A method, device, equipment and medium for predicting electric carbon factor of a hydroelectric power station

Through the combination of intelligent learning models and carbon satellite data, the BP neural network model is used to predict the electric carbon factor of hydropower stations, which solves the problems of complex and low accuracy of data collection in the existing technology, and achieves rapid carbon emissions of hydropower stations , accurate prediction.

CN118761521BActive Publication Date: 2024-11-19STATE GRID SICHUAN ELECTRIC POWER CORP ELECTRIC POWER RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411251767.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-11-19
Estimated Expiration
2044-09-09

AI Technical Summary

Technical Problem

The existing electrocarbon factor accounting technology of hydropower generation has problems such as difficulty in collecting data, complex accounting processes, and impacting accuracy, making it difficult to effectively predict carbon emissions of hydropower plants.

Method used

Using intelligent learning model combined with real-time monitoring data of carbon satellites, the installed capacity, dam type, generator set type, floor area and geographical location of hydropower stations are predicted through the BP neural network model, and the electric carbon factor is obtained, and carbon satellites are used. The data corrects the prediction value to improve the accuracy of the prediction.

Benefits of technology

It realizes rapid and accurate prediction of carbon factors of hydropower stations, simplifies the data collection and accounting process, improves the timeliness and accuracy of carbon emission forecasts, and provides technical support for carbon accounting.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118761521B_ABST
    Figure CN118761521B_ABST
Patent Text Reader

Abstract

The present application discloses a method, device, equipment and medium for predicting the electric carbon factor of a hydroelectric power station. The present application belongs to the field of carbon dioxide monitoring technology. The method first analyzes the mechanism of carbon dioxide generation in the hydroelectric power station, and comprehensively analyzes the influence of factors such as installed capacity, dam type, generator type, floor area and geographical location on the electric carbon factor, thereby constructing a corresponding electric carbon factor prediction model to obtain the annual total electric carbon factor prediction value. The required original data is easy to collect and has a strong timeliness. At the same time, the predicted value is corrected according to the carbon satellite measured data, which further improves the accuracy and reliability of the prediction, helps to improve the accuracy of carbon accounting, and provides technical support for greenhouse gas accounting of hydroelectric power stations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the technical field of carbon dioxide monitoring, and specifically relates to a method, device, equipment and medium for predicting the electric carbon factor of a hydroelectric power station. Background Art

[0002] In the current global environment of carbon emissions and environmental protection and energy conservation, hydropower has attracted much attention as a clean energy source. However, recent studies have pointed out that although hydropower is considered a low-carbon energy source and is considered environmentally friendly, it actually also has greenhouse gas emissions problems, and its life cycle greenhouse emissions cannot be ignored. For example, researchers have conducted in-depth sorting and research on the life cycle greenhouse gas emissions of several hydropower stations at home and abroad. The results show that the carbon emissions of hydropower stations over their entire life cycle are about 18,000 tons. In the context of increasing carbon emissions, the hydropower industry has made outstanding contributions to reducing carbon emissions and has greatly curbed the growth of carbon emissions. Although the hydropower industry has made important contributions to reducing carbon emissions, as carbon emissions continue to increase, people's demand for green energy is also growing. Therefore, it is necessary to predict the carbon emissions of hydropower generation to provide data support for further reducing carbon emissions.

[0003] Usually, the carbon emissions of hydropower are generally characterized by the hydropower electricity carbon factor, which generally adopts the LCA (life cycle assessment) full life cycle carbon accounting technology. This accounting technology needs to divide the life cycle of the hydropower station into several different stages and requires a lot of data for accounting. In addition, the data required for this accounting technology is not easy to collect, the accounting process is complicated, and the choice of emission boundaries directly affects the accuracy of the accounting. There are disadvantages such as many uncertain factors and a long accounting cycle. Summary of the invention

[0004] In order to solve the problems existing in the existing hydroelectric carbon factor accounting technology, such as the difficulty in collecting the required data and the complex accounting process, this application proposes a method, device, equipment and medium for predicting the carbon factor of a hydroelectric power station. This application comprehensively considers the installed capacity, dam type, generator type, floor area and geographical location of the hydroelectric power station and combines the intelligent learning model to obtain the predicted annual total carbon factor of the hydroelectric power station to be predicted. At the same time, it combines the real-time monitoring data of carbon satellites and related power and electrical data to correct and obtain a more accurate prediction of the carbon factor. This application is based on an intelligent algorithm to predict the carbon factor of a hydroelectric power station throughout its life cycle. The required prediction data is easy to collect and has a strong timeliness. At the same time, it also uses the real-time monitoring data of carbon satellites for correction, which further improves the accuracy of the carbon factor prediction and helps to improve the accuracy of carbon accounting.

[0005] This application is implemented through the following technical solutions:

[0006] A method for predicting electric carbon factor of a hydroelectric power station, the method comprising:

[0007] Obtaining relevant information of the hydroelectric power station to be predicted and performing preprocessing, wherein the relevant information includes: installed capacity, dam type, generator type, land area and geographical location;

[0008] Classifying the hydroelectric power station to be predicted according to its characteristics, and inputting the preprocessed data into the electric carbon factor prediction model of the corresponding category to obtain the electric carbon factor prediction value of the hydroelectric power station to be predicted;

[0009] According to the carbon satellite monitoring data, the carbon emissions of the hydroelectric power station to be predicted during the operation phase are obtained by inversion;

[0010] The predicted value of the electric carbon factor is merged with the carbon emissions of the hydroelectric power station to be predicted during the operation phase obtained by inversion, and finally the total electric carbon factor of the hydroelectric power station to be predicted is obtained.

