A coal-fired carbon emission data real-time accounting method, device, equipment and medium

Abstract

The present application belongs to the technical field of coal-fired carbon emission monitoring, and discloses a coal-fired carbon emission data real-time accounting method, device, equipment and medium. The real-time carbon emission amount during coal combustion is calculated, the basic comprehensive score is obtained by calculating the credibility comprehensive sub-score weighted summation of each key index parameter during coal combustion, the rationality score of parameter combination is obtained by correlation analysis of all key index parameters, the credibility comprehensive score of real-time carbon emission amount is calculated according to the basic comprehensive score and the rationality score, if the preset level is reached, the corresponding carbon emission data credibility report is generated, if the preset level is not reached, warning is performed according to the corresponding level, the credibility score of each key index parameter during coal combustion can be performed, the rationality score obtained by correlation analysis of all key index parameters is combined, the credibility comprehensive score is comprehensively calculated, thereby realizing real-time accounting of carbon emission data and improving the credibility of carbon emission data.

Description

Technical Field

[0001] This invention relates to the field of coal-fired carbon emission monitoring technology, specifically to a method, apparatus, equipment, and medium for real-time calculation of coal-fired carbon emission data. Background Technology

[0002] Currently, carbon emission data collection from coal-fired power plants relies on manual operation, which suffers from problems such as opaque sampling, susceptibility to data tampering, and delayed accounting, resulting in insufficient credibility of carbon emission data. Carbon emission accounting mostly adopts annual inventory and empirical factor methods, which are time-consuming and inaccurate, failing to meet the needs of real-time supervision and carbon market trading. On the other hand, the existing system suffers from severe data silos and lacks full-chain reliable traceability capabilities, leading to insufficient public trust in carbon data and posing risks in compliance reviews and carbon trading.

[0003] In summary, existing carbon emission data from coal-fired power plants cannot be calculated in real time and lacks reliability. Summary of the Invention

[0004] To address the aforementioned shortcomings, the present invention aims to provide a method, apparatus, electronic device, and storage medium for real-time calculation of coal-fired carbon emission data, which can achieve real-time calculation and improve reliability.

[0005] The first aspect of this invention discloses a method for real-time calculation of carbon emission data from coal combustion, comprising:

[0006] The emission factor is determined, the instantaneous flow rate of coal fed into the furnace during the combustion of coal-fired fossil fuels is obtained, and the real-time carbon emissions are calculated by multiplying the instantaneous flow rate of coal fed into the furnace by the emission factor.

[0007] Calculate the comprehensive sub-score of the reliability of various key indicator parameters during the combustion of coal and fossil fuels;

[0008] The basic comprehensive score is obtained by weighted summation of the credibility sub-scores of all the key indicator parameters.

[0009] A correlation analysis was performed on all the key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters.

[0010] Based on the basic comprehensive score and the reasonableness score, the credibility comprehensive score of the real-time carbon emissions is calculated.

[0011] If the overall credibility score reaches a preset level, a corresponding carbon emission data credibility report will be generated.

[0012] If the overall credibility score does not reach the preset level, an early warning will be issued according to the corresponding level.

[0013] Among them, a correlation analysis is performed on all the key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters, including:

[0014] Based on all the key indicator parameters, a multimodal time series vector is constructed;

[0015] The multimodal time-series vectors are input into a pre-trained scoring model, and the rationality score of the combination of key indicator parameters is obtained based on the output of the scoring model.

[0016] In some embodiments, determining the emission factor includes:

[0017] Real-time assessment of whether the testing process for coal and fossil fuel samples is complete;

[0018] If the testing process is not completed, the simulated carbon content is predicted, and the product of the simulated carbon content and the carbon oxidation rate is used as the emission factor.

[0019] If the testing process has been completed, obtain the measured elemental carbon content of the coal, and use the product of the elemental carbon content and the carbon oxidation rate as the emission factor.

[0020] In some embodiments, the simulated carbon content is predicted, including:

[0021] Determine whether a coal-fired power plant has online coal quality analysis equipment;

[0022] If the online coal quality analysis equipment is available, the simulated carbon content is determined based on the online detection results of the online coal quality analysis equipment.

[0023] If the online coal quality analysis equipment is not available, the simulated carbon content is determined based on the historical test results of a specified batch of incoming coal samples; or, the simulated carbon content is determined based on the historical test results of the previous period; or, a value that meets the conditions is selected from the local database as the simulated carbon content.

[0024] In some embodiments, the reliability comprehensive sub-score for various key indicator parameters during the combustion of coal-fired fossil fuels is calculated, including:

[0025] Determine whether the key parameters during the combustion of coal-fired fossil fuels are coal quality parameters;

[0026] If the key indicator parameter is not the coal quality indicator parameter, calculate the traceability credibility score of the key indicator parameter and use the traceability credibility score as its credibility comprehensive sub-score.

[0027] If the key indicator parameter is the coal quality indicator parameter, calculate the traceability credibility score and the indicator credibility score of the key indicator parameter, and use the sum of the two as its credibility comprehensive sub-score.

[0028] The key indicator parameters include at least fuel characteristic parameters, fuel flow parameters, energy flow parameters, and equipment operating parameters. The fuel characteristic parameters include at least the elemental carbon content of coal, the lower heating value of coal, the calorific value of coal, moisture, volatile matter, ash content, and hydrogen content. The fuel flow parameters include at least the instantaneous flow rate of coal entering the furnace. The energy flow parameters include at least the boiler load rate, the turbine power generation, and the grid-connected power generation. The coal quality indicator parameters include at least the elemental carbon content of coal, the calorific value of coal, the moisture content, the volatile matter, and the ash content.

[0029] In some embodiments, the scoring model includes a coupled deep convolutional autoencoder network and an adversarial generative network, wherein the deep convolutional autoencoder network includes an encoder and a decoder connected in sequence, and the adversarial generative network includes a generator and a discriminator connected in sequence; and, inputting the multimodal temporal vector into a pre-trained scoring model, obtaining a reasonableness score for the combination of key indicator parameters based on the output of the scoring model, including:

[0030] The multimodal temporal vector is input into a pre-trained scoring model. The encoder performs a convolution operation on the multimodal temporal vector to extract the target spatiotemporal features. The target spatiotemporal features are then compressed into the latent space to obtain key feature representations.

