Carbon emission monitoring method and device, storage medium and carbon emission terminal

Abstract

The embodiment of the invention discloses a carbon emission monitoring method and device, a storage medium and a carbon emission terminal, and relates to the field of environment monitoring. According to the invention, the total ammeter data is automatically collected and the pre-training model is utilized to accurately identify the type and the number of the devices, so that fine decomposition of energy consumption is realized. And accurately calculating the carbon emission of each device in combination with the energy consumption of the device and the carbon emission factor, and then summarizing to obtain the total carbon emission. The process reduces manual intervention, improves the accounting efficiency, and ensures the accuracy of data. According to the scheme, enterprises can easily master the full view of carbon emission, solid data support is provided for formulating energy-saving and emission-reducing measures, and implementation of green development and sustainable development strategies is powerfully promoted.

Description

Technical Field

[0001] The present application relates to the field of environmental monitoring, and in particular to a carbon emission monitoring method, device, storage medium and carbon emission terminal. Background Art

[0002] In the context of addressing global climate change and promoting sustainable development, accurate monitoring and management of carbon emissions has become key. Existing technologies mostly rely on offline sampling and laboratory analysis. These methods have problems such as low monitoring frequency, poor data timeliness, and inability to reflect carbon emission dynamics in real time. In addition, traditional monitoring methods are costly and difficult to deploy on a large scale, and data processing mostly relies on manual labor, which limits accuracy and efficiency. With the development of industrial automation and intelligence, there is an urgent need for a solution that can monitor the carbon emissions of energy consumption networks in real time, accurately, and efficiently. Summary of the invention

[0003] The embodiments of the present application provide a carbon emission monitoring method, device, storage medium and carbon emission terminal, which can solve the problem that the carbon emission cannot be accurately detected in the prior art. The technical solution is as follows:

[0004] In a first aspect, an embodiment of the present application provides a method for monitoring carbon emissions, the method comprising:

[0005] Acquire total electric meter characteristic data measured by a total electric meter in an energy consumption network; the total electric meter characteristic data includes: one or more of current value, voltage value, active power and reactive power;

[0006] Identify the total electric meter characteristic data according to a plurality of pre-trained device identification models to obtain all device types and device quantities in the energy consumption network; the plurality of device identification models are used to identify different device types;

[0007] Collecting the energy consumption values ​​measured by the total electric meter according to a preset time period, and generating an energy consumption time series according to the collected data; the energy consumption time series includes multiple time points and energy consumption values ​​associated with each time point;

[0008] Performing energy consumption decomposition on the energy consumption time series to obtain energy consumption values ​​of each identified device type;

[0009] Calculate the carbon emissions of each equipment type based on the energy consumption value and carbon emission factor of each equipment type;

[0010] The carbon emissions of each equipment type are accumulated to obtain the total carbon emissions of the energy consumption network.

[0011] In a second aspect, an embodiment of the present application provides a device for monitoring carbon emissions, the device comprising:

[0012] An acquisition unit, used to acquire total meter characteristic data measured by a total meter in an energy consumption network; the total meter characteristic data includes: one or more of current value, voltage value, active power and reactive power;

[0013] An identification unit, used to identify the characteristic data of the total electric meter according to a plurality of pre-trained device identification models, and obtain all device types and device quantities in the energy consumption network; the plurality of device identification models are used to identify different device types;

[0014] A collection unit, used to collect the energy consumption values ​​measured by the total electric meter according to a preset time period, and generate an energy consumption time series according to the collected data; the energy consumption time series includes multiple time points and energy consumption values ​​associated with each time point;

[0015] A decomposition unit, used for performing energy consumption decomposition on the energy consumption time series to obtain energy consumption values ​​of each identified device type;

[0016] A calculation unit, used for calculating the carbon emission of each equipment type according to the energy consumption value and carbon emission factor of each equipment type;

[0017] The summarizing unit is used to accumulate the carbon emissions of each device type to obtain the total carbon emissions of the energy consumption network.

[0018] In a third aspect, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above-mentioned method steps.