[0011] In some embodiments, the process of constructing the electric carbon factor prediction model specifically includes:

[0012] Obtaining basic information of several hydroelectric power stations and performing preprocessing; the basic information includes: installed capacity, dam type, generator type, land area, geographical location and electric carbon factor;

[0013] The preprocessed data are classified according to the characteristics of the hydroelectric power plants, and the corresponding electric carbon factor data sets are constructed according to the different types of hydroelectric power plants;

[0014] Normalizing the data in the electric-carbon factor data set, and dividing the normalized electric-carbon factor data set into a training set and a test set;

[0015] Construct a BP neural network model and initialize it;

[0016] The training set and the test set are used to train and test the BP neural network model to obtain an electric carbon factor prediction model.

[0017] In some embodiments, the pretreatment step specifically includes:

[0018] Missing value processing, outlier processing, data type conversion and deduplication processing.

[0019] In some embodiments, the step of classifying according to the characteristics of the hydroelectric power station specifically includes:

[0020] The basic data of the hydropower stations obtained are divided into three basic data: dam type, diversion type and hybrid type according to the type of hydropower station;

[0021] The three basic data are divided into three types: large hydropower stations, medium hydropower stations and small hydropower stations.

[0022] In some embodiments, the BP neural network model includes an input layer, a hidden layer, and an output layer;

[0023] The installed capacity, dam type, generator type, land area and geographical location are used as the input layer nodes, the annual total electricity carbon factor is used as the output layer node, and the number of nodes in the hidden layer is preliminarily determined according to the number of nodes in the input layer and the output layer;

[0024] The hidden layer activation function adopts a tangent S-type transfer function, the output layer activation function adopts a linear transfer function, the training function of the BP neural network model adopts a gradient descent BP algorithm training function, and the learning function of the BP neural network model adopts a momentum gradient descent weight and threshold learning function.

[0025] In some embodiments, the training and testing steps specifically include:

[0026] Inputting the sample data of the training set into the BP neural network model and outputting the predicted value;

[0027] By comparing the predicted value of the BP neural network model with the true value, the average relative error is calculated. If the accuracy requirement is not met and the number of iterations is within the set number, the BP neural network model is retrained after adjusting the model parameters until the training result meets the accuracy requirement or the number of training times reaches the number of iterations;

[0028] After the model training is completed, the sample data of the test set is input into the BP neural network model, and the predicted value is output. The predicted value of the BP neural network model is compared with the true value to calculate the average relative error. If the error requirement is not met, the model parameters are adjusted to retrain the BP neural network model until the model test meets the error requirement. At this time, the corresponding BP neural network model is the electric carbon factor prediction model.

[0029] In some embodiments, the inversion of the carbon emissions of the hydropower station to be predicted during the operation phase according to the carbon satellite monitoring data specifically includes:

[0030] Obtaining the average dry-air mixing ratio of the carbon dioxide column in the area where the hydroelectric power station to be predicted is located through a carbon satellite online monitoring system;

[0031] Calculating the total amount of dry air; the total amount of dry air is equal to the standard atmospheric pressure divided by the gravitational acceleration of the area where the hydroelectric power station to be predicted is located;

[0032] The carbon emissions in the area where the hydroelectric power station to be predicted is located are calculated based on the average dry air mixing ratio of the carbon dioxide column and the total amount of dry air; the carbon emissions in the area where the hydroelectric power station to be predicted is equal to the average dry air mixing ratio of the carbon dioxide column multiplied by the mass of dry air multiplied by the area of ​​land submerged by the hydroelectric power station to be predicted multiplied by the observation time multiplied by a coefficient;

[0033] According to the type of flooded land, the carbon emissions generated by the biodegradation of the flooded land during the operation phase of the predicted hydropower station are calculated;

[0034] The carbon emissions of the hydroelectric power station to be predicted in the operation phase are calculated based on the carbon emissions of the region where the hydroelectric power station to be predicted is located and the carbon emissions generated by biodegradation.

[0035] In some embodiments, the carbon emissions generated by biodegradation are calculated as follows:

[0036] ,

[0037] In the formula, Carbon emissions from biodegradation, is the carbon dioxide emission flux corresponding to the flooded land type, is the land area to be flooded by the hydroelectric power station to be predicted, is the degradation time;

[0038] The calculation method of the carbon emissions during the operation phase of the hydropower station to be predicted includes:

[0039] If the age of the hydropower station to be predicted is less than or equal to the degradation time, the carbon emissions calculation formula during the operation phase is:

[0040] ,

[0041] If the age of the hydropower station to be predicted is greater than the degradation time, the carbon emissions calculation formula in the operation stage is:

[0042] ,

[0043] in, is the carbon emissions during the operation phase, is the carbon emissions in the area where the hydroelectric power station to be predicted is located, is the service life of the hydroelectric power station to be predicted, is the degradation time, The observation time.

[0044] In some embodiments, the fusion process specifically includes:

[0045] Calculate the carbon emissions predicted by the model based on the predicted value of the electric carbon factor and the power data;

[0046] Calculate the total carbon emissions of the hydroelectric power station to be predicted based on the carbon emissions predicted by the model and the carbon emissions during the operation phase;

[0047] The total electric carbon factor is calculated according to the total carbon emissions, the power generation during the observation period, the service life of the hydroelectric power station to be predicted, and the observation period.

