A coal power enterprise pollution reduction and carbon reduction collaborative control effect accounting and evaluation system and method

Abstract

The present application provides a kind of coal power enterprise pollution reduction carbon reduction collaborative control effect accounting evaluation system and method, it is related to coal power technology field.The emission AI accounting evaluation model constructed by training is carried out intelligent identification to each pollution source type of coal power emission data with time stamp and recommends the output corresponding type of emission calculation strategy, carries out corresponding emission accounting based on the recommended emission calculation strategy, outputs corresponding emission accounting result, the emission accounting result of each pollution source type is counted, and the coal power emission accounting amount of enterprise is quickly obtained.The present application is based on the time stamp of the coal power emission data marked by public credit agency monitoring, and carries out data intelligent accounting calculation based on the coal power emission data marked by time stamp, not only can improve data public credit, but also can utilize AI model to carry out data classification calculation, to realize the intelligent accounting of coal power carbon and other pollutants, improve the accounting efficiency, shorten the accounting period, reduce the difficulty of checking of coal power enterprise pollution reduction carbon reduction.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of coal power technology, in particular to a coal power enterprise pollution reduction and carbon reduction collaborative control effect accounting and evaluation system and method. BACKGROUND

[0002] The coal power industry has long been plagued by high energy consumption and high emissions. How to achieve low-carbon transformation while considering the collaborative control of pollution reduction and carbon emission reduction is a key challenge that needs to be addressed in current coal power emission accounting.

[0003] Coal power emission accounting inventory plays a core role in the low-carbon transformation and collaborative control of pollution reduction and carbon reduction of coal power enterprises. By building an accurate emission inventory, the emission data of various pollutants and greenhouse gases of coal power enterprises can be systematically collected, and the emission reduction potential of the emission source can be quantified. Accurate accounting inventory can provide a clear technical path for the low-carbon transformation of coal power enterprises, and clearly identify which links have the greatest emission reduction potential and how to achieve the reduction of multiple pollutants through collaborative control.

[0004] Table 1 is the accounting range of various emission sources. The accounting substances of atmospheric pollutants generally include sulfur dioxide (SO2), nitrogen oxides (NOx), carbon monoxide (CO), volatile organic compounds (VOCs), ammonia (NH), total suspended particulate matter (TSP), inhalable particulate matter (PMo), fine particulate matter (PM), black carbon (BC), and organic carbon (OC). The accounting range of power and heat sources includes the atmospheric pollutant emissions from the production facilities of the thermal power generation (D4411), cogeneration (D4412), and biomass power generation (D4417) and heat production and supply (D4430) industries, as well as the CO2 and N2O emissions from the fuel combustion process. Facilities include coal-fired power plant coal-fired units and related facilities, coal transportation systems, waste gas treatment facilities, fuel storage systems, and other supporting facilities. Emissions include SO2, NOx, PM10, PM2.5, CO2, CH4, N2O, and other indirect carbon emissions from purchased power or heat. In terms of time, the annual time range is first established to constitute a complete accounting period, and the daily time range is changed according to data requirements.

[0005] Table 1: Accounting range of various emission sources

[0006]

[0007] In addition, for emission accounting methods, the traditional emission accounting method mainly calculates the atmospheric pollutant and greenhouse gas emissions according to the differences in pollution source types and emission characteristics. The calculation methods can be divided into monitoring method, material balance method, and emission coefficient method. The appropriate emission calculation method should be determined by referring to the calculation methods and priority suggested by the guidelines and combining with the actual situation. The current main determination is the following three methods:

[0008] Emission factor method: The emission amount is calculated by multiplying the data of a certain type of activity (such as fuel consumption) by the corresponding emission factor.

[0009] Material balance method: This method is suitable for situations where energy or material flow is clearly tracked. By tracking the input and output of materials within and outside the system, the emission of pollutants and greenhouse gases can be calculated. First, determine the input and output of materials (such as the input amount of coal and the emission amount of waste gas), and then calculate the emission amount by calculating the input, consumption and waste emission of the material.

[0010] Online monitoring method: For large emission sources (such as power plant boilers), the emission data is obtained directly through online monitoring equipment. Install real-time monitoring equipment to ensure that the equipment can continuously measure the main pollutants and greenhouse gases in the flue gas. At the same time, the monitoring equipment is calibrated regularly to ensure the accuracy of the data.

[0011] However, the above existing calculation methods may have some problems in data credibility. For example, the calculation methods for SO2, NOx, TSP and VOCs are usually calculated according to the automatic monitoring method or the production and emission coefficient method, or derived from the emission permit implementation report; the automatic monitoring method is preferred to use the automatic monitoring data that meets the specification requirements for calculation. If the equipment cannot meet the requirements, the production and emission coefficient method is used. The production and emission coefficient method mainly calculates the emission amount of pollutants through the production and emission coefficient method provided in the emission permit implementation report of the enterprise. However, this calculation process lacks supervision of the data.

[0012] In addition, for the emission accounting method of coal-fired power plant emission reduction projects, the traditional emission accounting method is not intelligent enough, and the administrator of the enterprise needs to manually process, collect and calculate a large amount of data. The administrator needs to select the corresponding emission calculation method according to the pollution source type and emission characteristics for calculation. Therefore, this process is relatively long, time-consuming, inefficient, and the accounting method is not intelligent enough. SUMMARY

[0013] In order to solve the technical problems existing in the prior art, the present application provides the following technical solutions:

[0014] On the one hand, a coal-fired power plant pollution reduction and carbon reduction collaborative control effect accounting and evaluation method is provided, which is realized by an electronic device, and the method comprises:

[0015] S1, real-time monitoring and collection of coal-fired power plant emission data set with time stamp, including at least the following data: fuel consumption data, equipment operation data and coal quality information;

[0016] S2, classify the coal power emission data set based on the LLM large language model to obtain coal power emission data subsets of different pollution source categories;

[0017] S3, import the coal power emission data subsets into a preset emission AI accounting and evaluation model, identify the emission characteristics of the coal power emission data subsets by the emission AI accounting and evaluation model, and recommend and output an emission quantity calculation strategy matched with the emission characteristics;

[0018] S4, based on the recommended emission quantity calculation strategy, perform corresponding emission quantity accounting on the coal power emission data subsets, and output corresponding emission quantity accounting results;

[0019] S5, statistics of the emission quantity accounting results of each pollution source category, and generating a corresponding emission source accounting list according to a preset report format.