[0019] In a fourth aspect, an embodiment of the present application provides a carbon emission terminal, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0020] The beneficial effects brought about by the technical solutions provided by some embodiments of the present application include at least:

[0021] By automatically acquiring the characteristic data measured by the total electric meter (such as current, voltage, active power, reactive power, etc.), and using multiple pre-trained device identification models for intelligent analysis, the automatic identification of the type and quantity of equipment in the energy consumption network is realized. This process greatly reduces manual intervention, improves the efficiency and accuracy of data processing, and enables carbon emission accounting to be carried out automatically and continuously. Using the trained device identification model, different types of equipment and their quantities in the energy consumption network can be accurately distinguished, providing a solid foundation for subsequent energy consumption decomposition. Through the construction and analysis of energy consumption time series, the accurate decomposition of energy consumption of various types of equipment is further realized, ensuring the precision and accuracy of carbon emission calculation. Combining the energy consumption value of each type of equipment and the corresponding carbon emission factor, this technical solution can accurately calculate the carbon emissions of each type of equipment. This accounting method based on actual energy consumption data avoids the errors that may be caused by traditional estimation methods, and provides more reliable data support for the formulation and implementation of energy conservation and emission reduction policies. By accumulating the carbon emissions of each type of equipment, the total carbon emissions of the energy consumption network are obtained, realizing a comprehensive control of the carbon emissions of the entire network. This result not only helps companies understand their own carbon emissions, but also provides a strong basis for formulating scientific emission reduction targets and strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 It is a schematic diagram of the network architecture provided by the embodiment of the present application;

[0024] Figure 2 It is a flow chart of a method for monitoring carbon emissions provided in an embodiment of the present application;

[0025] Figure 3 It is a schematic diagram of the framework of the energy consumption decomposition algorithm provided in the embodiment of the present application;

[0026] Figure 4 is a schematic diagram of a carbon emission topological structure diagram provided in an embodiment of the present application;

[0027] Figure 5 It is a structural schematic diagram of a carbon emission monitoring device provided by the present application;

[0028] Figure 6 It is a structural schematic diagram of a carbon emission terminal provided in this application. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0030] It should be noted that the carbon emission monitoring method provided in the present application is generally performed by a carbon emission terminal, and accordingly, the carbon emission monitoring device is generally arranged in the carbon emission terminal.

[0031] Figure 1 An exemplary system architecture that can be applied to the carbon emissions monitoring method or the carbon emissions monitoring device of the present application is shown.

[0032] like Figure 1 As shown, the system architecture may include: a user device 101 and a plurality of carbon emission terminals 102 deployed in a distributed manner. The user device 101 and the carbon emission terminal 102 may communicate with each other through a network, and the network is used as a medium for providing communication links between the above-mentioned units. The network may include various types of wired communication links or wireless communication links, for example: a wired communication link includes an optical fiber, a twisted pair or a coaxial cable, etc., and a wireless communication link includes a Bluetooth communication link, a wireless fidelity (WIreless-FIdelity, Wi-Fi) communication link or a microwave communication link, etc. The user device 101 may be a PC, a laptop, a mobile phone or other device.

[0033] Among them, the carbon emission terminal is deployed in a distributed manner at different locations in the energy consumption network. The carbon emission terminal includes a user interface, a self-organizing network module, an edge computing unit, a distributed data storage unit, a data acquisition unit and an energy consumption sensor. The energy consumption sensor is used to collect the total meter characteristic data of the energy consumption network, and each carbon emission terminal communicates through the self-organizing network module. The carbon emission terminal 102 is used to collect the total meter characteristic data of the energy consumption network in real time and identify all device types and the number of devices of each device type in the energy consumption network according to the total meter characteristic data, and then calculate the carbon emissions of each device type. The user device 101 reads the calculation results in the distributed data storage unit in the carbon emission terminal. The carbon emission terminal can periodically analyze the carbon emissions of each device type, for example: provide fine-grained (15 minutes) intelligent analysis of carbon emissions, and support analysis in space (region, device) and time (real-time, day, month, year) dimensions.

[0034] It should be understood that Figure 1 The number of carbon emission terminals, networks and energy consumption networks in the figure is only for illustration and can be any number of carbon emission terminals, networks and energy consumption networks according to the implementation requirements.

[0035] The following will be combined with the attached Figure 2, the carbon emission monitoring method provided in the embodiment of the present application is described in detail. Among them, the carbon emission monitoring device in the embodiment of the present application can be Figure 1 The carbon emission terminal is shown.

[0036] See also Figure 2 , which is a flow chart of a method for monitoring carbon emissions according to an embodiment of the present application. Figure 2 As shown, the method of the embodiment of the present application may include the following steps:

[0037] S201. Obtain total electricity meter characteristic data measured by a total electricity meter in an energy consumption network.