[0048] In a second aspect, the present application proposes a device for predicting the electric carbon factor of a hydroelectric power station, the device comprising:

[0049] A data acquisition and processing unit, the data acquisition and processing unit is used to acquire relevant information of the hydroelectric power station to be predicted and perform preprocessing; wherein the relevant information includes: installed capacity, dam type, generator type, land area and geographical location;

[0050] A first prediction unit, wherein the first prediction unit classifies the hydroelectric power station to be predicted according to its characteristics, and inputs the preprocessed data into a corresponding category of an electric carbon factor prediction model to obtain a predicted value of the electric carbon factor of the hydroelectric power station to be predicted;

[0051] a second prediction unit, wherein the second prediction unit inverts and obtains the carbon emissions of the hydropower station to be predicted during the operation phase according to the carbon satellite monitoring data;

[0052] And, a fusion correction unit, which fuses the predicted value of the electric carbon factor with the carbon emissions of the hydroelectric power station to be predicted during the operation phase obtained by inversion, and finally obtains the total electric carbon factor of the hydroelectric power station to be predicted.

[0053] In some embodiments, the device further comprises:

[0054] A model building unit is used to build the electric carbon factor prediction model based on a BP neural network.

[0055] In a third aspect, the present application proposes an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0056] In a fourth aspect, the present application proposes a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0057] The present application proposes a method, device, equipment and medium for predicting the electric carbon factor of a hydroelectric power station. First, the mechanism of carbon dioxide generation in the hydroelectric power station is analyzed, and the influence of factors such as installed capacity, dam type, generator type, land area and geographical location on the electric carbon factor is comprehensively considered, so as to construct a corresponding electric carbon factor prediction model and obtain the predicted value of the electric carbon factor. The required original data is easy to collect and has strong timeliness. At the same time, the carbon emissions of the hydroelectric power station in the operation stage are inverted according to the carbon satellite measured data, and the predicted value is corrected, which further improves the accuracy and reliability of the prediction, helps to improve the accuracy of carbon accounting, and provides technical support for greenhouse gas accounting of hydroelectric power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The drawings described herein are used to provide a further understanding of the embodiments of the present application, constitute a part of the present application, and do not constitute a limitation on the embodiments of the present application. In the drawings:

[0059] Figure 1 A flow chart of a method according to an embodiment of the present application;

[0060] Figure 2 A flow chart for model construction of an embodiment of the present application;

[0061] Figure 3 This is a flow chart of carbon satellite electric carbon factor prediction according to an embodiment of the present application;

[0062] Figure 4 This is a fusion flow chart of an embodiment of the present application;

[0063] Figure 5 This is a system principle block diagram of an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with examples and drawings. The illustrative implementation scheme of the present application and its description are only used to explain the present application and are not intended to limit the present application.

[0065] Embodiment: The current calculation of the carbon factor of hydroelectric power generation generally adopts the full life cycle assessment method. This method requires the life cycle of the hydroelectric power station to be divided into several different stages, and requires a lot of data, which is not easy to collect. The calculation process is complicated, and the selection of the emission boundary directly affects the accuracy of the calculation. In addition, there are many uncertain factors and a long calculation cycle.

[0066] The electricity carbon factor, also known as the grid emission factor or regional electricity carbon emission factor, is an indicator that measures the amount of carbon dioxide emissions per unit of electricity produced during electricity production. This factor is mainly used to estimate carbon dioxide emissions caused by electricity consumption and is an important parameter for greenhouse gas emission accounting. The electricity carbon factor is affected by many factors, including power generation structure (such as coal, gas, hydropower, nuclear energy, etc.), energy conversion efficiency, grid losses, and the proportion of renewable energy in electricity supply. Therefore, the existing carbon emission factor prediction technology based on historical electricity data and implemented using artificial intelligence algorithms is not suitable for the calculation of the electricity carbon factor for hydropower.

[0067] Based on this, this embodiment proposes a method for predicting the electric carbon factor of a hydroelectric power station. This embodiment analyzes the mechanism of carbon dioxide generation in hydroelectric power stations and classifies the factors affecting carbon emissions of hydroelectric power stations according to the carbon emission characteristics of hydroelectric power stations. For example, in the preparation stage, carbon emissions mainly come from the production of raw materials, and the quantity and type of raw materials produced have the greatest impact. The main characteristics affecting the type and quantity are installed capacity and dam type. In the construction and decommissioning stages, the main source of carbon dioxide emissions is generated during transportation and construction, and the amount of transportation directly depends on the size of the installed capacity and the type of dam. In the operation stage, the main source of carbon dioxide emissions is the decay and activity of underwater plants and animals after flooding of water bodies, but the flooding volume and area of ​​the dam will change in different seasons. Based on this, the greatest influencing factors, namely installed capacity, dam type, generator type, land area and geographical location, are comprehensively considered to influence the electric carbon factor. According to the historical impact data of these influencing factors on the annual total carbon factor, an electric carbon factor prediction model is constructed to realize the prediction of the electric carbon factor. At the same time, the carbon emissions of the hydropower station to be tested during the operation phase are obtained by inverting the carbon satellite monitoring data to obtain the final total electric carbon factor.

[0068] like Figure 1 As shown, the method proposed in this embodiment specifically includes the following steps:

[0069] Step 100, obtain relevant information of the hydroelectric power station to be predicted and perform preprocessing. The relevant information includes: installed capacity, dam type, generator type, land area and geographical location. Data preprocessing mainly includes: missing value processing, outlier processing, data type conversion, deduplication processing, etc.

[0070] Step 200, classify the hydroelectric power station to be predicted according to its characteristics, and input the preprocessed data into the electric carbon factor prediction model of the corresponding category to obtain the predicted value of the electric carbon factor of the hydroelectric power station to be predicted.

[0071] like Figure 2As shown, the electric carbon factor prediction model of this step is constructed based on the BP neural network, and the model construction process is as follows:

[0072] Step 201, basic information of several hydroelectric power stations is obtained and preprocessed.

[0073] The basic information includes: installed capacity, dam type, generator type, land area, geographical location and carbon factor.