[0020] Optionally, the S1, real-time monitoring and collecting coal power emission data sets with time stamps, including:

[0021] The coal power enterprise monitors and collects coal power emission data sets of the coal power enterprise according to a preset sampling frequency through a background server and reports to a government server of a public trust institution;

[0022] Through the government server, the coal power enterprise and the coal power emission data set reported by it are audited for public trust:

[0023] If the audit is passed, the coal power emission data set is time-stamped to generate a coal power emission data set with a time stamp;

[0024] If the audit is not passed, the corresponding audit opinion is issued to the background server of the coal power enterprise.

[0025] Optionally, the S2, based on the LLM large language model, classifies the coal power emission data set to obtain coal power emission data subsets of different pollution source categories, including:

[0026] Constructing classification keywords and classification logic of different pollution source categories;

[0027] Combined with the classification keywords and the classification logic, a large language model prompt word is constructed;

[0028] The large language model prompt word is input into a preset LLM large language model for prompt word training and learning of the LLM large language model;

[0029] The LLM large language model is used to read the coal power emission data set, and the coal power emission data set is classified based on the large language model prompt word to identify coal power emission data subsets of different pollution source categories;

[0030] output and save the subset of coal-fired power plant emission data of different pollution source categories.

[0031] Optionally, the method for generating the emission AI accounting evaluation model comprises:

[0032] collecting coal-fired power plant emission data of several different pollution source categories and preprocessing, wherein the coal-fired power plant emission data comprises historical fuel consumption data, equipment operation data and coal quality information of corresponding pollution sources;

[0033] extracting emission features from the coal-fired power plant emission data, wherein the emission features comprise emission source activity levels of corresponding pollution sources in the coal-fired power plant emission data within a preset time period, i.e. the difference between the emission amount of the pollution source and the emission threshold within the preset time period;

[0034] labeling the emission features, and the labeling information comprises pollution source attributes corresponding to the emission features and accounting strategies for calculating the emission amount of the corresponding pollution sources;

[0035] collecting the labeled emission features corresponding to different pollution source categories to construct a feature set;

[0036] dividing the feature set into a training set and a validation set according to a preset ratio;

[0037] inputting the training set into a preset random forest model for feature training and learning to generate the emission AI accounting evaluation model;

[0038] verifying the prediction performance of the emission AI accounting evaluation model using the validation set:

[0039] if the verification is passed, deploying the emission AI accounting evaluation model to a background server;

[0040] if the verification is not passed, repeating the above steps to re-construct the emission AI accounting evaluation model.

[0041] Optionally, the S3, importing the subset of coal-fired power plant emission data into a preset emission AI accounting evaluation model, identifying emission features of the subset of coal-fired power plant emission data by the emission AI accounting evaluation model and recommending and outputting emission amount calculation strategies matching the emission features, comprises:

[0042] outputting the subset of coal-fired power plant emission data of different pollution source categories to the emission AI accounting evaluation model in sequence by an LLM large language model;

[0043] The emission AI accounting evaluation model identifies the emission characteristics of the subset of coal-fired power plant emission data, and according to the emission characteristics and their labeled information, retrieves and outputs the corresponding emission calculation strategy;

[0044] The recommended emission calculation strategy is bound to the subset of coal-fired power plant emission data corresponding to the emission characteristics.

[0045] Optionally, S4, based on the recommended emission calculation strategy, performs corresponding emission accounting on the subset of coal-fired power plant emission data, and outputs the corresponding emission accounting result, including:

[0046] The emission calculation strategy is parsed to obtain the emission calculation formula therein;

[0047] The LLM large language model reads the emission parameters in the subset of coal-fired power plant emission data, and imports the emission calculation formula to calculate and output the corresponding emission accounting result;

[0048] The emission accounting result is bound to the corresponding pollution source and its subset of coal-fired power plant emission data.

[0049] On the other hand, a coal-fired power plant pollution reduction and carbon reduction collaborative control effectiveness accounting evaluation system is provided, which is used to implement the coal-fired power plant pollution reduction and carbon reduction collaborative control effectiveness accounting evaluation method described above. The system includes:

[0050] A coal-fired power plant data monitoring module is used to monitor and collect coal-fired power plant emission data sets with timestamps in real time, including at least the following data: activity level data, generation coefficient data, end-of-pipe treatment facility information, or automatic monitoring data;

[0051] An LLM classification module is used to classify the coal-fired power plant emission data set based on an LLM large language model to obtain subsets of coal-fired power plant emission data of different pollution source categories;

[0052] An accounting strategy recommendation module is used to import the subset of coal-fired power plant emission data into a pre-set emission AI accounting evaluation model, identify the emission characteristics of the subset of coal-fired power plant emission data by the emission AI accounting evaluation model, and recommend and output an emission calculation strategy matching the emission characteristics;

[0053] An emission calculation module is used to perform corresponding emission accounting on the subset of coal-fired power plant emission data based on the recommended emission calculation strategy, and output the corresponding emission accounting result;

[0054] A statistical module is used to statistically analyze the emission accounting results of each pollution source category, and generate a corresponding emission source accounting list according to a pre-set report format.

[0055] In another aspect, an electronic device is provided, comprising: a processor; a memory having computer readable instructions stored thereon, which, when executed by the processor, implement any one of the above coal power enterprise pollution reduction and carbon reduction collaborative control effectiveness accounting and evaluation methods.

[0056] In another aspect, a computer readable storage medium is provided, having at least one instruction stored therein, which is loaded and executed by a processor to implement any one of the above coal power enterprise pollution reduction and carbon reduction collaborative control effectiveness accounting and evaluation methods.