[0038] Among them, the carbon emission terminal needs to establish a communication connection with the total electricity meter in the energy consumption network, usually through Modbus, MQTT, HTTP and other protocols. According to the preset sampling frequency (such as once every 5 minutes), the current value, voltage value, active power, reactive power and other characteristic data are obtained through API calls or direct reading of the meter register. The collected data is stored in a local database or cloud storage service to ensure data persistence and queryability.

[0039] S202: Identify the total electricity meter characteristic data according to a plurality of pre-trained device identification models to obtain all device types and device quantities in the energy consumption network.

[0040] Among them, multiple equipment recognition models are pre-trained, and each model recognizes different types of equipment, such as air compressors, cranes, motors, etc. These models may be based on machine learning algorithms (such as SVM, decision trees) or deep learning models (such as CNN, LSTM). Key features for identification are extracted from the total meter feature data, such as power fluctuation patterns, current / voltage waveform features, etc. The extracted features are input into each equipment recognition model, and the model outputs the probability or confidence of each equipment type. According to the model output, combined with the preset threshold or logical judgment, the existence and quantity of each equipment type in the energy consumption network are determined.

[0041] In some embodiments of the present application, the total electric meter characteristic data is identified according to a plurality of pre-trained device identification models to obtain all device types and device quantities in the energy consumption network; the plurality of device identification models are used to identify different device types, including:

[0042] Acquire multiple instantaneous device numbers output by each device identification model within a preset time interval, and take the maximum value of the multiple instantaneous device numbers as the actual device number of the device identification model;

[0043] The actual number of devices in each device identification model is summarized to obtain all device types and device numbers in the energy consumption network.

[0044] Among them, the carbon emission terminal loads multiple pre-trained device identification models. These models may be machine learning models (such as random forests, SVM) or deep learning models (such as CNN, LSTM), which are specifically used to identify different types of energy-consuming devices. Allocate necessary computing resources (such as CPU, GPU, memory) to each model, and initialize necessary parameters and states. The carbon emission terminal extracts the characteristic data of the total electricity meter within a preset time interval from the database or real-time data stream, including current value, voltage value, active power, reactive power, etc. Remove outliers and fill in missing values ​​to ensure the quality of input data. According to the model requirements, transform the data or extract new features, such as power fluctuation features, time series features, etc.

[0045] The carbon emission terminal inputs the preprocessed feature data into each device identification model in batches. Each batch may contain data from multiple time points. Perform inference operations on each model and output the instantaneous number of each type of equipment at each time point. Since the model may output probability or confidence, it needs to be converted into a specific number of equipment through a preset threshold. For each device identification model, the instantaneous number of devices at all time points is collected within a preset time interval (such as one day or one week). Since the number of devices should be a fixed value or fluctuate within a certain range in practice, the maximum value of all instantaneous device numbers within the time interval is taken as the true number of devices for the device identification model. This step assumes that the device is active most of the time and the model can accurately identify its existence. The true number of devices obtained by each device identification model is summarized. Since each model is responsible for identifying different types of devices, the summary result will contain all device types and their corresponding numbers in the energy consumption network. Optionally, the summary result is verified, such as checking whether the device type is repeated and whether the number of devices is reasonable.

[0046] The carbon emission terminal will present the summarized equipment types and equipment quantities to the user or management system in the form of tables, charts or reports, and store the results in a database or file system for subsequent query and analysis.

[0047] Through the above steps, the carbon emission terminal can use multiple pre-trained device recognition models to identify the total electricity meter characteristic data and accurately determine all device types and their device quantities in the energy consumption network.

[0048] In some embodiments of the present application, a sample data set is obtained; the sample data set includes the number of devices of each device type and total electric meter characteristic data collected in multiple different time periods;

[0049] Using the SVM machine model algorithm to train the sample data set;

[0050] Using a random forest algorithm to train the sample data set;

[0051] A simple linear regression model is used to integrate the two training results to obtain an integrated device recognition model.

[0052] Among them, the carbon emission terminal obtains sample data sets from the historical data of the energy consumption network. These data should cover multiple different time periods to ensure that the model can learn the operation mode and characteristics of the equipment in different time periods.

[0053] The sample dataset should contain the following key information:

[0054] Timestamp: marks the time point when data is collected.

[0055] Equipment type and quantity: The actual number of each equipment type in a specific time period. This data may be obtained through manual recording, sensor monitoring, or intelligent recognition systems.

