Methods and systems for carbon emission prediction in the power industry
By modifying the support vector machine regression model using a genetic algorithm, and combining the Gaussian radial basis kernel function and the Lagrangian function, the problem of carbon emission prediction easily getting trapped in local optima is solved, achieving higher reliability and accuracy.
Patent Information
- Application Number
- CN202311455914.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-03
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-11-03
AI Technical Summary
Existing carbon emission prediction schemes based on support vector machines are prone to getting trapped in local optima, leading to a decline in the accuracy and reliability of carbon emission predictions in the power industry.
A genetic algorithm is used to modify the support vector machine regression model. A carbon emission prediction model is constructed by combining the Gaussian radial basis kernel function and the Lagrange function. The training dataset is obtained through data cleaning and normalization to avoid local optima.
This improves the reliability and accuracy of carbon emission forecasts, making the forecast results more objective and scientific.
Smart Images

Figure CN117493987B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of electrical automation, and specifically relates to a carbon emission prediction method and system for the power industry. Background Technology
[0002] With the development of economy and technology and the improvement of people's living standards, electricity has become an indispensable secondary energy source in people's production and life, bringing endless convenience to people's production and life.
[0003] Currently, the primary source of electricity worldwide is still thermal power generation. The main fuel for thermal power generation is fossil fuels such as coal. Therefore, with the proliferation of thermal power systems, fossil fuels are used extensively, leading to a year-on-year increase in carbon emissions and the gradual emergence of environmental problems. Therefore, forecasting carbon emissions in the power industry can assist in future development planning and help the industry formulate corresponding strategies to control the growth of carbon emissions.
[0004] Currently, traditional carbon emission prediction schemes for the power industry primarily employ support vector machine (SVM)-based approaches. These schemes can predict carbon emissions in the power industry relatively well based on limited sample data. However, SVM-based schemes are highly susceptible to getting trapped in local optima, leading to a decrease in prediction accuracy and reliability. Summary of the Invention
[0005] One of the objectives of this invention is to provide a highly reliable, accurate, and objective scientific method for predicting carbon emissions in the power industry.
[0006] The second objective of this invention is to provide a system for implementing the aforementioned carbon emission prediction method for the power industry.
[0007] The carbon emission prediction method for the power industry provided by this invention includes the following steps:
[0008] S1. Obtain historical carbon emission data for the power industry in the target area;
[0009] S2. Preprocess the data obtained in step S1 to obtain the training dataset;
[0010] S3. Construct a basic model for carbon emission prediction based on the support vector machine regression model;
[0011] S4. The constructed carbon emission prediction basic model is modified using a genetic algorithm, and the modified carbon emission prediction basic model is trained using the training dataset obtained in step S2 to obtain the carbon emission prediction model.
[0012] S5. Using the carbon emission prediction model obtained in step S4, predict carbon emissions for the power industry in the target region.
[0013] Step S1, which involves obtaining historical carbon emission data of the power industry in the target area, specifically includes the following steps:
[0014] The carbon emission sources of the power industry in the target area were identified as the combustion of fossil fuels and the purchased and used electricity.
[0015] Obtain historical data on coal consumption and corresponding carbon emissions, electricity purchases and corresponding carbon emissions, and electricity sales for the power industry in the target region for each year.
[0016] Step S2 involves preprocessing the data obtained in step S1 to obtain the training dataset, specifically including the following steps:
[0017] The data obtained in step S1 is cleaned and normalized to obtain the training dataset.
[0018] Step S3, which involves constructing a basic carbon emission prediction model based on a support vector machine regression model, specifically includes the following steps:
[0019] Based on the relationship between various carbon emission sources and carbon emission data, support vector machine regression models are selected respectively, along with corresponding kernel functions and parameters;
[0020] The kernel function for the support vector machine regression model was selected as the Gaussian radial basis function, and the following was adopted: Radial action constraint; Gaussian radial basis kernel function is expressed as:
[0021] In the formula It is the Gaussian radial basis kernel function; This is a radial range limiting parameter; Let i be the i-th sample point; Let j be the j-th sample point; To obtain the distance between two samples;
[0022] Setting kernel function parameters for ;
[0023] Introducing the Lagrange function, the objective function of the support vector machine regression model is:
[0024] In the formula The objective function for the support vector machine; The number of samples; Let i be the slack variable for the i-th sample point; Let be the symmetric relaxation variable for the i-th sample point; Let j be the slack variable for the j-th sample point; Let j be the symmetric slack variable for the j-th sample point; This is the range coefficient; The vertical coordinates of the sample points are 2D.