[0057] The technical solutions provided by the embodiments of the present application have at least the following beneficial effects:

[0058] The present application provides a coal power enterprise pollution reduction and carbon reduction collaborative control effectiveness accounting and evaluation method, which aims to use big data technology to intelligently identify and recommend output corresponding type of emission calculation strategy for each pollution source type of coal power emission data with time stamp by training and building an emission AI accounting and evaluation model, and to perform corresponding emission accounting based on the recommended emission calculation strategy and output the corresponding emission accounting result. Finally, the emission accounting results of each pollution source type are statistically counted to quickly obtain the coal power emission accounting amount of the enterprise. The present application can timestamp the coal power emission data monitored by a public trust agency, and perform data intelligent accounting calculation based on the timestamped coal power emission data. Not only can it improve data credibility, but also can use AI model for data classification calculation to realize intelligent accounting of coal power carbon and other pollutants, greatly improve the accounting efficiency, shorten the accounting period, and reduce the difficulty of checking the pollution reduction and carbon reduction of coal power enterprises. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0060] Figure 1 is a flow chart of the coal power enterprise pollution reduction and carbon reduction collaborative control effectiveness accounting and evaluation method provided by the embodiments of the present application;

[0061] Figure 2 is an interactive schematic diagram of the data processing system supervised by the public trust;

[0062] Figure 3 is a training flow chart of the emission AI accounting and evaluation model provided by the embodiments of the present application;

[0063] Figure 4 is a coal power enterprise pollution reduction and carbon reduction collaborative control effect accounting and evaluation system block diagram provided by an embodiment of the present application.

[0064] Figure 5 is a structural schematic diagram of an electronic device. DETAILED DESCRIPTION

[0065] The technical solutions in the present application will be described below with reference to the drawings.

[0066] In the embodiments of the present application, the words such as "example", "for example" are used to represent as an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0067] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0068] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1. When the distinction is not emphasized, the meanings expressed are consistent.

[0069] In order to make the technical problems, technical solutions and advantages to be solved by the present application more clear, the following will be described in detail with reference to the drawings and specific embodiments.

[0070] The embodiments of the present application provide a coal power enterprise pollution reduction and carbon reduction collaborative control effect accounting and evaluation method. The method can be realized by an electronic device, which can be a terminal or a server. As shown in a coal power enterprise pollution reduction and carbon reduction collaborative control effect accounting and evaluation method flowchart, the processing flow of the method can include the following steps: Figure 1

[0071] S1, real-time monitoring and collecting coal power emission data set with time stamp, including at least the following data: fuel consumption data, equipment operation data and coal quality information;

[0072] S2, classifying the coal power emission data set based on the LLM large language model, to obtain coal power emission data subsets of different pollution source categories;

[0073] ​S3, import the coal power emission data subset into a preset emission AI accounting evaluation model, identify the emission characteristics of the coal power emission data subset by the emission AI accounting evaluation model, and recommend and output an emission quantity calculation strategy matched with the emission characteristics;

[0074] S4, based on the recommended emission quantity calculation strategy, perform corresponding emission quantity accounting on the coal power emission data subset, and output a corresponding emission quantity accounting result;

[0075] S5, statistics of the emission quantity accounting result of each pollution source type, and generating a corresponding emission source accounting list according to a preset report format.

[0076] In the present application, the accounting range of coal power emission for different pollution source types can be shown in the accounting range table in Table 1. According to the accounting requirements, the sampling accounting can be performed. The first level emission source can be identified by collecting statistical yearbook, pollution discharge license and its implementation report data information, and environmental statistical data, etc. The types of atmospheric pollutants and greenhouse gases to be accounted for are determined according to the first level emission source.

[0077] For the coal power emission accounting strategy of different pollution source types, the present application combines AI algorithm and LLM large language model for joint calculation and processing, which replaces the corresponding accounting completed by artificial means, so as to realize the intelligent accounting process of coal power enterprise pollution reduction and carbon reduction collaborative control effect, and improve the accounting efficiency.

[0078] The steps S1-S5 will be described in detail below.

[0079] Optionally, the S1, real-time monitoring and collecting coal power emission data set with time stamp, including:

[0080] The coal power enterprise monitors and collects the coal power emission data set of the coal power enterprise according to the preset sampling frequency through the background server, and reports to the government server of the public credibility institution;

[0081] Through the government server, the coal power enterprise and the coal power emission data set reported by it are audited for public credibility:

[0082] If the audit is passed, the coal power emission data set is time-stamped, and a coal power emission data set with time stamp is generated;

[0083] If the audit is not passed, the corresponding audit opinion is issued to the background server of the coal power enterprise.

[0084] For example, Figure 2As shown, in order to solve the accuracy and effectiveness of the accounting data of coal power enterprises, the coal power enterprises communicate with the third-party public credibility institutions of government affairs, and can be monitored and collected by the background server according to the preset sampling frequency. The corresponding coal power emission data set needs to be reported to the government server, and the government server needs to be audited for the public credibility of the coal power emission data that needs to be calculated. The emission data of the coal power enterprise is given public credibility and authentication. When operating specifically, the government server can audit the coal power emission data set according to the corresponding audit conditions, and mark the data set with a timestamp, so that the coal power emission data collected each time has the corresponding public credibility authentication mark. Subsequently, when calculating, the authenticated coal power emission data set needs to be used for the next calculation authentication, so as to avoid enterprise emission data fraud and improve its public credibility and accuracy. The government server can complete the timestamp marking, and the steps of marking the timestamp after data auditing and authentication function can refer to the following steps:

[0085] 1. Time stamp application and verification operation steps

[0086] Log in to the time stamp management system:

[0087] First, log in to the official website of the time stamp service provider.

[0088] Select time stamp authentication:

[0089] In the login management interface, find the option related to time stamp authentication and click to enter.

[0090] Upload file and apply for time stamp:

[0091] Select the file that needs to apply for time stamp (usually the data file that has been audited), and click the button to apply for time stamp.

[0092] Download time stamp authentication compressed package:

[0093] After successful application, download the time stamp authentication compressed package according to the system guide. This compressed package usually contains the original file, time stamp.tsa file and time stamp authentication certificate.

[0094] Save and backup

[0095] Save the original file, time stamp.tsa file and time stamp authentication certificate properly to ensure the integrity and traceability of the data.

[0096] 2. Time stamp authentication function

[0097] Evidence preservation:

[0098] Timestamp authentication is a means of evidence preservation, which can provide reliable time proof and content authenticity and integrity proof for electronic files. Through timestamp, the generation time of electronic files can be determined, and they can be prevented from being tampered with.

[0099] Legal Effect:

[0100] In law, electronically timestamped data has the effect of proof. This means that these electronic data can be used as legal evidence when needed.

[0101] Reducing the cost of evidence:

[0102] As a means of fixing electronic data, timestamp technology is recognized in judicial practice. The timestamp authentication certificate issued by a trusted third-party electronic certification service agency can be recognized without notarization and judicial expertise, greatly reducing the cost of evidence and improving the efficiency of evidence.