[0074] Installed capacity refers to the rated active power of the generator sets actually installed in the hydroelectric power station. Dam types are divided into earth dams, gravity dams, concrete face rockfill dams, arch dams, etc. The power plant of some hydroelectric power stations is also part of the dam. A turbine is a hydraulic machine that converts water energy into mechanical energy. The equipment that uses a hydraulic machine to drive a generator to convert rotational mechanical energy into electrical energy becomes a hydroelectric generator set; according to the type of hydroelectric power station, the hydroelectric power station obtained is divided into three basic types: dam type, water diversion type and hybrid type. According to the provisions of the current "Classification and Design Standards for Water Conservancy and Hydropower Hub Projects" (SDJ12-78) in my country, hydroelectric power stations can be divided into three main types according to the scale of power generation: large hydroelectric power stations (>250MV), medium-sized hydroelectric power stations (25-250MV) and small hydroelectric power stations (0-25MV).

[0075] A hydroelectric power station is mainly composed of a dam, a reservoir, a power station, etc., and its land area is also determined by the total area of ​​the dam, the total area of ​​the reservoir and the total area of ​​the power station.

[0076] The geographical location refers to whether there is a river flowing through, whether there is a bag-shaped valley with a large storage capacity, whether the valley exit is narrow, etc.

[0077] This information is preprocessed to avoid the impact of defects and anomalies in the original data on the accuracy and reliability of the model. Specifically, it includes: (1) Missing value processing: Check whether there are missing values ​​in the data. For example, some hydropower stations may not provide all the information or some information is not recorded. For missing values, you can choose to delete the corresponding records, fill the missing values ​​(for example, use the mean or median) or use interpolation methods to fill. (2) Outlier processing: Identify and process outliers, which may have a negative impact on the performance of the model. Outliers may be data entry errors or represent real but rare situations. Common processing methods are to detect outliers through box plots or Z-score methods, and adjust or delete them according to actual conditions, such as processing points with large deviations in the data. (3) Data type conversion: Ensure the consistency of data types, such as converting text data into numerical data to facilitate subsequent modeling and analysis. (4) Deduplication processing: Check whether there are duplicate records. If so, delete the duplicate records to avoid introducing unnecessary bias in the modeling process.

[0078] Step 202, classify the preprocessed data according to the characteristics of the hydroelectric power station, and construct corresponding electric carbon factor data sets according to different types of hydroelectric power stations.

[0079] According to the carbon emission characteristics of hydroelectric power stations, their main carbon emissions are concentrated in the construction stage, operation stage and decommissioning stage. Since different types of hydroelectric power stations mainly affect the types of building consumables in the construction stage and recycled materials in the decommissioning stage, the basic data of hydroelectric power stations obtained are divided into three basic types according to the type of hydroelectric power station: dam type, diversion type and mixed type.

[0080] According to the provisions of the current "Classification and Design Standards for Water Conservancy and Hydropower Projects" (SDJ12-78) in my country, the above three basic types of data are further divided into three types: large hydropower stations (>250MV), medium hydropower stations (25-250MV) and small hydropower stations (0-25MV). Combined with the above types, diversion hydropower stations rely on the natural flow of rivers and do not have large-scale reservoirs to store water, so the installed capacity is usually small to medium, and they are divided into diversion small hydropower stations and diversion medium hydropower stations; dam-type hydropower stations can store a large amount of water, regulate water flow, and provide a stable high head, so the installed capacity is usually large, and dam-type hydropower stations can be divided into dam-type medium hydropower stations and dam-type large hydropower stations. Hybrid hydropower stations are usually used to regulate power grid loads, and large installed capacity is divided into hybrid medium hydropower stations and hybrid medium hydropower stations. This embodiment refines the classification of hydropower stations and constructs an electric carbon factor data set by class for subsequent model training to obtain prediction models of different types of hydropower stations, thereby improving prediction accuracy.

[0081] Step 203, normalize the data in the electric carbon factor data set, and divide the normalized electric carbon factor data set into a training set and a test set. In step 203, the multiple influencing factors have the same measurement scale through normalization, so as to avoid the accuracy and reliability of the model being affected by different measurement scales. The normalization method can adopt existing technologies such as the minimum-maximum normalization method and the Z-score normalization method, wherein the minimum-maximum normalization method uses formula (1) to scale the original data in the electric carbon factor data set to between [0,1].

[0082] (1)

[0083] In the formula, Indicates The original value of the feature, represents the minimum characteristic value, represents the maximum value of the feature, Indicates The normalized value of a feature. A feature is a type of basic information, such as installed capacity. This method is suitable for static data, that is, scenarios where the data does not change.

[0084] The Z-score normalization method uses formula (2) to scale the original data in the electro-carbon factor dataset to between [0, 1];

[0085] (2)

[0086] In the formula, Indicates The mean of the features, Indicates The standard deviation of the feature, Indicates The original value of the feature, Indicates This method is suitable for dynamic data, that is, scenarios where data changes.

[0087] This step 203 can be implemented by random partitioning or classification partitioning to divide the training set and the test set. Random partitioning is to divide most of the samples in the data set into the training set, and a small number of samples into the test set. The specific allocation ratio is set according to the specific problem and data size. Classification partitioning is to divide the samples of each category according to a certain ratio to avoid subsequent model deviation caused by sample imbalance.

[0088] Preferably, in this step, the normalized electric-carbon factor data set is randomly divided into a training set and a test set in a ratio of 80% and 20%, respectively. The training set is used to train the model, and the test set is used to evaluate the model performance.

[0089] Step 204: construct a BP neural network model and initialize it.

[0090] This step 204 selects to establish a three-layer BP neural network model, that is, the model includes an input layer, a hidden layer and an output layer. The characteristics of the hydropower station including installed capacity, dam type, generator type, land area and geographical location are used as input layer nodes, that is, as the input of the model, and the annual total electricity carbon factor is used as the node of the output layer, that is, as the output of the model. The number of nodes in the hidden layer can be calculated by an empirical formula to obtain an initial value, and the mean relative error MRE of the model is used for optimization and adjustment in the subsequent training process to obtain the optimal number of hidden layer nodes.