[0031] The key feature representation is upsampled by the decoder to reconstruct the target reconstruction vector;

[0032] The reconstruction error is calculated based on the difference between the target reconstruction vector and the multimodal time-series vector;

[0033] The discriminator outputs the confidence level that the target reconstructed vector belongs to the true distribution;

[0034] Based on the reconstruction error and the confidence level, a reasonableness score for the combination of key indicator parameters is output.

[0035] The second aspect of this invention discloses a real-time coal combustion carbon emission data calculation device, comprising:

[0036] Factor determination unit, used to determine emission factors;

[0037] The coal flow metering unit is used to obtain the instantaneous flow rate of coal entering the furnace during the combustion of coal and fossil fuels.

[0038] The calculation unit is used to calculate the real-time carbon emissions by multiplying the instantaneous flow rate of the coal entering the furnace by the emission factor;

[0039] The first scoring unit is used to calculate the comprehensive sub-score of the credibility of various key indicator parameters during the combustion of coal and fossil fuels;

[0040] The summation unit is used to perform a weighted summation of the credibility sub-scores of all the key indicator parameters to obtain the basic comprehensive score;

[0041] The correlation analysis unit is used to perform correlation analysis between all the key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters.

[0042] The second scoring unit is used to calculate the credibility score of the real-time carbon emissions based on the basic comprehensive score and the rationality score.

[0043] The report generation unit is used to generate a corresponding carbon emission data credibility report when the credibility comprehensive score reaches a preset level.

[0044] The early warning unit is used to issue an early warning according to the corresponding level when the overall credibility score does not reach the preset level.

[0045] The correlation analysis unit includes:

[0046] Construct sub-units to generate multimodal time-series vectors based on all the aforementioned key indicator parameters;

[0047] The scoring subunit is used to input the multimodal time-series vector into a pre-trained scoring model and obtain a reasonableness score for the combination of key indicator parameters based on the output of the scoring model.

[0048] In some embodiments, the factor determination unit includes:

[0049] The judgment subunit is used to determine in real time whether the testing process for coal-fired fossil fuel samples has been completed.

[0050] The simulation subunit is used to predict the simulated carbon content before the testing process is completed;

[0051] The first determining subunit is used to use the product of the simulated carbon content and the carbon oxidation rate as the emission factor;

[0052] The second determining subunit is used to obtain the measured carbon content of coal after the testing process is completed, and to use the product of the carbon content of coal and the carbon oxidation rate as the emission factor.

[0053] In some embodiments, the scoring model includes a coupled deep convolutional autoencoder network and an adversarial generative network, wherein the deep convolutional autoencoder network includes an encoder and a decoder connected in sequence, and the adversarial generative network includes a generator and a discriminator connected in sequence; the scoring subunit includes:

[0054] The input module is used to input the multimodal temporal vector into a pre-trained scoring model, extract target spatiotemporal features by performing convolution operations on the multimodal temporal vector through the encoder, and compress the target spatiotemporal features into the latent space to obtain key feature representations;

[0055] The reconstruction module is used to upsample the key feature representation through the decoder to reconstruct the target reconstruction vector;

[0056] An error calculation module is used to calculate the reconstruction error based on the difference between the target reconstruction vector and the multimodal temporal vector.

[0057] The discriminant module is used to output the confidence level that the target reconstructed vector belongs to the true distribution through the discriminator;

[0058] The output module is used to output a reasonableness score for the combination of key indicator parameters based on the reconstruction error and the confidence level.

[0059] A third aspect of the present invention discloses an electronic device, including a memory storing executable program code and a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the real-time calculation method for coal-fired carbon emission data disclosed in the first aspect.

[0060] The fourth aspect of this invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the real-time calculation method for coal-fired carbon emission data disclosed in the first aspect.

[0061] Compared with existing technologies, the beneficial effects of this invention are as follows: It calculates the real-time carbon emissions during the combustion of coal-fired fossil fuels; it calculates a basic comprehensive score by weighted summation of the comprehensive reliability sub-scores of various key indicator parameters during coal-fired fossil fuel combustion; it performs parameter correlation analysis on all key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters; based on the basic comprehensive score and the reasonableness score, it calculates the comprehensive reliability score of real-time carbon emissions; if the comprehensive reliability score reaches a preset level, a corresponding carbon emission data reliability report is generated; if the comprehensive reliability score does not reach the preset level, an early warning is issued according to the corresponding level. This invention can perform reliability scoring on various key indicator parameters during the combustion of coal-fired fossil fuels, and combine this with the reasonableness score obtained from parameter correlation analysis on all key indicator parameters to comprehensively calculate the comprehensive reliability score of real-time carbon emissions, thereby achieving real-time accounting of coal-fired carbon emission data and improving the reliability of coal-fired carbon emission data. Attached Figure Description

[0062] Figure 1This is a flowchart of a real-time calculation method for coal-fired carbon emissions disclosed in this invention;

[0063] Figure 2 This is a schematic diagram of the structure of a real-time coal-fired carbon emission data calculation device disclosed in this invention;

[0064] Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in this invention;

[0065] Figure 4 This is a schematic diagram of the structure of a computer device disclosed in this invention.

[0066] Explanation of reference numerals in the attached figures:

[0067] 201. Factor determination unit; 202. Coal flow measurement unit; 203. Accounting unit; 204. First scoring unit; 205. Summation unit; 206. Correlation analysis unit; 207. Second scoring unit; 208. Report generation unit; 209. Early warning unit; 301. Memory; 302. Processor. Detailed Implementation

[0068] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0069] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 110, 120, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0070] It will be understood by those skilled in the art that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0071] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those skilled in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0072] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Throughout the description, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0073] In this embodiment of the invention, data acquisition sensors are installed at key equipment nodes in coal-fired power plants, including coal weighbridges at the plant entrance and furnace feed, transfer vehicles, samplers, sample preparation machines, belt scales, coal feeders, coal quality analyzers, and sample storage cabinets, for coal processing, coal flow metering, and coal sample control. Through Internet of Things (IoT) technology and using common industrial protocols such as Modbus and TCP / IP, covering aspects such as equipment data interfaces, open protocols, and system compatibility, real-time automatic data acquisition from equipment terminals throughout the entire coal management process is achieved.