[0056] Total electric meter characteristic data: total electric meter characteristic data corresponding to the above time period, such as current value, voltage value, active power, reactive power, etc. For example, the various data in the sample data set are shown in Table 1.

[0057]

[0058] Table 1

[0059] The carbon emission terminal cleans and preprocesses the raw data, including removing noise, filling missing values, standardizing or normalizing features, etc., to improve data quality and reduce bias during model training.

[0060] Configure the SVM algorithm on the carbon emission terminal, and set key parameters such as the kernel function (such as linear kernel, polynomial kernel, RBF kernel, etc.), penalty parameter C, etc. Select features suitable for the SVM model from the preprocessed sample data set as input. These features should be able to distinguish the operating characteristics of different equipment types. Divide the sample data set into a training set and a validation set (or test set), and use the training set to train the SVM model. By adjusting the model parameters and iterative optimization process, the model can accurately identify different equipment types. Use the validation set or test set to evaluate the performance of the SVM model, including indicators such as accuracy, recall, and F1 score.

[0061] Configure the random forest algorithm on the carbon emission terminal and set key parameters such as the number of trees, maximum depth, and splitting criteria. Also select features suitable for the random forest algorithm from the preprocessed sample data set as input. Use the training set to train the random forest model. Random forest improves the accuracy and robustness of the model by building multiple decision trees and integrating their prediction results. Similar to the SVM model, use the validation set or test set to evaluate the performance of the random forest model.

[0062] Use a simple linear regression model to integrate the two training results:

[0063] First, obtain the identification results of the device type from the SVM model and the random forest model. These results may be the predicted value of the number of devices or the probability distribution of the device type. Make sure that the results output by the two models are one-to-one corresponding in time points and device types for subsequent integration processing. Although simple linear regression is usually used for the prediction of continuous values, it can be used here as a weighting or averaging strategy. Assign different weights to the outputs of the SVM model and the random forest model (these weights can be fixed or dynamically adjusted based on model performance), and then use a linear regression model (or a simpler weighted averaging method) to combine the outputs of the two models to obtain the final integrated device identification model output. Note: The "simple linear regression model" here may not actually perform the traditional linear regression fitting process, but is used as a metaphor for an integration strategy or result merging method. Finally, use an independent test set to evaluate the performance of the integrated device identification model to verify its accuracy and reliability in practical applications.

[0064] S203: Collect the energy consumption values ​​measured by the total electric meter according to a preset time period, and generate an energy consumption time series according to the collected data.

[0065] Among them, according to the preset time period (such as every 15 minutes, every hour), the energy consumption value of the total meter is collected through the API or directly reading the meter (usually the total electricity or total power integral). The collected energy consumption values ​​are arranged in chronological order to form an energy consumption time series. Each time point corresponds to an energy consumption value, which is recorded in the database.

[0066] S204: Decompose the energy consumption time series to obtain energy consumption values ​​of each identified device type.

[0067] Use machine learning-based decomposition algorithms (such as random forest regression, neural networks, etc.) or non-intrusive load monitoring (NILM) technology to decompose the energy consumption time series. If a machine learning method is used, it may be necessary to train the decomposition model using historical data and known equipment usage. The energy consumption time series is input into the decomposition algorithm, which decomposes the total energy consumption into the energy consumption of each equipment type based on information such as equipment type and operating characteristics.

[0068] Furthermore, in some embodiments of the present application, the energy consumption decomposition of the energy consumption time series to obtain the energy consumption value of each identified device type includes:

[0069] The attention mechanism and LSTM long-term memory network are used to decompose the energy consumption time series to obtain the energy consumption values ​​of each identified device type.

[0070] Among them, see Figure 3 The framework diagram of the energy consumption decomposition algorithm shown in FIG. 1 is a schematic diagram of the energy consumption decomposition algorithm. This application uses a combination of attention mechanism and LSTM to construct the energy consumption decomposition algorithm. LSTM is suitable for mining the time dependency of data, but its learning ability for large time intervals is insufficient. After adding the time attention mechanism, redundant features can be removed and different weights can be given to the relationship between different time points, thereby making up for the shortcomings of LSTM in extracting energy consumption time series features.

[0071] The following is a description of each module in the framework:

[0072] The convolution layer is used to initially integrate data features. Since the electricity data on which carbon emissions are based often has a high sampling rate, the dilated convolution algorithm is used for convolution calculation, so that the features of the data at a long distance can be extracted without consuming too many computing resources.