[0103] Ensure data integrity:

[0104] The application of timestamp can ensure the integrity of data from generation to storage and transmission. Any tampering with the data will result in timestamp verification failure, thus ensuring the authenticity and credibility of the data.

[0105] Notes

[0106] Before applying for a timestamp, you need to ensure that the uploaded file has been audited and meets the relevant requirements.

[0107] When saving and backing up timestamped files, appropriate measures should be taken to ensure the security and integrity of the files.

[0108] When using timestamp as legal evidence, you need to follow the relevant legal procedures and regulations.

[0109] Through the above steps and function introduction, it can be seen that the application of timestamp after data audit has important significance. It not only ensures the authenticity and integrity of the data, but also provides effective evidence support in law.

[0110] Optionally, the S2 classifies the coal power emission data set based on the LLM large language model to obtain coal power emission data subsets of different pollution source classes, including:

[0111] Constructing classification keywords and classification logic for different pollution source classes;

[0112] Constructing large language model prompt words in combination with the classification keywords and the classification logic;

[0113] inputting the large language model prompt word into a preset LLM large language model, and performing prompt word training and learning on the LLM large language model;

[0114] traversing and reading the coal power emission dataset through the LLM large language model, and classifying the coal power emission dataset based on the large language model prompt word, to identify coal power emission data subsets of different pollution source classes;

[0115] outputting and saving the coal power emission data subsets of different pollution source classes.

[0116] When classifying coal power emission data of different pollution source classes, the LLM large language model deployed in the background automatically performs corresponding classification operations (customized training or requests for calling LLM large language models from third parties through LLM API interfaces). The specific operation can be referred to the following steps:

[0117] 1. Constructing a large language model prompt word:

[0118] Clarify the task goal: First, the task goal needs to be clarified, which is to classify the coal power emission dataset and identify coal power emission data subsets of different pollution source classes.

[0119] Design the structure of the prompt word: According to the task goal, design the structure of the prompt word, including the instruction part and the input data part. The instruction part should clearly state the classification task and requirements, and the input data part should provide relevant information of the coal power emission dataset.

[0120] Write specific prompt words: For example, the following prompt word can be written: "Please classify the following coal power emission dataset according to pollution source classes, and identify coal power emission data subsets of different pollution source classes. The dataset contains [specific data description], please classify according to [specific classification standard]."

[0121] 2. Input the prompt word for training and learning:

[0122] Select LLM large language model: According to the task requirements and model performance, select a suitable LLM large language model, such as GPT series, PaLM, etc.

[0123] Input prompt word: Input the constructed prompt word into the selected LLM large language model.

[0124] Training and learning: Let the model learn according to the prompt word, so that it can understand the task requirements and learn to classify the coal power emission dataset.

[0125] 3. Traverse and read the coal power emission dataset:

[0126] Load the dataset: Load the coal power emission dataset into the LLM large language model.

[0127] Traverse data: Use the model to traverse and read each piece of data in the dataset.

[0128] Classification based on prompt words:

[0129] Apply prompt words: During the process of traversing data, constantly apply the constructed prompt words to guide the model to classify each piece of data.

[0130] Identify pollution source class: According to the classification standard in the prompt words, the model should be able to identify the pollution source class to which each piece of data belongs.

[0131] Generate data subsets: Classify the identified coal-fired power plant emission data of different pollution source classes into corresponding data subsets.

[0132] Through the above steps, the coal-fired power plant emission data set can be classified by using large language model prompt words, and the coal-fired power plant emission data subsets of different pollution source classes can be identified. In this process, the generalization ability, few-shot learning ability and context learning ability of the large language model will be fully utilized, so as to realize the accurate classification and identification of the coal-fired power plant emission data.

[0133] The coal-fired power plant emission data subsets of each pollution source class can be classified and saved in the background.

[0134] The coal-fired power plant emission data set can include:

[0135] Fuel consumption data: amount of coal used, combustion efficiency, etc.

[0136] Equipment operation data: workload of generator set, use time

[0137] Coal quality information: carbon content, ash content, volatile content and other index data.

[0138] The collection index can be set according to the demand.

[0139] The following is the training process of the AI algorithm model of the present application, which can be understood in combination with the model structure of random deep forest algorithm and its training application principle.

[0140] Optionally, the generating method of the emission AI accounting and evaluation model comprises:

[0141] Collect and preprocess coal-fired power plant emission data of several different pollution source classes, wherein the coal-fired power plant emission data includes historical fuel consumption data, equipment operation data and coal quality information of corresponding pollution sources;

[0142] Feature extraction is performed on the coal power emission data to extract emission features in the coal power emission data, wherein the emission features include emission source activity levels of corresponding pollution sources in the coal power emission data within a preset time period, i.e., the difference between the emission amount of a pollution source and the emission threshold within a preset time period;

[0143] The emission features are labeled with feature labels, and the labeling information includes pollution source attributes corresponding to the emission features and an emission amount calculation strategy for accounting for the pollution sources corresponding to the emission features;

[0144] The labeled emission features corresponding to different pollution source categories are collected to construct a feature set;

[0145] The feature set is divided into a training set and a validation set according to a preset ratio;

[0146] The training set is input into a preset random forest model for feature training and learning to generate the emission AI accounting and evaluation model;

[0147] The validation set is used to verify the prediction performance of the emission AI accounting and evaluation model:

[0148] If the verification is passed, the emission AI accounting and evaluation model is deployed to a background server;

[0149] If the verification fails, the above steps are repeated to rebuild the emission AI accounting and evaluation model.

[0150] As shown in Figure 3 The coal power emission data corresponding to different pollution source categories can be extracted from the enterprise background log library and preprocessed. The LLM can collect coal power emission data of several different pollution source categories from the log library based on data collection prompts in the manner of the above-mentioned LLM based on prompt words. The preprocessing can be cleaning and the like.

[0151] Next, the random deep forest algorithm is used to train the emission AI accounting and evaluation model. Random forest is an ensemble learning algorithm that builds multiple decision trees and combines their results to improve prediction accuracy and stability. In practical applications, the parameters and structure of the random forest model can be flexibly adjusted according to the specific problem requirements and data characteristics.