[0074] The Internet of Things (IoT) sensing technology enables comprehensive data capture of equipment information throughout the entire fuel process, from sampling, sample preparation, and testing to sample storage. Information and automation deployment is implemented for coal sampling and preparation equipment, coal flow metering equipment, and coal quality testing equipment, ensuring seamless integration between these devices and the fuel information management platform. This includes data collection and uploading of equipment verification / calibration status, maintenance / repair records, and operational information. Data acquisition sensors are deployed on key testing equipment such as industrial analyzers and calorimeters to collect real-time data on coal consumption and coal quality indicators, ensuring interconnectivity between coal flow metering equipment, coal quality testing equipment, and the fuel information management platform.

[0075] Throughout the entire process of sampling, sample preparation, testing, and storage, especially in key production stages of the power plant such as fuel transportation, boiler operation, and flue gas emission monitoring points, high-definition video monitoring systems and intelligent video analysis systems are deployed. The high-definition video monitoring system uses a 5G network to transmit real-time monitoring videos of each process operation and coal sample status images, improving the overall control and supervision capabilities. The intelligent video analysis system, based on computer vision technology, automatically identifies abnormal equipment conditions, such as belt misalignment, pipeline leaks, and personnel violations, and issues real-time alarms. Simultaneously, regular spot checks of the videos are conducted, and timestamps are correlated with carbon emission data to ensure the authenticity of the data collection process.

[0076] In power plant processes such as fuel acceptance and material entry / exit, high-resolution cameras are deployed to capture images of paper ledgers in real time. Optical character recognition (OCR) technology is then used to automatically identify these images, with deep learning models further enhancing accuracy. The identified ledger data is compared with automatically collected data, and any anomalies trigger a manual review process. Simultaneously, biometric technologies such as facial recognition and fingerprint recognition are used to verify the identity of personnel entering data for manual review, requiring electronic signature confirmation upon completion.

[0077] Intelligent control robots and drones are deployed in key locations such as sampling rooms, sample preparation rooms, and testing rooms to perform regular or pre-defined inspections, recording equipment operation and personnel procedures during carbon emission processes. Inspection frequency is dynamically adjusted based on risk levels. When data reliability is detected to be at risk, inspection frequency and monitoring intensity are increased for specific stages and operational locations.

[0078] Automatically collected sensor data, including carbon measurement data, monitoring videos, and captured images, is automatically transmitted, uploaded, encrypted, and stored tamper-proofly. Specifically, a private blockchain network is constructed, utilizing the distributed ledger and encryption mechanisms of blockchain to store key information such as raw carbon emission data, inspection and measurement process logs, and image recognition results in block format. Furthermore, hash encryption technology is used to sign operation logs, recording user operation time, content, IP address, and other information. Operation logs are stored on blockchain nodes, with precise timestamps and continuous log storage, ensuring data traceability. This allows for the traceability of a single sample's entire lifecycle from entry into the factory to submission for testing, improving the credibility of carbon data and providing an immutable chain of legal evidence for carbon accounting disputes.

[0079] The system protects the security and privacy of data during operation. Through encryption isolation and hierarchical access control mechanisms, it provides a legally compliant and fully traceable emergency monitoring channel for carbon emission data verification, ensuring the data sovereignty and privacy of enterprises unconditionally, thus achieving a balance between data protection and controllable supervision.

[0080] Real-time monitoring, fault early warning, and remote monitoring are employed to ensure the safety and stability of energy use in power plants. A data disaster recovery deployment is implemented, with two independent backup databases deployed locally for the power plant's carbon metering equipment. Real-time data replication technology is used to achieve data synchronization, ensuring continuous system operation and data security, and improving the stability and reliability of carbon data storage.

[0081] Please see Figure 1 This invention discloses a method for real-time calculation of coal-fired carbon emissions. The method can be executed by electronic devices such as computers, laptops, or tablets, or by a real-time coal-fired carbon emissions calculation device embedded in an electronic device; this invention does not limit the specific device to these. The electronic device is communicatively connected to the aforementioned fuel information management platform.

[0082] like Figure 1 As shown, the method includes the following steps 110-170:

[0083] 110. Determine the emission factor, obtain the instantaneous flow rate of coal entering the furnace during the combustion of coal-fired fossil fuels, and calculate the real-time carbon emissions by multiplying the instantaneous flow rate of coal entering the furnace by the emission factor.

[0084] The instantaneous flow rate of coal fed into the furnace is used to characterize the real-time consumption of coal, which can be detected in real time by coal flow metering equipment, such as belt scales. The real-time carbon emissions are calculated by multiplying the instantaneous flow rate of coal fed into the furnace by an emission factor, where the emission factor is the product of the elemental carbon content of the coal and the carbon oxidation rate. The carbon oxidation rate is usually assumed to be 99%, while the elemental carbon content of the coal needs to be obtained through national standard methods of laboratory chemical analysis. However, due to the time lag of chemical analysis methods, it is not yet possible to output the elemental carbon content test results in real time. Therefore, before the coal sample test data is generated, a simulated carbon content can be temporarily used to determine the emission factor, depending on the actual situation.

[0085] Specifically, in step 110, the implementation method for determining the emission factor may include:

[0086] The system can determine in real time whether the testing process for coal fossil fuel samples is complete. If the testing process is not complete, the simulated carbon content is predicted, and the product of the simulated carbon content and the carbon oxidation rate is used as the emission factor. If the testing process is complete, the system obtains the measured elemental carbon content of the coal, and the product of the elemental carbon content of the coal and the carbon oxidation rate is used as the emission factor.

[0087] The process of predicting the simulated carbon content includes the following steps (not shown): S11-S13.

[0088] S11. Determine whether the coal-fired power plant has online coal quality analysis equipment; wherein, the online coal quality analysis equipment performs online detection based on technologies such as laser-induced breakdown spectroscopy, near-infrared spectroscopy, or X-ray fluorescence spectroscopy. If yes, proceed to step S12; otherwise, proceed to step S13.

[0089] S12. If online coal quality analysis equipment is available, the simulated carbon content shall be determined based on the online detection results of the online coal quality analysis equipment.

[0090] S13. If there is no online coal quality analysis equipment, the simulated carbon content shall be determined based on the historical test results of the specified batch of incoming coal samples; or, the simulated carbon content shall be determined based on the historical test results of the previous period; or, a value that meets the conditions shall be selected from the local database as the simulated carbon content.

[0091] By predicting and simulating carbon content, the data gaps during fuel sample testing can be compensated for. Once the coal sample testing data for the corresponding period is officially generated, the emission factor is determined based on the measured elemental carbon content of the coal, thereby achieving real-time calculation and output of carbon emissions.

[0092] 120. Calculate the comprehensive sub-score of the reliability of various key indicator parameters during the combustion of coal fossil fuels.