[0073] The feature self-attention mechanism eliminates redundant features by assigning different weights to primary features, and further extracts high-level features related to emission decomposition. The feature self-attention mechanism simplifies the general self-attention algorithm, reduces the computational complexity, and improves the computational efficiency to meet the needs of real-time decomposition. Different from the general self-attention mechanism algorithm's processing of query vector Q, key vector K, and value vector V, the feature self-attention mechanism randomly selects nlnn vectors from K, reorganizes them in order to obtain K' for querying the sparsity M of Q, and then extracts the vector composition Q' from Q according to the index of the first clnn maximum values ​​in M ​​to calculate the feature weight. The feature weight W can be expressed as:

[0074]

[0075] , where d is the feature dimension. The primary features are added according to W to obtain the high-level features.

[0076] The LSTM layer is used to extract primary temporal features. Through memory units and gated network structures, it captures short-term to medium-term temporal dependencies, selectively forgets irrelevant information, and retains relevant temporal dependency features. This decomposition algorithm adopts the LSTM-batch normalization-LSTM structure, in which the double-layer LSTM structure increases the depth of the LSTM network and enhances the network's memory, allowing the network to learn more complex and abstract features. The batch normalization module reduces the internal covariate drift, reduces the gradient explosion and vanishing problems, enhances the generalization ability of the algorithm, and improves the training speed of the network.

[0077] The Temporal Pattern Attention module is used to make up for the shortcomings of the aforementioned LSTM network in extracting ultra-long-distance features. By calculating the hidden state vector output by the LSTM, the Temporal Pattern Attention mechanism assigns different weights to all time steps, thereby extracting more global features and more flexibly screening and integrating primary temporal features into advanced temporal features, thereby forming a feature vector closely related to the decomposition task.

[0078] As the number of network layers increases, the later network structures are more difficult to be effectively trained. Therefore, a residual connection structure is added between the aforementioned network layers, that is, a skip connection is introduced, thereby reducing the path of gradient propagation, so that the deep network structure can also be fully trained.

[0079] The final fully connected layer of the algorithm is used to linearly map the features output by the temporal attention mechanism to form the final decomposition result as the output result of the energy consumption decomposition algorithm.

[0080] S205. Calculate the carbon emission of each equipment type according to the energy consumption value and carbon emission factor of each equipment type.

[0081] Among them, the carbon emission terminal can obtain the carbon emission factor of each equipment type from the authoritative database or official source, that is, the carbon dioxide emissions corresponding to the unit energy consumption. The energy consumption value of each equipment type is multiplied by its corresponding carbon emission factor to obtain the carbon emissions of each equipment type.

[0082] In an embodiment of the present application, a carbon emission topology diagram is generated based on the calculated carbon emissions of each device type; the carbon emission topology diagram is composed of multiple layers, each layer includes at least one node, each node is associated with a carbon emission, and each node in the bottom layer is associated with a device type.

[0083] Among them, the carbon emission association network between devices is automatically constructed by using the calculated carbon emissions of each device type. The carbon emission topology diagram can help users understand the mutual influence between devices and manage carbon emissions more effectively. After the topology is built, its accuracy and reliability need to be verified through actual data. This step is key to ensuring that we can make decisions based on accurate information. Visualizing the carbon emission topology diagram can not only help users understand the distribution of carbon emissions more intuitively, but also provide intuitive support for decision makers.

[0084] For example, see Figure 4 In the carbon emission topology diagram shown, the energy consumption network is an electrolytic aluminum production line. The root node of the structure diagram is the total emission of the energy consumption network, the second-level node is the equipment category, and the third-level node is all the equipment types identified in the energy consumption network. When it is detected that the user clicks on a node or moves the mouse to a node, the carbon emission of the node will be displayed. The carbon emission can also be displayed directly on the node. The carbon emission of a node is the sum of the carbon emissions of all its child nodes.

[0085] S206: Accumulate the carbon emissions of each device type to obtain the total carbon emissions of the energy consumption network.

[0086] The carbon emission terminal accumulates the carbon emissions of each type of equipment to obtain the total carbon emissions of the entire energy consumption network. The total carbon emissions are presented to users or management systems in the form of visual reports, email notifications or API interfaces for subsequent carbon emission management and optimization.