[0091] The number of nodes in the hidden layer can be calculated using any of the empirical formulas shown in equations (3) to (5) to obtain the initial value:

[0092] (3)

[0093] (4)

[0094] (5)

[0095] In the formula, represents the number of nodes in the hidden layer, Indicates the number of input layer nodes, represents the number of nodes in the output layer, is a constant between [1-10].

[0096] After determining the BP neural network structure, step 204 also needs to determine the activation function, training function, learning function and training parameters. The activation function is used to map the input of the neuron to an important function of the output. The training function adjusts the weights and thresholds from a global perspective, and the learning function adjusts the weights and thresholds from a local perspective. Preferably, in step 204, the tansig function, i.e., the tangent S-type transfer function, is used as the hidden layer activation function, and the purelin function, i.e., the linear transfer function, is used as the output layer activation function to process nonlinear data, convert the input data into useful output data, output linear data, and improve the performance and expression ability of the BP neural network model. The training function uses the traingd function, i.e., the BP algorithm training function of gradient descent, and considers the overall error to be minimal, and adjusts the weights and thresholds from a global perspective. The learning function selects the learngdm function, i.e., the momentum gradient descent weight and threshold learning function, and considers the local error to be minimal, and adjusts the weights and thresholds from a local perspective.

[0097] The output of the learning function is the increment of the weight and threshold, and the output of the training function is the trained network and training records. During the training process, the training function continuously calls the learning function to correct the weight and threshold, and ends the training by checking the set number of training steps or the error calculated by the performance function is less than the set error.

[0098] Determine various training parameters, including: training target error, learning rate, maximum number of training times, training time, and maximum number of iterations. These training parameters can be selected from experience. In addition, in order to improve the training efficiency and stability of the model, optimization and adjustment can also be performed in the subsequent training process to find the best parameter combination.

[0099] Step 205, train and test the BP neural network model using the training set and the test set to obtain the electric carbon factor prediction model. This step inputs the training set samples into the BP neural network model, outputs the predicted value, compares the predicted value of the model with the true value, and calculates the average relative error. If the accuracy requirement is not met and within the set number of iterations, the model parameters (including the number of hidden layer nodes, activation function, training function, learning function, and training parameters, etc.) are adjusted and the network model is retrained until the training result meets the accuracy requirement or the number of training reaches the number of iterations, and the model training phase is completed. Then enter the model testing phase, input the test set samples into the BP neural network model, output the predicted value, compare the predicted value of the model with the true value, verify the accuracy of the predicted value, calculate the average relative error, if the error requirement is not met, adjust the model parameters and retrain the network model until the error requirement is met in the model testing phase, and the model testing phase is completed. At this time, the corresponding BP neural network model is the electric carbon factor prediction model. The trained electric carbon factor prediction model is used to predict the electric carbon factor of the hydropower station that needs to be predicted.

[0100] Step 300: Invert the carbon emissions of the hydroelectric power station to be predicted during the operation phase based on the carbon satellite monitoring data. Figure 3 As shown, the specific process is as follows:

[0101] Step 301, obtaining the average dry-air mixing ratio of the carbon dioxide column in the area where the hydroelectric power station to be predicted is located through the carbon satellite online monitoring system.

[0102] Step 302, based on the average dry air mixing ratio of the carbon dioxide column, calculate the carbon emissions of the area where the hydroelectric power station to be predicted is located. The specific process is as follows:

[0103] First, calculate the total amount of dry air in the area where the hydroelectric power station to be predicted is located , the calculation formula is:

[0104] (6)

[0105] In the formula, is the gravity acceleration of the area where the hydroelectric power station to be predicted is located; is standard atmospheric pressure.

[0106] Then, the carbon emissions of the area where the hydropower station to be predicted is calculated by formula (7): :

[0107] (7)

[0108] In the formula, is the average dry air mixing ratio of the carbon dioxide column, The area of ​​land flooded for hydroelectric power stations, The observation time.

[0109] Step 303, based on the type of flooded land, calculate the carbon emissions generated by the biodegradation of the flooded land during the operation phase of the hydropower station to be predicted.

[0110] In the early stage of operation of a hydroelectric power station (for example, the operation time is less than or equal to 10 years), a large amount of carbon dioxide will be produced due to the biological degradation of organic matter in the flooded land. After a certain period of time, according to statistical analysis, generally after 10-15 years, the biological degradation of organic matter in the flooded land will basically no longer produce carbon dioxide. The carbon emissions calculated by inversion of carbon satellite observation data include the carbon dioxide produced by biodegradation throughout the life cycle of the hydroelectric power station. Therefore, it is necessary to consider the carbon emissions produced by biodegradation of flooded land and correct the carbon emissions obtained by inversion calculation.

[0111] Specifically, the calculation formula for carbon emissions from biodegradation is:

[0112] (8)

[0113] In the formula, Carbon emissions from biodegradation, is the carbon dioxide emission flux corresponding to the flooded land type, is the land area to be flooded by the hydroelectric power station to be predicted, is the degradation time.

[0114] Among them, the carbon dioxide emission flux can be obtained by looking up the table. Table 1 shows the carbon emission flux corresponding to various land types.

[0115] Table 1

[0116]

[0117] Step 304, the carbon emissions of the hydroelectric power station to be predicted in the operation phase are calculated based on the carbon emissions of the area where the hydroelectric power station to be predicted is located and the carbon emissions caused by biodegradation.