[0025] Step S4 involves using a genetic algorithm to correct the constructed basic carbon emission prediction model, and then using the training dataset obtained in step S2 to train the corrected basic carbon emission prediction model to obtain the carbon emission prediction model. This process specifically includes the following steps:
[0026] The genetic algorithm was used to revise the basic carbon emission prediction model.
[0027] Population initialization for the genetic algorithm, kernel function parameters The penalty parameter C is used as the objective function, and chromosome encoding is performed.
[0028] The fitness function value is calculated to determine the quality of individuals in the population: the model's prediction accuracy is used as the fitness function value, resulting in the fitness function F. n is the number of individuals. The predicted value of the model. This is the actual value. It is an absolute value function;
[0029] Selection is performed using roulette wheel selection, mutation using uniform mutation, and crossover using linear combination. The selection, crossover, and mutation operations are repeatedly iterated through the population.
[0030] The determination is as follows: If the number of iterations meets the set requirements, the best individual is selected, and the corresponding optimal kernel function parameters are obtained. And the penalty parameter C; if the number of iterations does not meet the set requirement, continue to the next round of iterations;
[0031] Finally, the revised basic model for carbon emission prediction is expressed as follows:
[0032] In the formula For the basic model of carbon emissions; This is a correction constant;
[0033] Finally, the modified carbon emission prediction base model is trained using the training dataset obtained in step S2 to obtain the carbon emission prediction model.
[0034] Step S5, which uses the carbon emission prediction model obtained in step S4 to predict carbon emissions for the power industry in the target region, specifically includes the following steps:
[0035] Based on the data obtained in step S1, determine the growth rate of each data item;
[0036] The forecast data for the power industry in the target region for the forecast year is determined based on the growth rate.
[0037] The predicted data is input into the carbon emission prediction model obtained in step S4 to complete the carbon emission prediction for the power industry in the target region.
[0038] This invention also provides a system for implementing the carbon emission prediction method for the power industry, comprising a data acquisition module, a data processing module, a model building module, a model training module, and a carbon emission prediction module; the data acquisition module, data processing module, model building module, model training module, and carbon emission prediction module are connected in series; the data acquisition module is used to acquire historical carbon emission data of the power industry in the target area and upload the data to the data processing module; the data processing module is used to preprocess the received data to obtain a training dataset and upload the data to the model building module; the model building module is used to construct a basic carbon emission prediction model based on a support vector machine regression model based on the received data and upload the data to the model training module; the model training module is used to correct the constructed basic carbon emission prediction model using a genetic algorithm based on the received data, and train the corrected basic carbon emission prediction model using the training dataset to obtain a carbon emission prediction model and upload the data to the carbon emission prediction module; the carbon emission prediction module is used to predict carbon emissions of the power industry in the target area based on the received data and the carbon emission prediction model.
[0039] The carbon emission prediction method and system for the power industry provided by this invention avoids getting trapped in local optima by using a genetic algorithm to correct the support vector machine regression model. Therefore, this invention has higher reliability, better accuracy, and is more objective and scientific. Attached Figure Description
[0040] Figure 1 This is a schematic diagram of the method flow of the present invention.
[0041] Figure 2 This is a schematic diagram of the functional modules of the system of the present invention. Detailed Implementation
[0042] like Figure 1The diagram shown is a flowchart of the method of the present invention: The carbon emission prediction method for the power industry disclosed in this invention includes the following steps:
[0043] S1. Obtain historical carbon emission data for the power industry in the target region; specifically including the following steps:
[0044] The carbon emission sources of the power industry in the target area were identified as the combustion of fossil fuels and the purchased and used electricity.
[0045] Obtain historical data on coal consumption and corresponding carbon emissions, electricity purchases and corresponding carbon emissions, and electricity sales in the target region's power industry for each year.
[0046] S2. Preprocess the data obtained in step S1 to obtain the training dataset; specifically, this includes the following steps:
[0047] The data obtained in step S1 is cleaned and normalized to obtain the training dataset.