[0152] The steps of building a model using the random forest algorithm generally include the following main links:

[0153] Data preprocessing:

[0154] The original data set is cleaned to handle missing values, outliers, etc.

[0155] Standardization or normalization of data ensures that the numerical ranges of different features are on the same order of magnitude, preventing certain features from having an excessive impact on the model.

[0156] Divide the dataset into training and testing sets (or training, validation, and testing sets) for model training and evaluation.

[0157] Parameter settings:

[0158] Determine the number of decision trees in the random forest (n_estimators). More trees can improve the stability and accuracy of the model, but also increase the computational cost.

[0159] Set the maximum depth of each tree (max_depth) to help prevent overfitting.

[0160] Determine the maximum number of features considered when splitting each node (max_features), which can increase the diversity between trees.

[0161] Set other related parameters such as the minimum number of samples to split a node (min_samples_split), the minimum number of leaf node samples (min_samples_leaf), etc.

[0162] Train the model:

[0163] Use the training set (divide the feature set according to 8:2) to train the random forest model. During training, each tree is independently trained on its corresponding bootstrap sample (with replacement), and a subset of features is randomly selected for consideration when splitting each node.

[0164] Model evaluation:

[0165] Evaluate the trained model using the test set (or validation set) to calculate performance indicators such as accuracy, recall, F1 score, etc.

[0166] Out-of-bag error (OOB Error) can be used to evaluate the generalization ability of the model, which is an evaluation method that does not require additional reserved test samples.

[0167] Feature importance evaluation:

[0168] After training, the feature importance output by the model can be used to evaluate the contribution of each feature to the model's prediction results.

[0169] Model tuning:

[0170] Based on the evaluation results and feature importance evaluation, the model parameters are tuned to improve the performance of the model.

[0171] Model application:

[0172] Apply the trained model to new data for tasks such as prediction or classification.

[0173] When extracting features, the supervised learning algorithm can be used to extract the corresponding emission features from the coal power emission data:

[0174] Coal power emission data includes information such as the concentration, emission amount, and timestamp of various emissions.

[0175] Label the data, i.e., divide the data into different categories or labels according to the emission characteristics (such as the type of emissions, concentration range, etc.).

[0176] Divide the dataset into training set, validation set and test set, in order to train and evaluate the model.

[0177] Data preprocessing:

[0178] Clean the data, handle missing values, outliers, etc.

[0179] Standardize or normalize the data to ensure that the numerical range of different features is on the same order of magnitude.

[0180] Feature engineering may be required, such as feature extraction (e.g., mapping original data to low-dimensional space through mathematical transformation) and feature selection (e.g., selecting the most useful feature subset for prediction tasks).

[0181] Model selection and construction:

[0182] Select appropriate supervised learning algorithms, such as logistic regression, support vector machines, decision trees, random forests, neural networks, etc.

[0183] Construct the prediction model according to the selected algorithm and set the corresponding hyperparameters (such as learning rate, iteration number, regularization strength, etc.).

[0184] Model training:

[0185] Use the training set data to train the model, continuously adjust the model parameters through optimization methods (such as gradient descent), minimize the loss function (such as cross-entropy, mean square error, etc.), and improve the prediction ability of the model.

[0186] Model validation and tuning:

[0187] Use the validation set data to validate the model and evaluate its performance (such as accuracy, recall, F1 score, etc.).

[0188] According to the verification results, the model is optimized, such as adjusting hyperparameters, replacing algorithms, etc., to improve the generalization ability of the model.

[0189] Feature extraction:

[0190] During model training, supervised learning algorithms automatically learn and extract the most useful features from raw data for prediction tasks (such as the type of emissions, concentration range, etc.).

[0191] These features can be used to explain the model's prediction results and help understand the internal rules of coal-fired power plant emissions data.

[0192] Model testing and application:

[0193] Use the test set data to conduct final testing on the model to evaluate its performance in actual application.

[0194] If the model performs well, it can be deployed to real-world applications for real-time monitoring and prediction of coal-fired power plant emissions data.

[0195] The performance of supervised learning algorithms depends largely on the quality and quantity of training data, and administrators need to do a good job of preprocessing in the preprocessing stage. Data quality and quantity can be supervised in combination with LLM. Specific judgment and completion by the administrator.

[0196] The emission characteristics described herein include the emission source activity level of the corresponding pollution source in the coal-fired power plant emissions data within a predetermined time period, i.e., the difference between the emission amount of the pollution source and the emission threshold within the predetermined time period. The emission characteristics can be the emission concentration, emission amount (taking the maximum value), or timestamp of the corresponding pollutant, etc. Preferably, the difference between the emission amount of the pollution source and the emission threshold (such as the maximum emission value of sulfur dioxide specified in the current quarter) within the current time period is used as the emission characteristic, which is used as the emission characteristic for feature recognition and matching strategy.

[0197] The pollution source attribute labeled for the emission characteristics and the emission amount calculation strategy for the pollution source corresponding to the emission characteristics can be configured according to the following strategy conditions, which facilitates subsequent model matching according to the identified emission characteristics:

[0198] The emission amount calculation strategy for feature labeling of different pollution source classes can be labeled and associated with the corresponding emission amount calculation strategy by the administrator according to the pollution source type and its emission characteristics. The emission amount calculation strategy contains the emission amount calculation formula of the corresponding pollution source, which can be configured by the administrator according to the industry / national accounting standards. For example, the following are several different pollution source accounting strategies:

[0199] 1. SO2, NOx, TSP and VOCs emission amount calculation method

[0200] SO2, NOx, TSP and VOCs emissions are calculated by automatic monitoring method or production and emission coefficient method, or from the discharge permit implementation report.

[0201] The automatic detection method is preferred to use the automatic monitoring data that meets the specification requirements for accounting. The automatic monitoring equipment should meet the relevant laws and regulations of installation, debugging, acceptance and operation and maintenance, and ensure that the quarterly effective capture rate is not less than 75%. There is no record of monitoring data falsification within three years. If the equipment cannot meet the requirements, the production and emission coefficient method is used. The production and emission coefficient method mainly calculates the pollutant emissions through the production and emission coefficient method provided in the discharge permit implementation report of the enterprise.