[0087] In some embodiments of the present application, it also includes:

[0088] Obtain a historical carbon emission time series for each equipment type; wherein the historical carbon emission time series includes multiple historical time points and the carbon emission amount at each historical time point;

[0089] Predicting the carbon emissions of the equipment type at a future time point based on the historical carbon emissions time series;

[0090] If the predicted carbon emissions are greater than the preset threshold, an early warning message will be issued.

[0091] Among them, in order to understand the carbon emissions of equipment types in different time periods, their historical carbon emissions data are collected. These data include multiple historical time points (such as daily, weekly, monthly, etc.) and the carbon emissions at the corresponding time points. These data are the basis for subsequent forecasting and analysis. Historical carbon emission data may come from the calculation results of S205, the sensors that come with the equipment, environmental monitoring stations, and the carbon emission recording system within the enterprise. Using algorithms such as statistics, machine learning or deep learning, the historical carbon emission time series is analyzed to identify trends, seasonal changes, periodic laws, etc. Then, based on these laws, a prediction model is built to predict carbon emissions at a certain point in the future. The prediction accuracy is affected by many factors, including the integrity of historical data, the complexity of the prediction model, and the interference of external factors (such as policy changes, technological advances, etc.).

[0092] According to the carbon emission standards stipulated by the enterprise, industry or government, a reasonable warning threshold is set. This threshold can be an absolute value (such as no more than XX tons per year) or a relative value (such as a reduction of XX% compared to last year). When the forecast results show that the carbon emissions at a certain point in the future will exceed the preset threshold, the system automatically triggers the warning mechanism and sends a warning message to relevant personnel (such as environmental protection departments, enterprise managers, equipment operators, etc.). The method of sending the warning message can be text messages, emails, or application push messages. After receiving the warning message, the relevant personnel can take corresponding countermeasures, such as adjusting production plans, optimizing equipment efficiency, upgrading low-carbon technologies, etc., to reduce carbon emissions and ensure that the company meets environmental protection requirements. This embodiment collects historical carbon emission data of equipment types, uses a prediction model to predict future carbon emissions, and issues a warning prompt when the prediction results exceed the preset threshold, providing enterprises with an effective means of carbon emission management. This helps enterprises better understand carbon emissions, take timely countermeasures, and achieve low-carbon production and sustainable development.

[0093] The embodiments of the present application have the following beneficial effects when evaluating the risk of a bond:

[0094] By automatically acquiring the characteristic data measured by the total electric meter (such as current, voltage, active power, reactive power, etc.), and using multiple pre-trained device identification models for intelligent analysis, the automatic identification of the type and quantity of equipment in the energy consumption network is realized. This process greatly reduces manual intervention, improves the efficiency and accuracy of data processing, and enables carbon emission accounting to be carried out automatically and continuously. Using the trained device identification model, different types of equipment and their quantities in the energy consumption network can be accurately distinguished, providing a solid foundation for subsequent energy consumption decomposition. Through the construction and analysis of energy consumption time series, the accurate decomposition of energy consumption of various types of equipment is further realized, ensuring the precision and accuracy of carbon emission calculation. Combining the energy consumption value of each type of equipment and the corresponding carbon emission factor, this technical solution can accurately calculate the carbon emissions of each type of equipment. This accounting method based on actual energy consumption data avoids the errors that may be caused by traditional estimation methods, and provides more reliable data support for the formulation and implementation of energy conservation and emission reduction policies. By accumulating the carbon emissions of each type of equipment, the total carbon emissions of the energy consumption network are obtained, realizing a comprehensive control of the carbon emissions of the entire network. This result not only helps companies understand their own carbon emissions, but also provides a strong basis for formulating scientific emission reduction targets and strategies.

[0095] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0096] See also Figure 5 , which shows a schematic diagram of the structure of a carbon emission monitoring device provided by an exemplary embodiment of the present application, hereinafter referred to as device 5. The device 5 can be implemented as all or part of a carbon emission terminal through software, hardware or a combination of both. The device 5 includes: an acquisition unit 501, an identification unit 502, a collection unit 503, a decomposition unit 504, a calculation unit 505 and a summary unit 506.