[0048] S3. Construct a basic model for carbon emission prediction based on the support vector machine regression model; specifically including the following steps:
[0049] Based on the relationship between various carbon emission sources and carbon emission data, support vector machine regression models are selected respectively, along with corresponding kernel functions and parameters;
[0050] The kernel function for the support vector machine regression model was selected as the Gaussian radial basis function, and the following was adopted: Radial action constraint; Gaussian radial basis kernel function is expressed as:
[0051] In the formula It is the Gaussian radial basis kernel function; This is a radial range limiting parameter; Let i be the i-th sample point; Let j be the j-th sample point; To obtain the distance between two samples;
[0052] Setting kernel function parameters for ;
[0053] Introducing the Lagrange function, the objective function of the support vector machine regression model is:
[0054] In the formula The objective function for the support vector machine; The number of samples; Let i be the slack variable for the i-th sample point; Let be the symmetric relaxation variable for the i-th sample point; Let j be the slack variable for the j-th sample point; Let j be the symmetric slack variable for the j-th sample point; This is the range coefficient; The two-dimensional ordinate of the sample point;
[0055] S4. The constructed carbon emission prediction base model is modified using a genetic algorithm, and the modified carbon emission prediction base model is trained using the training dataset obtained in step S2 to obtain the carbon emission prediction model; specifically, the following steps are included:
[0056] The genetic algorithm was used to revise the basic carbon emission prediction model.
[0057] Population initialization for the genetic algorithm, kernel function parameters The penalty parameter C is used as the objective function, and chromosome encoding is performed.
[0058] The fitness function value is calculated to determine the quality of individuals in the population: the model's prediction accuracy is used as the fitness function value, resulting in the fitness function F. n is the number of individuals. The predicted value of the model. This is the actual value. It is an absolute value function;
[0059] Selection is performed using roulette wheel selection, mutation using uniform mutation, and crossover using linear combination. The selection, crossover, and mutation operations are repeatedly iterated through the population.
[0060] The determination is as follows: If the number of iterations meets the set requirements, the best individual is selected, and the corresponding optimal kernel function parameters are obtained. And the penalty parameter C; if the number of iterations does not meet the set requirement, continue to the next round of iterations;
[0061] Finally, the revised basic model for carbon emission prediction is expressed as follows:
[0062] In the formula For the basic model of carbon emissions; This is a correction constant;
[0063] Finally, the modified carbon emission prediction base model is trained using the training dataset obtained in step S2 to obtain the carbon emission prediction model.
[0064] Ultimately, two prediction models will be established: a coal consumption-carbon emission prediction model and an electricity purchase-carbon emission prediction model.
[0065] S5. Using the carbon emission prediction model obtained in step S4, predict carbon emissions for the power industry in the target region; specifically including the following steps:
[0066] Based on the data obtained in step S1, determine the growth rate of each data item; in specific implementation, determine the growth rate of coal consumption and electricity purchase.
[0067] The forecast data for the power industry in the target region in the forecast year is determined based on the growth rate; in practice, this mainly involves determining the coal consumption and electricity purchases of the power industry in the target region in the forecast year.
[0068] The predicted data is input into the carbon emission prediction model obtained in step S4 to complete the carbon emission prediction for the power industry in the target area. In specific implementation, the carbon emissions predicted by the coal consumption-carbon emission prediction model and the carbon emissions predicted by the electricity purchase-carbon emission prediction model are added together to obtain the final total carbon emission prediction value.
[0069] like Figure 2 The diagram shows the functional modules of the system of this invention: The system for implementing the carbon emission prediction method for the power industry disclosed in this invention includes a data acquisition module, a data processing module, a model building module, a model training module, and a carbon emission prediction module; the data acquisition module, data processing module, model building module, model training module, and carbon emission prediction module are connected in series; the data acquisition module is used to acquire historical carbon emission data of the power industry in the target area and upload the data to the data processing module; the data processing module is used to preprocess the data information according to the received data to obtain a training dataset and upload the data to the model building module; the model building module is used to construct a basic carbon emission prediction model based on a support vector machine regression model according to the received data and upload the data to the model training module; the model training module is used to correct the constructed basic carbon emission prediction model using a genetic algorithm according to the received data and train the corrected basic carbon emission prediction model using the training dataset to obtain a carbon emission prediction model and upload the data to the carbon emission prediction module; the carbon emission prediction module is used to predict carbon emissions of the power industry in the target area according to the received data and the carbon emission prediction model.