[0202] 2. PM10, PM2.5, BC and OC emission calculation method

[0203] The emissions of PM10 and PM2.5 can be calculated according to the emissions of TSP and the proportion of particles in a certain particle size range (such as PM2.5 and PM10) in total particles (E PM ), as shown in formula (4-1); the emissions of BC and OC can be calculated according to the emissions of PM2.5 and the proportions of BC and OC in PM2.5 (E BC and E OC ), as shown in formulas (4-2) and (4-3):

[0204] E PM = E TSP x f PM (4-1),

[0205] E BC = E PM2.5 x f BC (4-2),

[0206] E OC = E PM2.5 x f OC (4-3),

[0207] In the formula, E TSP is the emission of TSP, fpm is the proportion of particles in a certain particle size range (such as PM5 and PM10) in the emitted TSP, and f BC and f OC are the proportions of BC and OC in PM2.5, respectively.

[0208] 3. CO emission calculation method

[0209] The CO emissions of the power and heat source combustion process can be calculated according to the production and emission coefficient method, and the calculation formula of the production and emission coefficient method is as follows:

[0210] E = A x EF (4-4)

[0211] In the formula, A is the activity level of the emission source; EF is the generation coefficient of the pollutant (which can be obtained by referring to the corresponding coefficient table).

[0212] 4. CO2 emission calculation method

[0213] The CO2 emissions generated by the combustion of fossil fuels in the power and heat sources and the indirect CO2 emissions generated by the purchase of electricity and heat are calculated according to the "Guidelines for Enterprise Greenhouse Gas Emission Accounting and Reporting for Power Generation Facilities". When calculating indirect emissions, it is recommended to use the latest national grid emission factor published by the Ministry of Ecology and Environment.

[0214] For key enterprises included in the scope of greenhouse gas emission reporting and verification, CO2 emissions can be directly used for verification data.

[0215] 5. N2O emission calculation method

[0216] N2O emissions generated by fuel combustion are calculated according to the method in the "Provincial Greenhouse Gas Emission Inventory Compilation Guidelines (Trial)".

[0217] After the above training, the emission AI accounting evaluation model can be deployed to the background server for subsequent strategy recommendation.

[0218] Optionally, the S3 imports the coal-fired power plant emission data subset into a pre-set emission AI accounting evaluation model, identifies the emission characteristics of the coal-fired power plant emission data subset by the emission AI accounting evaluation model, and recommends and outputs an emission calculation strategy matching the emission characteristics.

[0219] The LLM large language model sequentially outputs the coal-fired power plant emission data subset of different pollution sources to the emission AI accounting evaluation model;

[0220] The emission AI accounting evaluation model identifies the emission characteristics of the coal-fired power plant emission data subset, and according to the emission characteristics and their labeled information, retrieves and recommends and outputs the corresponding emission calculation strategy;

[0221] The recommended emission calculation strategy is bound to the coal-fired power plant emission data subset corresponding to the emission characteristics.

[0222] After the large language model classification, the large language model can be directly communicated with the emission AI accounting evaluation model, and the coal-fired power plant emission data subsets of various types of pollution sources are input into the emission AI accounting evaluation model in turn. After the model identifies and recommends the emission calculation strategy for the current type of pollution source, the corresponding type of pollution source is recommended. The data communication between the large language model and the AI model can be carried out according to the corresponding transmission port of the model deployment.

[0223] After the large language model classification, the AI model can be directly input for strategy recommendation, and the recommended strategy for the corresponding pollution source and the corresponding subset are bound on the background server, and then the emission accounting calculation is carried out.

[0224] Optionally, S4, based on the recommended emission calculation strategy, the coal-fired power plant emission data subset is subjected to corresponding emission accounting, and the corresponding emission accounting result is output, comprising:

[0225] The emission calculation strategy is analyzed to obtain the emission calculation formula therein;

[0226] The emission parameters in the coal-fired power plant emission data subset are read by the LLM large language model, and the emission calculation formula is imported to calculate and output the corresponding emission accounting result;

[0227] The emission accounting result is bound with the corresponding pollution source and its coal-fired power plant emission data subset.

[0228] S5, the emission accounting results of each type of pollution source are counted, and the corresponding emission source accounting list is generated according to the preset report format.

[0229] When the emission accounting is carried out, the strategy is analyzed to obtain the emission calculation formula therein, because the subset data of different pollution sources is subjected to corresponding strategy calculation, and the specific formula of each type is described above.

[0230] And the subset data also contains the emission parameters of each index of the corresponding pollutant, so it needs to be extracted and imported into the formula for calculation. Here, the large language model is used to read the emission parameters in the coal-fired power plant emission data subset, and the corresponding emission parameters are imported into the corresponding calculation factors in the corresponding emission calculation formula for automatic accounting calculation.

[0231] The step of using the LLM large language model to read the corresponding emission parameters and import them into the corresponding calculation factors in the formula can refer to the above process of the large language model based on the prompt word for automatic operation. The keywords and calculation logic of the corresponding calculation factor can be defined in advance by the administrator, which can let the large language model extract the corresponding emission parameters based on the above logic and keywords, and import the parameters into the corresponding calculation factor, so as to calculate the corresponding emission according to the corresponding execution logic.

[0232] The step of using the LLM large language model to read the corresponding emission parameters in the data set and import them into the corresponding calculation factors in the preset formula can be performed according to the following process:

[0233] 1. Data preparation:

[0234] Ensure that the coal and electricity emission data set has been sorted into a format that the LLM large language model can understand, such as CSV, JSON, etc.

[0235] The data set should contain the required emission parameters, such as the type of emission, concentration, emission amount, etc.

[0236] Model selection and configuration:

[0237] Select a suitable LLM large language model, such as GPT series, BERT, etc.

[0238] Configure the model to ensure that it can process and understand text data.

[0239] Write instructions and scripts:

[0240] Write detailed instructions to tell the LLM large language model which emission parameters need to be read from the data set.

[0241] Write scripts or programs to connect the output of the LLM large language model with the preset formula.

[0242] 2. Read data:

[0243] Read the emission parameters in the data set through the LLM large language model. This may require inputting the data set as text into the model and instructing the model to extract specific information.

[0244] The model may return a piece of text containing the required parameters, or return a structured data format (such as JSON).

[0245] Data parsing and mapping:

[0246] Parse the output of the LLM large language model to extract specific emission parameter values.

[0247] Map these parameter values to corresponding calculation factors in a preset formula.

[0248] Formula calculation:

[0249] According to the mapped parameter values, execute the preset formula for calculation.