[0093] Key indicators include fuel characteristic parameters, fuel flow parameters, energy flow parameters, and equipment operating parameters. Among them, fuel characteristic parameters include at least the elemental carbon content of coal, lower heating value of coal, calorific value of coal, moisture, volatile matter, ash content, and hydrogen content. Fuel flow parameters include at least the instantaneous flow rate of coal entering the furnace. Energy flow parameters include at least the boiler load rate, turbine power generation, and grid-connected power generation.

[0094] Specifically, step 120 may include the following steps 1201-1203 (not shown):

[0095] 1201. Determine whether the key parameters during the combustion of coal fossil fuels are coal quality parameters;

[0096] 1202. If the key indicator parameter is not a coal quality indicator parameter, calculate the traceability credibility score of the key indicator parameter and use the traceability credibility score as its credibility comprehensive sub-score.

[0097] 1203. If the key indicator parameter is a coal quality indicator parameter, calculate the traceability credibility score and the indicator credibility score of the key indicator parameter, and add the two together as its credibility comprehensive sub-score.

[0098] It should be noted that for coal quality parameters, such as elemental carbon content, calorific value, moisture, volatile matter, and ash content, the traceability reliability score must also be considered. and indicator credibility score The two scores are added together to obtain a comprehensive reliability sub-score. By deriving a quantifiable reliability score for the coal sample, it is possible to assess whether there are carbon data quality issues, such as unreasonable coal quality test results.

[0099] Furthermore, the specific calculation method for the traceability credibility score of key indicator parameters may include:

[0100] The target data collection method for key indicator parameters is identified, which can be either automated machine collection or manual collection. Then, specific evaluation items corresponding to the target collection method are determined and scored item by item. These scores are then summed to obtain the traceability and reliability score for the key indicator parameters. Particular emphasis is placed on the difference in data collection methods; the initial score assigned to parameters collected automatically by machine is higher than the initial score assigned to parameters collected manually.

[0101] Specifically, source tracing credibility score The calculation formula is shown in equation (1) below:

[0102] (1)

[0103] in, , The scores represent the scores for different specified evaluation items.

[0104] If the target data acquisition method is automatic machine acquisition, the corresponding specified evaluation items include, but are not limited to:

[0105] The data is automatically collected by the machine and assigned an initial score of 25 points.

[0106] : Is the data uploaded and stored in real time? If so, do not process it; otherwise, deduct 5 points.

[0107] : Is the data transmitted in an encrypted manner? If yes, do not process it; otherwise, deduct 3 points.

[0108] : Check whether the indoor temperature, humidity and other environmental conditions are normal when the corresponding testing equipment is running. If so, no action is taken; otherwise, deduct 2 points.

[0109] : Check if the corresponding testing equipment is within the validity period of verification / calibration. If so, no action will be taken; otherwise, deduct 5 points.

[0110] If the target data collection method is manual data collection, the corresponding specified evaluation items include, but are not limited to:

[0111] The data was collected manually and assigned an initial score of 20.

[0112] If the data comes from a third-party testing agency and a third-party testing report has been uploaded and stored, no points will be deducted; otherwise, 10 points will be deducted.

[0113] If the data entry has been reviewed and verified at different levels and has personnel signature verification, then no action will be taken; otherwise, 5 points will be deducted.

[0114] If there is video or image evidence to support the data collection, testing, and manual entry process, no action will be taken; otherwise, 5 points will be deducted.

[0115] : Check if the corresponding testing equipment is within the validity period of verification / calibration. If so, no action will be taken; otherwise, deduct 5 points.

[0116] It should be understood that the specified evaluation items in this embodiment are merely exemplary and not exhaustive. Those skilled in the art can expand or adjust the specified evaluation items based on the teachings herein and in combination with specific application scenarios.

[0117] Furthermore, if the key indicator parameter is a coal quality indicator parameter, its reliability score also needs to be calculated. The credibility score of this indicator The calculation methods can specifically include:

[0118] Normal distribution analysis is performed on historical coal quality indicators of coal samples from the same commercial coal type and origin. The mean and standard deviation of each historical coal quality indicator are calculated, and a reasonable fluctuation range for each historical coal quality indicator is set based on the mean and standard deviation. Key indicator parameters are compared with the corresponding reasonable fluctuation range. If the key indicator parameter does not exceed the reasonable fluctuation range, a preset confidence score is determined as its indicator confidence score. If the key indicator parameter exceeds the reasonable fluctuation range, the distance metric between the key indicator parameter and the corresponding reasonable fluctuation range is calculated, and the distance confidence score is calculated based on the distance metric as its indicator confidence score. The smaller the distance metric, the higher the distance confidence score, which is equivalent to a smaller deviation from the reasonable fluctuation range and a higher confidence level. However, it should be noted that the distance confidence score is less than the preset confidence score, which can be set to 90%, 93%, or 95% of the full score.

[0119] 130. Calculate the basic comprehensive score by weighted summation of the credibility sub-scores of all key indicator parameters.

[0120] Among them, the basic comprehensive score The calculation of 0 is shown in the following formula (2):

[0121] (2)

[0122] in, These are the weighting coefficients for different types of key indicator parameters. Generally, the instantaneous flow rate of coal entering the furnace, the carbon content of the coal, and the calorific value of the coal are considered to have the dominant weights, which can be set higher, while the weighting coefficients for other data can be set lower. The weighting coefficients of all key indicator parameters are summed to 1. n represents the total number of key indicator parameters. The comprehensive sub-score represents the credibility of key indicator parameters, and all key indicator parameters have a traceable credibility score. It can be directly used as a sub-score of its credibility. For key indicators that are also coal quality parameters, it is necessary to... Add a credibility score to the indicator. The credibility sub-score was obtained. .

[0123] 140. Perform correlation analysis between all key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters.

[0124] By employing methods such as correlation analysis and mutual information calculation, the intrinsic driving relationships among core operating parameters are quantitatively revealed. Furthermore, based on machine learning and deep learning algorithms, such as random forests, XGBoost, and neural networks, a data expert analysis model is constructed. This model is used to assess the rationality of power plant operating parameters, automatically identifying whether operating parameters match fuel characteristics. For example, it checks whether the instantaneous coal quantity deviates from a reasonable range under given coal quality and power generation load. This allows for a quantitative evaluation of the rationality of key indicator parameter combinations and predicts and outputs a rationality score. .

