbon emission flows in a power distribution network as described in claim 4, characterized in that, Before the step of performing a weighted aggregation operation on the power-side carbon intensity in the enhanced feature based on the dynamic carbon flow allocation weights to obtain the initial carbon intensity data, the method further includes: Based on the preset node attribute identifier, determine whether the target node corresponding to the carbon intensity on the power supply side is a distributed power node; When the target node is determined to be a distributed power node, the carbon intensity value of the power supply side is locked to zero according to the preset zero-carbon energy mapping rule. The step of performing a weighted aggregation operation on the power-side carbon intensity in the enhanced feature based on the dynamic carbon flow allocation weights to obtain initial carbon intensity data further includes: Based on the dynamic carbon flow allocation weights, a weighted aggregation operation is performed on the locked power supply side carbon intensity values ​​to obtain initial carbon intensity data.

6. The method for calculating the carbon intensity of carbon emission flows in a power distribution network as described in claim 1, characterized in that, The training steps for the carbon flow mapping model include: The AC optimal power flow program, which includes reactive power and network loss calculations, is invoked to perform simulation calculations on the preset operating condition samples of the distribution network, and obtain the node power distribution data corresponding to each sample. Based on the preset proportional sharing principle, the power distribution data of each node is analyzed to generate the true value of node carbon intensity as labeled data; The reactive power and node type identifier in the operating condition sample are used as input, and the actual value of node carbon intensity is used as the target label to train the predicted carbon flow mapping initial model until convergence, thus obtaining the carbon flow mapping model.

7. A carbon intensity calculation device for carbon emission flows in a power distribution network, characterized in that, The device includes: The acquisition module is used to acquire enhanced features of the distribution network to be processed, including reactive power and node type identifiers; The extraction module is used to extract the dynamic carbon flow allocation weights of each node in the distribution network to be processed from the enhanced features based on the pre-trained carbon flow mapping model. The carbon flow mapping model is obtained by offline training based on the labeled data generated by the AC optimal power flow program that includes reactive power and network loss solutions. The dynamic carbon flow allocation weights are used to characterize the power transmission and carbon flow coupling relationship under the distribution network topology. The weighting module is used to perform weighted aggregation of the power source-side carbon intensity in the enhanced features based on the dynamic carbon flow allocation weights to obtain a carbon intensity prediction value on a standardized scale. The reduction module is used to reverse the carbon intensity prediction value to obtain the real-time carbon intensity of all network nodes, including the actual physical dimensions. The steps for extracting dynamic carbon flow allocation weights for each node in the distribution network to be processed from the enhanced features based on the pre-trained carbon flow mapping model include: inputting the enhanced features into the fully connected layer of the pre-trained carbon flow mapping model to generate a latent feature matrix; performing two independent linear transformations on the latent feature matrix based on the MHA in the carbon flow mapping model to generate a query matrix representing the electricity demand of the nodes and a key matrix representing the topological features of the nodes; sequentially performing association strength calculations on the transpose of the query matrix and the key matrix, and converting the calculation results into dynamic carbon flow allocation weights representing the proportion of carbon emission contribution of power generation nodes to load nodes.

8. A carbon intensity calculation device for carbon emission flows in a power distribution network, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the method for calculating the carbon intensity of carbon emission flows in a distribution network as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the carbon intensity calculation method for carbon emission flows of the distribution network as described in any one of claims 1 to 6.

Images