[0250] The formula can be a simple mathematical operation or a more complex model or algorithm.

[0251] 3. Result output and application:

[0252] Output the calculation result, which can be a specific numerical value, a set of numerical values, or a report in some form, determined according to the strategy.

[0253] Therefore, the large language model is used here to automatically perform the corresponding calculation process, thereby realizing the automatic emission accounting calculation process and improving the accounting efficiency.

[0254] In order to count the results of this emission accounting evaluation, the emission accounting results of each pollution source can be sequentially input into a preset report, and the corresponding emission source accounting list can be generated according to the report format, and the background server can synchronize the generated accounting list to the government server.

[0255] Figure 4 is a coal-fired power plant pollution reduction and carbon reduction collaborative control effectiveness accounting evaluation system block diagram according to an exemplary embodiment, which is used for coal-fired power plant pollution reduction and carbon reduction collaborative control effectiveness accounting evaluation method. Referring to Figure 4 , the system includes a coal-fired power plant data monitoring module 510, an LLM classification module 520, an accounting strategy recommendation module 530, and an emission calculation module 540 and a statistical module 550. Among them:

[0256] The coal-fired power plant data monitoring module 510 is used for real-time monitoring and collecting coal-fired power plant emission data sets with time stamps, including at least the following data: activity level data, generation coefficient data, end-of-pipe treatment facility information or automatic monitoring data;

[0257] The LLM classification module 520 is used for classifying the coal-fired power plant emission data set based on the LLM large language model to obtain coal-fired power plant emission data subsets of different pollution sources;

[0258] The accounting strategy recommendation module 530 is used for importing the coal-fired power plant emission data subset into a preset emission AI accounting evaluation model, identifying the emission characteristics of the coal-fired power plant emission data subset by the emission AI accounting evaluation model, and outputting and recommending an emission calculation strategy matched with the emission characteristics;

[0259] The emission calculation module 540 is used to calculate the corresponding emission amount of the coal-fired power emission data subset based on the recommended emission calculation strategy, and output the corresponding emission calculation result.

[0260] The statistics module 550 is used to calculate the emission amounts of each type of pollution source and generate a corresponding emission source accounting list according to a preset report format.

[0261] Please refer to the corresponding procedures and content in the above methods and steps to understand the functions and interactions of each module; they will not be repeated here.

[0262] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, such as... Figure 5 As shown, the electronic device may include the above-mentioned Figure 4 The system shown is an evaluation system for assessing the effectiveness of coordinated pollution reduction and carbon reduction control in coal-fired power plants. Optionally, electronic device 410 may include a first processor 2001.

[0263] Optionally, the electronic device 410 may also include a memory 2002 and a transceiver 2003.

[0264] The first processor 2001, memory 2002, and transceiver 2003 can be connected via a communication bus.

[0265] The following is combined with Figure 5 A detailed description of each component of electronic device 410 is provided below:

[0266] The first processor 2001 is the control center of the electronic device 410. It can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0267] Optionally, the first processor 2001 can perform various functions of the electronic device 410 by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0268] In a particular implementation, as an example, the first processor 2001 can include one or more CPUs, such as CPU0 and CPU1 shown in FIG. 20. Figure 5 In a particular implementation, as an example, the first processor 2001 can include one or more CPUs, such as CPU0 and CPU1 shown in FIG. 20.

[0269] In a particular implementation, as an example, the electronic device 410 can also include multiple processors, such as the first processor 2001 and the second processor 2004 shown in FIG. 20. Figure 5 In a particular implementation, as an example, the first processor 2001 can include one or more CPUs, such as CPU0 and CPU1 shown in FIG. 20.

[0270] The memory 2002 is configured to store a software program for implementing the solutions of the present application, and the first processor 2001 is configured to control the execution of the software program. The specific implementation can refer to the above-mentioned method embodiments, and will not be described here.

[0271] Alternatively, the memory 2002 can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, a magneto-optical disk, a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory 2002 can be integrated with the first processor 2001, or can exist independently and be coupled to the first processor 2001 through the interface circuit (not shown in FIG. 20) of the electronic device 410. The embodiments of the present application do not make a specific limitation in this regard. Figure 5 The transceiver 2003 is configured to communicate with a network device or a terminal device.

[0272] The transceiver 2003 is configured to communicate with a network device or a terminal device.

[0273] Optionally, the transceiver 2003 can include a receiver and a transmitter (not shown separately in FIG. 20). The receiver is configured to implement the receiving function, and the transmitter is configured to implement the transmitting function. Figure 5 The transceiver 2003 is configured to communicate with a network device or a terminal device.

[0274] Optionally, the transceiver 2003 can be integrated with the first processor 2001, or exist independently, and be coupled with the first processor 2001 through an interface circuit (not shown in the figure) of the electronic device 410, and the embodiments of the present application do not make a specific limitation hereon. Figure 5

[0275] It should be noted that, Figure 5 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router, and an actual knowledge structure identification device can include more or fewer components than those shown in the figure, or combine certain components, or different component arrangements.

[0276] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the coal enterprise pollution reduction and carbon reduction collaborative control effectiveness accounting and evaluation method described in the above method embodiments, which will not be described here again.

[0277] It should be understood that the first processor 2001 in the embodiments of the present application can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0278] ​It should also be understood that the memory in the embodiments of the present application can be volatile or nonvolatile memory, or can include both volatile and nonvolatile memory. The nonvolatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), or flash memory. The volatile memory can be random access memory (RAM) used as external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0279] The above-described embodiments can be implemented in whole or in part by software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center through wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid state disk.

[0280] It should be understood that the term "and / or" herein merely describes an association relationship of associated objects, which means that there can be three relationships, for example, A and / or B can represent three cases of A alone, A and B together, and B alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects, but can also represent an "and / or" relationship, which can be understood in the context before and after.

[0281] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or the like means any combination of the items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0282] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined by their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0283] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0284] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the devices, systems and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0285] In several embodiments provided by the present application, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are merely schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be through some interfaces, indirect coupling or communication connection between the systems or units, which can be electrical, mechanical or other forms.