[0118] This step uses the carbon emissions generated by biodegradation to correct the carbon emissions obtained by inverting carbon satellite monitoring data to obtain the carbon emissions in the operation phase. The specific process is as follows:

[0119] If the age (i.e., operating time) of the hydropower station to be predicted is less than or equal to the degradation time, the calculation formula for the carbon emissions during the operating phase of the entire life cycle of the hydropower station to be predicted is:

[0120] (9)

[0121] If the age of the hydropower station to be predicted is greater than the degradation time, the calculation formula for the carbon emissions during the operation phase of the entire life cycle of the hydropower station to be predicted is:

[0122] (10)

[0123] in, is the carbon emissions during the operation phase, is the carbon emissions in the area where the hydroelectric power station to be predicted is located, is the service life of the hydroelectric power station to be predicted, is the degradation time, The observation time.

[0124] The calculation principle of the above formula is: when the age of the hydropower station is less than or equal to the degradation time, the carbon emissions calculated by inversion of carbon satellite monitoring data include the carbon emissions generated by biodegradation. Therefore, when calculating the carbon emissions generated in the operation stage of the entire life cycle, the carbon emissions generated by biodegradation are included. However, biodegradation only produces carbon dioxide within a certain period of time. Therefore, it is necessary to exclude the carbon dioxide generated by biodegradation after the degradation time has exceeded. Therefore, formula (9) is used to calculate the carbon emissions in the operation stage. When the age of the hydropower station is greater than the degradation time, the carbon emissions calculated by inversion of carbon satellite monitoring data do not include the carbon emissions generated by biodegradation. Therefore, when calculating the carbon emissions generated in the operation stage of the entire life cycle, the carbon emissions generated by biodegradation are not considered. However, in the early stage of the operation of the hydropower station, biodegradation produces a large amount of carbon emissions, so it needs to be taken into account. Therefore, formula (10) is used to calculate the carbon emissions in the operation stage.

[0125] Step 400, the predicted value of the electric carbon factor is merged with the carbon emissions of the hydropower station to be predicted during the operation phase obtained by inversion, and finally the total electric carbon factor of the hydropower station to be predicted is obtained. This step corrects the predicted value of the electric carbon factor by using carbon satellite monitoring data, thereby obtaining a more accurate total electric carbon factor.

[0126] like Figure 4 As shown, step 400 specifically includes the following process:

[0127] Step 401, calculate the carbon emissions predicted by the model based on the predicted value of the electric carbon factor and the power data. The calculation formula is:

[0128] (11)

[0129] in, is the predicted value of the electric carbon factor, is the power generation during the observation time, is the carbon emissions predicted by the model.

[0130] Step 402, based on the carbon emissions predicted by the model and the carbon emissions during the operation phase, the total carbon emissions of the hydropower station to be predicted are calculated. The calculation formula is:

[0131] (12)

[0132] in, is the total carbon emissions.

[0133] Step 403, calculate the total electricity carbon factor based on the total carbon emissions. The calculation formula is:

[0134]

[0135] in, Compared with the traditional accounting method, the method proposed in this embodiment is simple to collect data, the accounting process is also relatively simple, and has strong timeliness. It can provide a reference for the calculation of the electric carbon factor of the hydropower station and help improve the accuracy of carbon accounting.

[0136] Based on the same technical concept as above, this embodiment also proposes a device for predicting the carbon factor of a hydroelectric power station. Figure 5 As shown, the device specifically includes:

[0137] A data acquisition and processing unit, the data acquisition and processing unit is used to acquire relevant information of the hydroelectric power station to be predicted and pre-process the relevant information. The relevant information includes: installed capacity, dam type, generator type, land area and geographical location;

[0138] The first prediction unit classifies the hydroelectric power station to be predicted according to its characteristics, and inputs the preprocessed data into the electric carbon factor prediction model of the corresponding category to obtain the predicted value of the electric carbon factor of the hydroelectric power station.

[0139] The second prediction unit uses the carbon satellite monitoring data to invert and obtain the carbon emissions of the hydroelectric power station to be predicted during the operation phase.

[0140] And, a fusion correction unit, which fuses the predicted value of the electric carbon factor with the carbon emissions of the hydropower station to be predicted during the operation phase obtained by inversion, and finally obtains the total electric carbon factor of the hydropower station to be predicted.

[0141] Furthermore, the device also includes:

[0142] The model building unit builds an electric carbon factor prediction model based on the BP neural network. The specific process is as described in steps 201 to 205 of the above method, and will not be repeated here.

[0143] It should be noted that the specific implementation method of each functional module unit of the device is as described in the corresponding steps of the above method, and will not be repeated here.

[0144] This embodiment also provides a computer device for executing the above method of this embodiment.

[0145] Computer devices include processors, memory, and system buses; various device components including memory and processors are connected to the system bus. The processor is a piece of hardware used to execute computer program instructions through basic arithmetic and logical operations in a computer system. Memory is a physical device used to temporarily or permanently store computer programs or data (e.g., program state information). The system bus can be any of the following types of bus structures, including a memory bus or memory controller, a peripheral bus, and a local bus. The processor and memory can communicate data through the system bus. The memory includes read-only memory (ROM) or flash memory (not shown in the figure), and random access memory (RAM), RAM usually refers to the main memory loaded with the operating system and computer programs.

[0146] Computer equipment generally includes a storage device. The storage device can be selected from a variety of computer-readable media, which refers to any available media that can be accessed by the computer equipment, including both removable and fixed media. For example, computer-readable media include, but are not limited to, flash memory (micro SD card), CD-ROM, digital versatile disc (DVD) or other optical disk storage, magnetic tape cassette, magnetic tape, magnetic disk storage or other magnetic storage device, or any other medium that can be used to store the desired information and can be accessed by the computer equipment.