[0125] Specifically, step 140 may include the following steps 1401-1402 (not shown):

[0126] 1401. Based on all key indicator parameters, construct the multimodal time series vector;

[0127] 1402. Input the multimodal time series vector into the pre-trained scoring model, and obtain the rationality score of the combination of key indicator parameters based on the output of the scoring model.

[0128] The scoring model is trained based on a network model combining a deep convolutional autoencoder (DCE) and a generative adversarial network (GAN). First, a network model combining a DCE and a GAN is constructed. This model includes a coupled DCE network and a GAN. The DCE network consists of an encoder and a decoder connected in sequence, while the GAN consists of a generator and a discriminator connected in sequence. The encoder uses a multi-layer convolutional layer + pooling layer structure to extract the spatiotemporal features of the original input vector layer by layer, and then compresses the spatiotemporal features into the latent space through fully connected layers. The decoder uses transposed convolutional layers to upsample the features and reconstruct the original input vector to obtain the reconstructed vector. The generator receives random noise vectors and generates pseudo-samples similar to the distribution of real samples. The discriminator uses a convolutional neural network structure to distinguish whether the input sample is a real sample or a pseudo-sample generated by the generator.

[0129] Then, the constructed network model is trained to obtain the scoring model. During training, the entire network model can be trained end-to-end without manual annotation. By optimizing the reconstruction loss and adversarial loss simultaneously through backpropagation, it can automatically learn the joint distribution characteristics of multi-source data and quantitatively reveal the intrinsic driving relationship between key indicator parameters. Using adversarial training of the generative adversarial network as a regularization term can prevent overfitting and improve the model's discriminative ability.

[0130] Based on this, step 1402 may specifically include the following steps S21 to S25 (not shown):

[0131] S21. Input the multimodal temporal vector into the pre-trained scoring model, extract the target spatiotemporal features by performing convolution operation on the multimodal temporal vector through the encoder, and compress the target spatiotemporal features into the latent space to obtain the key feature representation.

[0132] S22. The key feature representation is upsampled by the decoder to reconstruct the target reconstruction vector.

[0133] S23. The reconstruction error is calculated based on the difference between the target reconstruction vector and the multimodal time series vector.

[0134] For the input multimodal time-series vector x, the target reconstruction vector The reconstruction error MSE(x) is calculated as shown in equation (3):

[0135] MSE(x) = (1 / n) Σ(x i - i )²(3)

[0136] Where n is the feature dimension of the multimodal time-series vector, x i and i These are the original features and the reconstructed features, respectively.

[0137] S24. Output the confidence level that the target reconstruction vector belongs to the true distribution through the discriminator.

[0138] S25. Based on the reconstruction error and confidence level, output the rationality score of the combination of key indicator parameters.

[0139] To eliminate the influence of different feature dimensions, the reconstruction error MSE(x) is standardized as shown in equation (4):

[0140] E recon (x) = MSE(x) / σ² (4)

[0141] Where σ² is the variance of the reconstruction error of the training set, E recon (x) represents the standardized result.

[0142] Furthermore, the confidence level is mapped to the interval [0, 1] as shown in equation (5) below:

[0143] C norm (x) = (C(x) - min C ) / (max C - min C (5)

[0144] Where, min Cand max C Let C(x) be the minimum and maximum confidence values ​​of the training set, and let C(x) be the confidence value. norm (x) represents the mapping result.

[0145] Finally, a weighted summation method is used to fuse the reconstruction error standardization results and the confidence mapping results, as shown in the following equation (6):

[0146] Score(x) = α (1 - E recon (x)) + β C norm (x)(6)

[0147] Here, α and β are weighting coefficients that satisfy α + β = 1, typically α = 0.6 and β = 0.4.

[0148] This implementation method is applicable to scenarios such as industrial equipment operation status monitoring and system anomaly detection, and can effectively evaluate the rationality of operating parameters.

[0149] 150. Based on the basic comprehensive score and the rationality score, the credibility comprehensive score of the real-time carbon emissions is calculated.

[0150] Optionally, the basic comprehensive score can be... 0 and reasonableness score The overall credibility score is obtained by directly adding them together. 综合 Alternatively, the basic comprehensive score can be used. 0 and reasonableness score The overall credibility score is obtained by weighted summation. 综合 .

[0151] 160. If the overall credibility score reaches the preset level, a corresponding carbon emission data credibility report will be generated.

[0152] In this embodiment of the invention, a target level can be determined based on a comprehensive credibility score. At least three levels can be set: green, yellow, and red, and the target level can be any one of them. The green level corresponds to... 综合 In cases where the percentage is greater than 85%, the yellow light level corresponds to a percentage of 75% or less. 综合 In cases where the percentage is ≤85%, the red light level corresponds to: 综合 <75% of the cases.

[0153] The preset level corresponds to 综合A green light rating of over 85% indicates that the carbon emission data is highly reliable and requires no warning. This allows for the generation of a reliable carbon emission data report, which can serve as supporting evidence for carbon emission verification and carbon quota trading.

[0154] 170. If the overall credibility score does not reach the preset level, an early warning will be issued according to the corresponding level.

[0155] Under the yellow light warning level, a level two data risk warning is triggered, that is, a level two warning report is generated and sent to all front-line production departments, such as the fuel testing department and the operations management department, as a warning.

[0156] Under the red light warning level, a Level 1 data risk warning is triggered, which means that a Level 1 warning report is generated and the associated data chain is automatically traced and locked. The Level 1 warning report is directly communicated to the company's senior management, such as the carbon asset management leadership team and the vice president of production, to immediately activate company-wide carbon emission risk emergency control measures, identify the root cause, and take relevant actions within 24 hours.

[0157] This invention constructs a "green-yellow-red" three-color early warning mechanism, which automatically triggers different response processes based on a comprehensive credibility score; by automatically generating a suspicious list, limiting response time limits, and directly reaching the corresponding level of responsibility, it achieves accurate positioning, rapid response, and closed-loop management of data risks.

[0158] As an optional implementation method, a list of warning content can also be set up, as shown in Table 1 below. Through the list of warning content, various warning content can be monitored dynamically in real time.

[0159] Table 1 List of Warning Contents

[0160]

[0161] Optionally, real-time monitoring and early warning can be implemented for all items in the early warning list. Key indicators to monitor include the calibration status of coal flow metering equipment, elemental carbon content of coal, lower heating value of coal, and instantaneous flow rate of coal entering the furnace. If any item deviates from the indicators or fails to meet requirements, an early warning notification will be immediately sent to the relevant production management department, and a report on non-compliance with carbon emission management will be generated, specifying the deviation from the accounting guidelines. Responsible personnel are required to complete verification and on-site correction within a specified timeframe, such as 4 hours, and report the results back to the system, forming a closed-loop management system.