[0286] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0287] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0288] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0289] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for evaluating the effectiveness of coordinated pollution reduction and carbon reduction control in coal-fired power plants, characterized in that, The method includes: S1. Monitor and collect timestamped coal-fired power plant emission datasets in real time, including at least the following data: fuel consumption data, equipment operation data, and coal quality information; S2. Based on the LLM large language model, classify the coal-fired power plant emission dataset to obtain coal-fired power plant emission data subsets of different pollution source classes; S3. Import the coal-fired power plant emission data subset into a preset emission AI accounting and evaluation model. The emission AI accounting and evaluation model identifies the emission characteristics of the coal-fired power plant emission data subset and recommends an emission calculation strategy that matches the emission characteristics. S4. Based on the recommended emission calculation strategy, calculate the corresponding emission amount for the coal-fired power emission data subset and output the corresponding emission calculation result; S5. Calculate the emission accounting results for each type of pollution source and generate the corresponding emission source accounting list according to the preset report format; The method for generating the emission AI accounting and assessment model includes: Collect and preprocess coal-fired power plant emission data from several different pollution sources, wherein the coal-fired power plant emission data includes historical fuel consumption data, equipment operation data and coal quality information of the corresponding pollution sources; Feature extraction is performed on the coal-fired power plant emission data to extract emission features from the coal-fired power plant emission data. The emission features include the emission source activity level of the corresponding pollution source in the coal-fired power plant emission data within a preset time period, that is: the difference between the emission amount of the pollution source and the emission threshold within the preset time period. The emission characteristics are labeled, and the labeling information includes: the pollution source attributes corresponding to the emission characteristics and the calculation strategy for calculating the emission amount of the pollution source corresponding to the emission characteristics; The emission characteristics corresponding to different pollution source types after labeling are collected to construct a feature set; The feature set is divided into a training set and a validation set according to a preset ratio; The training set is input into a preset random forest model for feature training and learning to generate the emission AI accounting and evaluation model. The predictive performance of the emissions AI accounting and assessment model was verified using the validation set: If the verification is successful, the emission AI accounting and assessment model will be deployed to the backend server; If the verification fails, repeat the above steps to rebuild the emission AI accounting and assessment model.

2. The method for evaluating the effectiveness of coordinated pollution reduction and carbon reduction control in coal-fired power plants according to claim 1, characterized in that, S1, real-time monitoring and collection of timestamped coal-fired power plant emission datasets, includes: Coal-fired power plants monitor and collect coal-fired power plant emission datasets through a back-end server at a preset sampling frequency and then report them to the government server of a credible institution. The credibility of coal-fired power companies and their reported coal-fired power emission datasets is verified through the aforementioned government server. If the review is approved, the coal-fired power plant emission dataset will be timestamped to generate a coal-fired power plant emission dataset with timestamps. If the review fails, the corresponding review comments will be sent to the coal and power company's back-end server.

3. The method for evaluating the effectiveness of coordinated pollution reduction and carbon reduction control in coal-fired power plants according to claim 1, characterized in that, S2, based on the LLM large language model, classifies the coal-fired power plant emission dataset to obtain subsets of coal-fired power plant emission data for different pollution source classes, including: Construct classification keywords and classification logic for different pollution source categories; By combining the aforementioned classification keywords and classification logic, a large language model prompt word is constructed; The prompt words of the large language model are input into the preset LLM large language model, and the prompt words are trained and learned by the LLM large language model. The coal-fired power plant emission dataset is traversed and read using the LLM large language model, and the coal-fired power plant emission dataset is classified based on the prompt words of the large language model to identify coal-fired power plant emission data subsets of different pollution source classes; Output and save subsets of coal-fired power plant emission data for the different pollution source types.

4. The method for evaluating the effectiveness of coordinated pollution reduction and carbon reduction control in coal-fired power plants according to claim 1, characterized in that, Step S3 involves importing the coal-fired power plant emission data subset into a preset emission AI accounting and evaluation model. The emission AI accounting and evaluation model identifies the emission characteristics of the coal-fired power plant emission data subset and recommends an emission calculation strategy that matches these characteristics. The LLM large language model sequentially outputs the coal-fired power emission data subsets of different pollution source types to the emission AI accounting and evaluation model; The emission AI accounting and evaluation model identifies the emission characteristics of the coal-fired power emission data subset, and based on the emission characteristics and their annotation information, retrieves and recommends the corresponding emission calculation strategy. The recommended emission calculation strategy is bound to the coal-fired power emission data subset corresponding to the emission characteristics.

5. The method for evaluating the effectiveness of coordinated pollution reduction and carbon reduction control in coal-fired power plants according to claim 4, characterized in that, Step S4 involves calculating the emissions of the coal-fired power plant emission data subset based on the recommended emission calculation strategy, and outputting the corresponding emission calculation results, including: The emission calculation strategy is analyzed to obtain the emission calculation formula. The emission parameters in the coal-fired power emission data subset are read using the LLM large language model, and the emission calculation formula is imported to calculate and output the corresponding emission calculation results. The emission calculation results are linked to the corresponding pollution sources and their coal-fired power plant emission data subsets.

6. A system for calculating and evaluating the effectiveness of coordinated pollution reduction and carbon reduction in coal-fired power plants, wherein the system is used to implement the method for calculating and evaluating the effectiveness of coordinated pollution reduction and carbon reduction in coal-fired power plants as described in any one of claims 1-5, characterized in that, The system includes: The coal-fired power plant data monitoring module is used to monitor and collect timestamped coal-fired power plant emission datasets in real time, including at least the following data: fuel consumption data, equipment operation data, and coal quality information; The LLM classification module is used to classify the coal-fired power plant emission dataset based on the LLM large language model to obtain coal-fired power plant emission data subsets of different pollution source classes; The accounting strategy recommendation module is used to import the coal-fired power emission data subset into a preset emission AI accounting evaluation model, and the emission AI accounting evaluation model identifies the emission characteristics of the coal-fired power emission data subset and recommends an emission calculation strategy that matches the emission characteristics. The emission calculation module is used to calculate the corresponding emission amount of the coal-fired power emission data subset based on the recommended emission calculation strategy, and output the corresponding emission calculation result. The statistics module is used to calculate the emission amounts of each type of pollution source and generate a corresponding emission source accounting list according to a preset report format.

7. An electronic device, characterized in that, The electronic device includes: processor; A memory storing computer-readable instructions that, when executed by the processor, implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains program code that can be invoked by a processor to execute the method as described in any one of claims 1 to 5.

Images