[0097] The acquisition unit 501 is used to acquire total meter characteristic data measured by a total meter in an energy consumption network; the total meter characteristic data includes: one or more of current value, voltage value, active power and reactive power;

[0098] The identification unit 502 is used to identify the total electric meter characteristic data according to a plurality of pre-trained device identification models to obtain all device types and device quantities in the energy consumption network; the plurality of device identification models are used to identify different device types;

[0099] The collection unit 503 is used to collect the energy consumption values ​​measured by the total electric meter according to a preset time period, and generate an energy consumption time series according to the collected data; the energy consumption time series includes multiple time points and energy consumption values ​​associated with each time point;

[0100] A decomposition unit 504 is used to perform energy consumption decomposition on the energy consumption time series to obtain energy consumption values ​​of each identified device type;

[0101] A calculation unit 505 is used to calculate the carbon emission of each device type according to the energy consumption value and carbon emission factor of each device type;

[0102] The summarizing unit 506 is used to accumulate the carbon emissions of each device type to obtain the total carbon emissions of the energy consumption network.

[0103] In one or more possible embodiments, the total electric meter characteristic data is identified according to a plurality of pre-trained device identification models to obtain all device types and device quantities in the energy consumption network; the plurality of device identification models are used to identify different device types, including:

[0104] Acquire multiple instantaneous device numbers output by each device identification model within a preset time interval, and take the maximum value of the multiple instantaneous device numbers as the actual device number of the device identification model;

[0105] The actual number of devices in each device identification model is summarized to obtain all device types and device numbers in the energy consumption network.

[0106] In one or more possible embodiments, it further includes:

[0107] A training unit, used to obtain a sample data set; the sample data set includes the number of devices of each device type and total electric meter characteristic data collected in multiple different time periods;

[0108] Using the SVM machine model algorithm to train the sample data set;

[0109] Using a random forest algorithm to train the sample data set;

[0110] A simple linear regression model is used to integrate the two training results to obtain an integrated device recognition model.

[0111] In one or more possible embodiments, the step of performing energy consumption decomposition on the energy consumption time series to obtain energy consumption values ​​of each identified device type includes:

[0112] The attention mechanism and LSTM long-term memory network are used to decompose the energy consumption time series to obtain the energy consumption values ​​of each identified device type.

[0113] In one or more possible embodiments, it further includes:

[0114] A generating unit is used to generate a carbon emission topology diagram according to the calculated carbon emissions of each device type; the carbon emission topology diagram is composed of multiple layers, each layer includes at least one node, each node is associated with a carbon emission, and each node in the bottom layer is associated with a device type.

[0115] In one or more possible embodiments, it further includes:

[0116] An early warning unit, used to obtain a historical carbon emission time series for each equipment type; wherein the historical carbon emission time series includes multiple historical time points and the carbon emission amount at each historical time point;

[0117] Predicting the carbon emissions of the equipment type at a future time point based on the historical carbon emissions time series;

[0118] If the predicted carbon emissions are greater than the preset threshold, an early warning message will be issued.

[0119] It should be noted that when the device 5 provided in the above embodiment executes the carbon emissions monitoring method, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above functions. In addition, the carbon emissions monitoring device provided in the above embodiment and the carbon emissions monitoring method embodiment belong to the same concept. The implementation process is detailed in the method embodiment, which will not be repeated here.

[0120] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0121] The present application also provides a computer storage medium, which can store multiple instructions, and the instructions are suitable for being loaded and executed by a processor as described above. Figure 2 The method steps of the embodiment shown in the figure can be found in the specific implementation process. Figure 2 The specific description of the illustrated embodiment will not be repeated here.

[0122] The present application also provides a computer program product, which stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the carbon emissions monitoring method described in the above embodiments.

[0123] See also Figure 6 , which is a schematic diagram of the structure of a carbon emission terminal provided in the embodiment of the present application. Figure 6 As shown, the carbon emission terminal 600 may include: at least one processor 601 , at least one network interface 604 , a user interface 603 , a memory 605 , and at least one communication bus 602 .

[0124] The communication bus 602 is used to realize the connection and communication between these components.

[0125] The user interface 603 may include a display screen (Display) and a camera (Camera), and the optional user interface 603 may also include a standard wired interface and a wireless interface.

[0126] The network interface 604 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0127] Among them, the processor 601 may include one or more processing cores. The processor 601 uses various interfaces and lines to connect the various parts of the entire carbon emission terminal 600, and executes various functions and processes data of the carbon emission terminal 600 by running or executing instructions, programs, code sets or instruction sets stored in the memory 605, and calling data stored in the memory 605. Optionally, the processor 601 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 601 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 601, and it can be implemented separately through a chip.

[0128] Among them, the memory 605 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 605 includes a non-transitory computer-readable storage medium. The memory 605 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 605 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 605 may optionally be at least one storage device located away from the aforementioned processor 601. As Figure 6 As shown, the memory 605 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program.