[0162] If at least one item in the alert list is detected, a yellow alert level is triggered. If at least three items in the alert list are detected, or if a particular alert is detected for two consecutive days, a red alert level is triggered.

[0163] As an optional implementation method, a monthly carbon emission data credibility report can be output, which displays the evaluation results of each coal-fired management link in detail throughout the entire process, serving as supporting material for carbon emission verification and carbon quota trading; at the same time, the monthly carbon emission data credibility report and the real-time carbon emission calculation results are uploaded to the data management terminal for secure storage.

[0164] As shown in Table 2 below, the monthly carbon emission data credibility report details the comprehensive credibility sub-scores of each key indicator parameter and the warning status for the month. Table 2 is a model case, which selects key indicator parameters such as elemental carbon content of coal, lower heating value of coal, and instantaneous flow rate of coal entering the furnace as examples.

[0165] Table 2 Reliable Report on Monthly Carbon Emission Data

[0166]

[0167] Furthermore, based on factors such as the credibility sub-score and early warning status, suggested penalties can be assigned according to different situations. For example, if the number of early warning days for the belt scale calibration date is 0, and the number of early warning days for the belt scale calibration accuracy is greater than 3 days, then the suggested penalty for coal flow metering = monthly coal consumption × (1 + calibration accuracy during the early warning period - specified accuracy). If the number of early warning days for the belt scale calibration date exceeds 3 days, then the suggested penalty for coal flow metering = monthly coal consumption × (1 + specified accuracy).

[0168] If the monthly average score of elemental carbon content in coal is less than 20 and the number of abnormal warning days is greater than 3 days, the recommended penalty for the emission factor is calculated as emission factor × (100 + number of warning days) / 100; if the monthly average score of elemental carbon content in coal is less than 20 and the number of abnormal warning days is greater than 5 days, the emission factor will directly adopt the default value specified in the guidelines.

[0169] Furthermore, considering that in practice, some power plants still struggle to achieve full automation and information coverage in coal sample collection, testing, and subsequent processes, this invention focuses on the process of sending coal samples to a third-party testing institution after sampling. For steps in the testing process that still require manual intervention, such as sample packaging, handover, and transportation, the invention assesses whether the operations comply with regulations and meet reliability criteria. Building upon the power plant's existing full-process information-based monitoring, and combining this with personnel operation records, the invention conducts a standardized inspection of the coal sample testing process to determine whether each step follows the corresponding operating procedures, thereby reflecting the reliability of the testing data issued by a third party in the coal consumption ledger.

[0170] Specifically, the following steps can be performed during the coal sample submission process:

[0171] During the coal sample delivery and testing process, images of operating tables, monitoring videos, and environmental information are acquired. The table images are obtained by photographing paper operation record forms, and the environmental information includes at least temperature and humidity. Text is extracted from the table images through text recognition. Video information is obtained through video recognition. Based on the text, video, and environmental information, each specification item is checked. If more than two specification items fail the check, a carbon data risk warning is triggered. The table images can be taken and uploaded by the operator using a handheld mobile device.

[0172] The standardization items include, but are not limited to: whether the paper operation record forms and electronic record forms for coal samples are consistent; whether there are any missing key information, including but not limited to the time, location, personnel, and superior review signatures included in the sampling, preparation, and storage operation records of each batch of coal samples; whether the personnel's attire during the operation is compliant; whether the operators have been trained and meet the job requirements; whether the equipment maintenance and operating environment of the sampling machine and sample preparation machine meet the requirements; and whether the stored coal samples have been prepared into samples of the corresponding particle size according to the corresponding needs.

[0173] As can be seen, implementing this embodiment of the invention achieves reliable management of key carbon emission data throughout the entire process, from source collection and transmission to evidence storage. This method supports real-time collection and dynamic accounting of carbon emission data, significantly improving the efficiency and frequency of carbon emission accounting for power plants. Through system-level integration of image recognition technology and artificial intelligence algorithms, an integrated intelligent carbon data monitoring model of "front-end perception - back-end monitoring" is created, realizing automated real-time monitoring, intelligent cross-validation, and anomaly identification of carbon emission ledger data, as well as proactive identification and early warning of accounting compliance risks. It can assist enterprises in improving the standardization and scientific nature of energy metering, data recording, and system regulations in fuel management and use, and improve the enterprise's coal management standard system; it helps enterprises improve the quality and reliability of carbon data, comprehensively consider carbon emission data, and promptly identify and warn of deficiencies in the standardization of carbon emission data.

[0174] like Figure 2 As shown in the figure, this embodiment of the invention also discloses a real-time coal combustion carbon emission data accounting device, including a factor determination unit 201, a coal flow measurement unit 202, an accounting unit 203, a first scoring unit 204, a summation unit 205, a correlation analysis unit 206, a second scoring unit 207, a report generation unit 208, and an early warning unit 209, wherein,

[0175] Factor determination unit 201 is used to determine emission factors;

[0176] The coal flow metering unit 202 is used to obtain the instantaneous flow rate of coal entering the furnace during the combustion of coal-fired fossil fuels;

[0177] The calculation unit 203 is used to calculate the real-time carbon emissions by multiplying the instantaneous flow rate of coal entering the furnace by the emission factor.

[0178] The first scoring unit 204 is used to calculate the comprehensive sub-score of the credibility of various key indicator parameters during the combustion of coal and fossil fuels;

[0179] Summation unit 205 is used to perform a weighted summation of the credibility sub-scores of all key indicator parameters to obtain the basic comprehensive score;

[0180] The correlation analysis unit 206 is used to perform correlation analysis between parameters of all key indicators to obtain a reasonableness score of the combination of key indicator parameters.

[0181] The second scoring unit 207 is used to calculate the credibility comprehensive score of real-time carbon emissions based on the basic comprehensive score and the rationality score.

[0182] The report generation unit 208 is used to generate a corresponding carbon emission data credibility report when the comprehensive credibility score reaches a preset level.

[0183] The early warning unit 209 is used to issue an early warning according to the corresponding level when the overall credibility score does not reach the preset level.