[0147] Computer devices can be logically connected to one or more network terminals in a network environment. Network terminals can be personal computers, servers, routers, smart phones, tablets, or other public network nodes. Computer devices are connected to network terminals through network interfaces (local area network LAN interfaces). Local area network (LAN) refers to a network of computers interconnected within a limited area, such as a home, school, computer laboratory, or office building using network media. WiFi and twisted pair cabling Ethernet are the two most commonly used technologies for building LANs.

[0148] It should be noted that other computer systems including more or fewer subsystems than a computer device may also be suitable for the invention.

[0149] As described in detail above, the computer device applicable to the present embodiment can perform the specified operations of the method for predicting the electric carbon factor of a hydroelectric power station. The computer device performs these operations in the form of software instructions running on a computer-readable medium by a processor. These software instructions can be read into the memory from a storage device or from another device through a local area network interface. The software instructions stored in the memory enable the processor to execute the above-mentioned method for predicting the electric carbon factor of a hydroelectric power station. In addition, the present invention can also be implemented by hardware circuits or hardware circuits combined with software instructions. Therefore, the implementation of this embodiment is not limited to any specific combination of hardware circuits and software.

[0150] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for predicting the electric carbon factor of a hydroelectric power station, characterized in that: The method comprises: Obtaining relevant information of the hydroelectric power station to be predicted and performing preprocessing, wherein the relevant information includes: installed capacity, dam type, generator type, land area and geographical location; Classifying the hydroelectric power station to be predicted according to its characteristics, and inputting the preprocessed data into the electric carbon factor prediction model of the corresponding category to obtain the electric carbon factor prediction value of the hydroelectric power station to be predicted; According to the carbon satellite monitoring data, the carbon emissions of the hydroelectric power station to be predicted during the operation phase are obtained by inversion; The predicted value of the electric carbon factor is integrated with the carbon emissions of the hydropower station to be predicted during the operation phase obtained by inversion, and finally the total electric carbon factor of the hydropower station to be predicted is obtained; the construction process of the electric carbon factor prediction model specifically includes: Obtaining basic information of several hydroelectric power stations and performing preprocessing; the basic information includes: installed capacity, dam type, generator type, land area, geographical location and electric carbon factor; The preprocessed data are classified according to the characteristics of the hydroelectric power plants, and the corresponding electric carbon factor data sets are constructed according to the different types of hydroelectric power plants; Normalizing the data in the electric-carbon factor data set, and dividing the normalized electric-carbon factor data set into a training set and a test set; Construct BP neural network model and initialize it; Using the training set and the test set, the BP neural network model is trained and tested to obtain an electric carbon factor prediction model; The carbon emissions of the hydropower station to be predicted during the operation phase are obtained by inverting the carbon satellite monitoring data, specifically including: Obtaining the average dry-air mixing ratio of the carbon dioxide column in the area where the hydroelectric power station to be predicted is located through a carbon satellite online monitoring system; Calculating the total amount of dry air; the total amount of dry air is equal to the standard atmospheric pressure divided by the gravitational acceleration of the area where the hydroelectric power station to be predicted is located; The carbon emissions in the area where the hydroelectric power station to be predicted is located are calculated based on the average dry air mixing ratio of the carbon dioxide column and the total amount of dry air; the carbon emissions in the area where the hydroelectric power station to be predicted is equal to the average dry air mixing ratio of the carbon dioxide column multiplied by the mass of dry air multiplied by the area of ​​land submerged by the hydroelectric power station to be predicted multiplied by the observation time multiplied by a coefficient; According to the type of flooded land, the carbon emissions generated by the biodegradation of the flooded land during the operation phase of the predicted hydropower station are calculated; Calculate the carbon emissions of the hydroelectric power station to be predicted during the operation phase according to the carbon emissions of the area where the hydroelectric power station to be predicted is located and the carbon emissions generated by biodegradation; The carbon emission calculation formula generated by biodegradation is: E jiangjie =F*A*T*8760*10 6 In the formula, E jiangjie is the carbon emission generated by biodegradation, F is the carbon dioxide emission flux corresponding to the submerged land type, A is the submerged land area of ​​the hydropower station to be predicted, and T is the degradation time; The calculation method of the carbon emissions during the operation phase of the hydropower station to be predicted includes: If the age of the hydropower station to be predicted is less than or equal to the degradation time, the carbon emissions calculation formula during the operation phase is: If the age of the hydropower station to be predicted is greater than the degradation time, the carbon emissions calculation formula in the operation stage is: Among them, E yuny is the carbon emissions during the operation phase, is the carbon emission of the area where the hydroelectric power station to be predicted is located, LT is the service life of the hydroelectric power station to be predicted, T is the degradation time, and Δt is the observation time; The fusion process specifically includes: Calculate the carbon emissions predicted by the model based on the predicted value of the electric carbon factor and the power data; Calculate the total carbon emissions of the hydroelectric power station to be predicted based on the carbon emissions predicted by the model and the carbon emissions during the operation phase; The total electric carbon factor is calculated according to the total carbon emissions, the power generation during the observation period, the service life of the hydroelectric power station to be predicted, and the observation period.

2. A method for predicting the electric carbon factor of a hydroelectric power station according to claim 1, characterized in that: The pre-processing step specifically includes: Missing value processing, outlier processing, data type conversion and deduplication processing.

3. A method for predicting the electric carbon factor of a hydroelectric power station according to claim 1, characterized in that: The step of classifying according to the characteristics of the hydroelectric power station specifically includes: The basic data of the hydropower stations obtained are divided into three basic data: dam type, diversion type and hybrid type according to the type of hydropower station; The three basic data are divided into three types: large hydropower stations, medium hydropower stations and small hydropower stations.