[0129] exist Figure 6 In the carbon emission terminal 600 shown in FIG. 1 , the user interface 603 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 601 can be used to call the application stored in the memory 605 and specifically execute the following steps: Figure 6 The specific process can be referred to Figure 2 As shown, no further description is given here.

[0130] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0131] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A method for monitoring carbon emissions, characterized in that: include: Obtaining total electric meter characteristic data measured by a total electric meter in an energy consumption network; The total electric meter characteristic data includes: one or more of current value, voltage value, active power and reactive power; Identify the total electric meter characteristic data according to a plurality of pre-trained device identification models to obtain all device types and device quantities in the energy consumption network; the plurality of device identification models are used to identify different device types; Collecting the energy consumption values ​​measured by the total electric meter according to a preset time period, and generating an energy consumption time series according to the collected data; the energy consumption time series includes multiple time points and energy consumption values ​​associated with each time point; Performing energy consumption decomposition on the energy consumption time series to obtain energy consumption values ​​of each identified device type; Calculate the carbon emissions of each equipment type based on the energy consumption value and carbon emission factor of each equipment type; The carbon emissions of each equipment type are accumulated to obtain the total carbon emissions of the energy consumption network.

2. The method according to claim 1, characterized in that The total electric meter characteristic data is identified according to a plurality of pre-trained device identification models to obtain all device types and device quantities in the energy consumption network; The multiple device identification models are used to identify different device types, including: Acquire multiple instantaneous device numbers output by each device identification model within a preset time interval, and take the maximum value of the multiple instantaneous device numbers as the actual device number of the device identification model; The actual number of devices in each device identification model is summarized to obtain all device types and device numbers in the energy consumption network.

3. The method according to claim 1 or 2, characterized in that: Before obtaining the total electric meter characteristic data measured by the total electric meter in the energy consumption network, the method further includes: Acquire a sample data set; the sample data set includes the number of devices of each device type and total electric meter characteristic data collected in multiple different time periods; Using the SVM machine model algorithm to train the sample data set; Using a random forest algorithm to train the sample data set; A simple linear regression model is used to integrate the two training results to obtain an integrated device recognition model.

4. The method according to claim 3, characterized in that The energy consumption decomposition of the energy consumption time series to obtain energy consumption values ​​of each identified device type includes: The attention mechanism and LSTM long-term memory network are used to decompose the energy consumption time series to obtain the energy consumption values ​​of each identified device type.

5. The method according to claim 1, 2 or 4, characterized in that: Also includes: Generate a carbon emission topology diagram based on the calculated carbon emission of each equipment type; The carbon emission topology structure diagram is composed of multiple layers, each layer includes at least one node, each node is associated with a carbon emission amount, and each node in the bottom layer is associated with a device type.

6. The method according to claim 5, characterized in that Also includes: Obtain a historical carbon emission time series for each equipment type; wherein the historical carbon emission time series includes multiple historical time points and the carbon emission amount at each historical time point; Predicting the carbon emissions of the equipment type at a future time point based on the historical carbon emissions time series; If the predicted carbon emissions are greater than the preset threshold, an early warning message will be issued.

7. The method according to claim 6, characterized in that The warning prompt message is a text message, email or application push message.

8. A carbon emission monitoring device, characterized in that: include: An acquisition unit, used for acquiring total electric meter characteristic data measured by a total electric meter in an energy consumption network; The total electric meter characteristic data includes: one or more of current value, voltage value, active power and reactive power; An identification unit, used to identify the characteristic data of the total electric meter according to a plurality of pre-trained device identification models, and obtain all device types and device quantities in the energy consumption network; the plurality of device identification models are used to identify different device types; A collection unit, used to collect the energy consumption values ​​measured by the total electric meter according to a preset time period, and generate an energy consumption time series according to the collected data; the energy consumption time series includes multiple time points and energy consumption values ​​associated with each time point; A decomposition unit, used for performing energy consumption decomposition on the energy consumption time series to obtain energy consumption values ​​of each identified device type; A calculation unit, used for calculating the carbon emission of each equipment type according to the energy consumption value and carbon emission factor of each equipment type; The summarizing unit is used to accumulate the carbon emissions of each device type to obtain the total carbon emissions of the energy consumption network.

9. A computer storage medium, characterized in that The computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 7.

10. A carbon emission terminal, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method steps as claimed in any one of claims 1 to 7.

Images