[0184] As an optional implementation, the factor determination unit 201 includes:

[0185] The judgment subunit is used to determine in real time whether the testing process for coal-fired fossil fuel samples has been completed.

[0186] The simulation subunit is used to predict the simulated carbon content before the testing process is completed;

[0187] The first defined sub-unit is used to take the product of simulated carbon content and carbon oxidation rate as the emission factor;

[0188] The second determining subunit is used to obtain the measured carbon content of coal after the testing process is completed, and to use the product of the carbon content of coal and the carbon oxidation rate as the emission factor.

[0189] Further optional, the simulation sub-unit includes:

[0190] The judgment module is used to determine whether a coal-fired power plant has online coal quality analysis equipment;

[0191] The first determining module is used to determine the simulated carbon content based on the online detection results of the online coal quality analysis equipment when the judgment module determines that an online coal quality analysis equipment is available.

[0192] The second determining module is used to determine the simulated carbon content based on the historical test results of a specified batch of incoming coal samples when the judgment module determines that there is no online coal quality analysis equipment; or, based on the historical test results of the previous period; or, to select a value that meets the conditions from the local database as the simulated carbon content.

[0193] As an optional implementation, the first scoring unit 204 includes:

[0194] The judgment subunit is used to determine whether the key index parameters during the combustion of coal fossil fuels are coal quality index parameters.

[0195] The first calculation subunit is used to calculate the traceability credibility score of the key indicator parameter when the judgment subunit determines that the key indicator parameter is not a coal quality indicator parameter, and uses the traceability credibility score as its credibility comprehensive sub-score.

[0196] The second calculation subunit is used to calculate the traceability credibility score and the index credibility score of the key index parameter when the judgment subunit determines that the key index parameter is a coal quality index parameter, and the sum of the two is used as its credibility comprehensive sub-score.

[0197] As an optional implementation, the correlation analysis unit 206 includes:

[0198] Construct sub-units to generate multimodal time-series vectors based on all key indicator parameters;

[0199] The scoring subunit is used to input multimodal time-series vectors into a pre-trained scoring model and obtain a reasonableness score for the combination of key indicator parameters based on the output of the scoring model.

[0200] Further optionally, the scoring model includes a coupled deep convolutional autoencoder network and an adversarial generative network. The deep convolutional autoencoder network includes an encoder and a decoder connected in sequence, and the adversarial generative network includes a generator and a discriminator connected in sequence. Therefore, the aforementioned scoring subunit specifically includes:

[0201] The input module is used to input multimodal temporal vectors into a pre-trained scoring model. The encoder performs convolution operations on the multimodal temporal vectors to extract target spatiotemporal features, and compresses the target spatiotemporal features into the latent space to obtain key feature representations.

[0202] The reconstruction module is used to upsample the key feature representations through the decoder to reconstruct the target reconstruction vector;

[0203] The error calculation module is used to calculate the reconstruction error based on the difference between the target reconstruction vector and the multimodal temporal vector.

[0204] The discriminant module is used to output the confidence level that the reconstructed target vector belongs to the true distribution through the discriminator;

[0205] The output module is used to output a reasonableness score for the combination of key indicator parameters based on the reconstruction error and confidence level.

[0206] like Figure 3 As shown, this embodiment of the invention also discloses an electronic device, including a memory 301 storing executable program code and a processor 302 coupled to the memory 301;

[0207] The processor 302 calls the executable program code stored in the memory 301 to execute the real-time calculation method for coal-fired carbon emission data described in the above embodiments.

[0208] like Figure 4 As shown in the illustration, this invention also discloses a computer device. This computer device includes a processor, memory, network interface, display screen, and input device connected via a system bus. The processor in this computer design provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores relevant data for a real-time calculation method for coal-fired carbon emissions. The network interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the real-time calculation method for coal-fired carbon emissions described in the above embodiments.

[0209] This invention also discloses a computer-readable storage medium storing a computer program that causes a computer to execute the real-time coal combustion carbon emission data calculation method described in the above embodiments. The storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc.

[0210] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0211] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0212] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A method for real-time calculation of carbon emissions from coal combustion, characterized in that, include: The emission factor is determined, the instantaneous flow rate of coal fed into the furnace during the combustion of coal-fired fossil fuels is obtained, and the real-time carbon emissions are calculated by multiplying the instantaneous flow rate of coal fed into the furnace by the emission factor. Calculate the comprehensive sub-score of the reliability of various key indicator parameters during the combustion of coal and fossil fuels; The basic comprehensive score is obtained by weighted summation of the credibility sub-scores of all the key indicator parameters. A correlation analysis was performed on all the key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters. Based on the basic comprehensive score and the reasonableness score, the credibility comprehensive score of the real-time carbon emissions is calculated. If the overall credibility score reaches a preset level, a corresponding carbon emission data credibility report will be generated. If the overall credibility score does not reach the preset level, an early warning will be issued according to the corresponding level. Among them, a correlation analysis is performed on all the key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters, including: Based on all the key indicator parameters, a multimodal time series vector is constructed; The multimodal time-series vector is input into a pre-trained scoring model, and the rationality score of the combination of key indicator parameters is obtained based on the output of the scoring model. The comprehensive sub-score for the reliability of various key indicator parameters during the combustion of coal-fired fossil fuels includes: Determine whether the key parameters during the combustion of coal-fired fossil fuels are coal quality parameters; If the key indicator parameter is not the coal quality indicator parameter, calculate the traceability credibility score of the key indicator parameter and use the traceability credibility score as its credibility comprehensive sub-score. If the key indicator parameter is the coal quality indicator parameter, calculate the traceability credibility score and the indicator credibility score of the key indicator parameter, and use the sum of the two as its credibility comprehensive sub-score. The calculation method for the traceability credibility score of the key indicator parameters includes: Identify the target acquisition method for the key indicator parameters, wherein the target acquisition method is either automatic machine acquisition or manual acquisition. The designated evaluation items corresponding to the target collection method are identified and scored item by item. The scores of each item are accumulated and summed to obtain the traceability credibility score of the key indicator parameters.

2. The real-time calculation method for coal-fired carbon emissions according to claim 1, characterized in that, Determining emission factors includes: Real-time assessment of whether the testing process for coal and fossil fuel samples is complete; If the testing process is not completed, the simulated carbon content is predicted, and the product of the simulated carbon content and the carbon oxidation rate is used as the emission factor. If the testing process has been completed, obtain the measured elemental carbon content of the coal, and use the product of the elemental carbon content and the carbon oxidation rate as the emission factor.