4. A method for predicting the electric carbon factor of a hydroelectric power station according to claim 1, characterized in that: The BP neural network model includes an input layer, a hidden layer and an output layer; The installed capacity, dam type, generator type, land area and geographical location are used as the input layer nodes, the annual total electricity carbon factor is used as the output layer node, and the number of nodes in the hidden layer is preliminarily determined according to the number of nodes in the input layer and the output layer; The hidden layer activation function adopts a tangent S-type transfer function, the output layer activation function adopts a linear transfer function, the training function of the BP neural network model adopts a gradient descent BP algorithm training function, and the learning function of the BP neural network model adopts a momentum gradient descent weight and threshold learning function.

5. A method for predicting the electric carbon factor of a hydroelectric power station according to claim 1, characterized in that: The training and testing steps specifically include: Inputting the sample data of the training set into the BP neural network model and outputting the predicted value; By comparing the predicted value of the BP neural network model with the true value, the average relative error is calculated. If the accuracy requirement is not met and the number of iterations is within the set number, the BP neural network model is retrained after adjusting the model parameters until the training result meets the accuracy requirement or the number of training times reaches the number of iterations; After the model training is completed, the sample data of the test set is input into the BP neural network model, and the predicted value is output. The predicted value of the BP neural network model is compared with the true value to calculate the average relative error. If the error requirement is not met, the model parameters are adjusted to retrain the BP neural network model until the model test meets the error requirement. At this time, the corresponding BP neural network model is the electric carbon factor prediction model.

6. A device for predicting carbon factor of a hydroelectric power station, characterized in that: The device comprises: A data acquisition and processing unit, the data acquisition and processing unit is used to acquire relevant information of the hydroelectric power station to be predicted and perform preprocessing; wherein the relevant information includes: installed capacity, dam type, generator type, land area and geographical location; A first prediction unit, wherein the first prediction unit classifies the hydroelectric power station to be predicted according to its characteristics, and inputs the preprocessed data into a corresponding category of an electric carbon factor prediction model to obtain a predicted value of the electric carbon factor of the hydroelectric power station to be predicted; a second prediction unit, wherein the second prediction unit inverts and obtains the carbon emissions of the hydropower station to be predicted during the operation phase according to the carbon satellite monitoring data; and, a fusion correction unit, which fuses the predicted value of the electric carbon factor with the carbon emissions of the hydropower station to be predicted during the operation phase obtained by inversion, and finally obtains the total electric carbon factor of the hydropower station to be predicted; The construction process of the electric carbon factor prediction model specifically includes: Obtaining basic information of several hydroelectric power stations and performing preprocessing; the basic information includes: installed capacity, dam type, generator type, land area, geographical location and electric carbon factor; The preprocessed data are classified according to the characteristics of the hydroelectric power plants, and the corresponding electric carbon factor data sets are constructed according to the different types of hydroelectric power plants; Normalizing the data in the electric-carbon factor data set, and dividing the normalized electric-carbon factor data set into a training set and a test set; Construct BP neural network model and initialize it; Using the training set and the test set, the BP neural network model is trained and tested to obtain an electric carbon factor prediction model; The carbon emissions of the hydropower station to be predicted during the operation phase are obtained by inverting the carbon satellite monitoring data, specifically including: Obtaining the average dry-air mixing ratio of the carbon dioxide column in the area where the hydroelectric power station to be predicted is located through a carbon satellite online monitoring system; Calculating the total amount of dry air; the total amount of dry air is equal to the standard atmospheric pressure divided by the gravitational acceleration of the area where the hydroelectric power station to be predicted is located; The carbon emissions in the area where the hydroelectric power station to be predicted is located are calculated based on the average dry air mixing ratio of the carbon dioxide column and the total amount of dry air; the carbon emissions in the area where the hydroelectric power station to be predicted is equal to the average dry air mixing ratio of the carbon dioxide column multiplied by the mass of dry air multiplied by the area of ​​land submerged by the hydroelectric power station to be predicted multiplied by the observation time multiplied by a coefficient; According to the type of flooded land, the carbon emissions generated by the biodegradation of the flooded land during the operation phase of the predicted hydropower station are calculated; Calculate the carbon emissions of the hydroelectric power station to be predicted during the operation phase according to the carbon emissions of the area where the hydroelectric power station to be predicted is located and the carbon emissions generated by biodegradation; The carbon emission calculation formula generated by biodegradation is: E jiangjie =F*A*T*8760*10 6 In the formula, E jiangjie is the carbon emission generated by biodegradation, F is the carbon dioxide emission flux corresponding to the submerged land type, A is the submerged land area of ​​the hydropower station to be predicted, and T is the degradation time; The calculation method of the carbon emissions during the operation phase of the hydropower station to be predicted includes: If the age of the hydropower station to be predicted is less than or equal to the degradation time, the carbon emissions calculation formula during the operation phase is: If the age of the hydropower station to be predicted is greater than the degradation time, the carbon emissions calculation formula in the operation stage is: Among them, E yuny is the carbon emissions during the operation phase, is the carbon emission of the area where the hydroelectric power station to be predicted is located, LT is the service life of the hydroelectric power station to be predicted, T is the degradation time, and Δt is the observation time; The fusion process specifically includes: Calculate the carbon emissions predicted by the model based on the predicted value of the electric carbon factor and the power data; Calculate the total carbon emissions of the hydroelectric power station to be predicted based on the carbon emissions predicted by the model and the carbon emissions during the operation phase; The total electric carbon factor is calculated according to the total carbon emissions, the power generation during the observation period, the service life of the hydroelectric power station to be predicted, and the observation period.

7. A device for predicting electric carbon factor of a hydroelectric power station according to claim 6, characterized in that: The device also includes: A model building unit is used to build the electric carbon factor prediction model based on a BP neural network.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.