3. The real-time calculation method for coal-fired carbon emissions according to claim 2, characterized in that, The predicted simulated carbon content includes: Determine whether a coal-fired power plant has online coal quality analysis equipment; If the online coal quality analysis equipment is available, the simulated carbon content is determined based on the online detection results of the online coal quality analysis equipment. If the online coal quality analysis equipment is not available, the simulated carbon content is determined based on the historical test results of a specified batch of incoming coal samples; or, the simulated carbon content is determined based on the historical test results of the previous period; or, a value that meets the conditions is selected from the local database as the simulated carbon content.

4. The method for real-time calculation of coal-fired carbon emission data according to claim 1, characterized in that, The key indicator parameters include at least fuel characteristic parameters, fuel flow parameters, energy flow parameters, and equipment operating parameters. The fuel characteristic parameters include at least the elemental carbon content of coal, the lower heating value of coal, the calorific value of coal, moisture, volatile matter, ash content, and hydrogen content. The fuel flow parameters include at least the instantaneous flow rate of coal entering the furnace. The energy flow parameters include at least the boiler load rate, the turbine power generation, and the grid-connected power generation. The coal quality indicator parameters include at least the elemental carbon content of coal, the calorific value of coal, the moisture content, the volatile matter, and the ash content.

5. The method for real-time calculation of coal-fired carbon emission data according to any one of claims 1 to 4, characterized in that, The scoring model includes a coupled deep convolutional autoencoder network and an adversarial generative network. The deep convolutional autoencoder network includes an encoder and a decoder connected in sequence, and the adversarial generative network includes a generator and a discriminator connected in sequence. The model inputs the multimodal temporal vector into a pre-trained scoring model and obtains a reasonableness score for the combination of key indicator parameters based on the output of the scoring model, including: The multimodal temporal vector is input into a pre-trained scoring model. The encoder performs a convolution operation on the multimodal temporal vector to extract the target spatiotemporal features. The target spatiotemporal features are then compressed into the latent space to obtain key feature representations. The key feature representation is upsampled by the decoder to reconstruct the target reconstruction vector; The reconstruction error is calculated based on the difference between the target reconstruction vector and the multimodal time-series vector; The discriminator outputs the confidence level that the target reconstructed vector belongs to the true distribution; Based on the reconstruction error and the confidence level, a reasonableness score for the combination of key indicator parameters is output.

6. A real-time calculation device for coal-fired carbon emissions, characterized in that, include: Factor determination unit, used to determine emission factors; The coal flow metering unit is used to obtain the instantaneous flow rate of coal entering the furnace during the combustion of coal and fossil fuels. The calculation unit is used to calculate the real-time carbon emissions by multiplying the instantaneous flow rate of the coal entering the furnace by the emission factor; The first scoring unit is used to calculate the comprehensive sub-score of the credibility of various key indicator parameters during the combustion of coal and fossil fuels; The summation unit is used to perform a weighted summation of the credibility sub-scores of all the key indicator parameters to obtain the basic comprehensive score; The correlation analysis unit is used to perform correlation analysis between all the key indicator parameters to obtain a reasonableness score for the combination of key indicator parameters. The second scoring unit is used to calculate the credibility score of the real-time carbon emissions based on the basic comprehensive score and the rationality score. The report generation unit is used to generate a corresponding carbon emission data credibility report when the credibility comprehensive score reaches a preset level. The early warning unit is used to issue an early warning according to the corresponding level when the overall credibility score does not reach the preset level. The correlation analysis unit includes: Construct sub-units to generate multimodal time-series vectors based on all the aforementioned key indicator parameters; The scoring subunit is used to input the multimodal time-series vector into a pre-trained scoring model and obtain a reasonableness score for the combination of key indicator parameters based on the output of the scoring model. The first scoring unit includes: The judgment subunit is used to determine whether the key index parameters during the combustion of coal fossil fuels are coal quality index parameters. The first calculation subunit is used to calculate the traceability credibility score of the key indicator parameter when the judgment subunit determines that the key indicator parameter is not a coal quality indicator parameter, and to use the traceability credibility score as its credibility comprehensive sub-score. The second calculation subunit is used to calculate the traceability credibility score and the index credibility score of the key index parameter when the judgment subunit determines that the key index parameter is a coal quality index parameter, and to use the sum of the two as its credibility comprehensive sub-score. The calculation method for the traceability credibility score of the key indicator parameter includes: identifying the target collection method of the key indicator parameter, wherein the target collection method is an automatic machine collection method or a manual collection method; determining the designated evaluation item corresponding to the target collection method and assigning a score to each item; and summing the scores of each item to obtain the traceability credibility score of the key indicator parameter.

7. The real-time coal-fired carbon emission data calculation device according to claim 6, characterized in that, The factor determination unit includes: The judgment subunit is used to determine in real time whether the testing process for coal-fired fossil fuel samples has been completed. The simulation subunit is used to predict the simulated carbon content before the testing process is completed; The first determining subunit is used to use the product of the simulated carbon content and the carbon oxidation rate as the emission factor; The second determining subunit is used to obtain the measured carbon content of coal after the testing process is completed, and to use the product of the carbon content of coal and the carbon oxidation rate as the emission factor.

8. The real-time coal-fired carbon emission data calculation device according to claim 6, characterized in that, The scoring model includes a coupled deep convolutional autoencoder network and an adversarial generative network. The deep convolutional autoencoder network includes an encoder and a decoder connected in sequence, and the adversarial generative network includes a generator and a discriminator connected in sequence. The scoring subunit includes: The input module is used to input the multimodal temporal vector into a pre-trained scoring model, extract target spatiotemporal features by performing convolution operations on the multimodal temporal vector through the encoder, and compress the target spatiotemporal features into the latent space to obtain key feature representations; The reconstruction module is used to upsample the key feature representation through the decoder to reconstruct the target reconstruction vector; An error calculation module is used to calculate the reconstruction error based on the difference between the target reconstruction vector and the multimodal temporal vector. The discriminant module is used to output the confidence level that the target reconstructed vector belongs to the true distribution through the discriminator; The output module is used to output a reasonableness score for the combination of key indicator parameters based on the reconstruction error and the confidence level.

9. An electronic device, characterized in that, It includes a memory storing executable program code and a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the real-time calculation method for coal-fired carbon emission data according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program causes a computer to execute the real-time calculation method for coal-fired carbon emission data according to any one of claims 1 to 5.

Images