Claims
1. A method for predicting carbon emissions in the power industry, comprising the following steps: S1. Obtain historical carbon emission data for the power industry in the target region; specifically including the following steps: The carbon emission sources of the power industry in the target area were identified as the combustion of fossil fuels and the purchased and used electricity. Obtain historical data on coal consumption and corresponding carbon emissions, electricity purchases and corresponding carbon emissions, and electricity sales in the target region's power industry for each year. S2. Preprocess the data obtained in step S1 to obtain the training dataset; S3. Construct a basic model for carbon emission prediction based on the support vector machine regression model; S4. The constructed carbon emission prediction base model is modified using a genetic algorithm, and the modified carbon emission prediction base model is trained using the training dataset obtained in step S2 to obtain the carbon emission prediction model; specifically, the following steps are included: The genetic algorithm was used to revise the basic carbon emission prediction model. Population initialization is performed on the genetic algorithm, kernel function parameters and penalty parameter C as the objective function, and chromosome coding is performed; A fitness function value is calculated for judging the pros and cons of the population individuals: the prediction accuracy of the model is taken as the fitness function value, and a fitness function F is obtained , n is the number of individuals, is the prediction value of the model, is the actual value, is the absolute value function; Selection is performed using roulette wheel selection, mutation using uniform mutation, and crossover using linear combination. The selection, crossover, and mutation operations are repeatedly iterated through the population. The determination is as follows: If the number of iterations meets the set requirements, the best individual is selected, and the corresponding optimal kernel function parameters are obtained. And the penalty parameter C; if the number of iterations does not meet the set requirement, continue to the next round of iterations; Finally, the revised basic model for carbon emission prediction is expressed as follows: In the formula For the basic model of carbon emissions; This is a correction constant; The number of samples; Let i be the slack variable for the i-th sample point; Let be the symmetric relaxation variable for the i-th sample point; It is the Gaussian radial basis kernel function; Let i be the i-th sample point; Finally, the modified carbon emission prediction base model is trained using the training dataset obtained in step S2 to obtain the carbon emission prediction model. S5. Using the carbon emission prediction model obtained in step S4, predict carbon emissions for the power industry in the target region; specifically including the following steps: Based on the data obtained in step S1, determine the growth rate of each data item; The forecast data for the power industry in the target region for the forecast year is determined based on the growth rate. The predicted data is input into the carbon emission prediction model obtained in step S4 to complete the carbon emission prediction for the power industry in the target region.
2. The carbon emission prediction method for the power industry according to claim 1, characterized in that... Step S2 involves preprocessing the data obtained in step S1 to obtain the training dataset, specifically including the following steps: The data obtained in step S1 is cleaned and normalized to obtain the training dataset.
3. The carbon emission prediction method for the power industry according to claim 2, characterized in that... Step S3, which involves constructing a basic carbon emission prediction model based on a support vector machine regression model, specifically includes the following steps: Based on the relationship between various carbon emission sources and carbon emission data, support vector machine regression models are selected respectively, along with corresponding kernel functions and parameters; The kernel function for the support vector machine regression model was selected as the Gaussian radial basis function, and the following was adopted: Radial action constraint; Gaussian radial basis kernel function is expressed as: In the formula It is the Gaussian radial basis kernel function; This is a radial range limiting parameter; Let i be the i-th sample point; Let j be the j-th sample point; To obtain the distance between two samples; Setting kernel function parameters for ; Introducing the Lagrange function, the objective function of the support vector machine regression model is: In the formula The objective function for the support vector machine; The number of samples; Let i be the slack variable for the i-th sample point; Let be the symmetric relaxation variable for the i-th sample point; Let j be the slack variable for the j-th sample point; Let j be the symmetric slack variable for the j-th sample point; This is the range coefficient; The vertical coordinates of the sample points are 2D.
4. A system for implementing the carbon emission prediction method for the power industry according to any one of claims 1 to 3, characterized in that... It includes a data acquisition module, a data processing module, a model building module, a model training module, and a carbon emission prediction module; the data acquisition module, data processing module, model building module, model training module, and carbon emission prediction module are connected in series; the data acquisition module is used to acquire historical carbon emission data of the power industry in the target area and upload the data to the data processing module; The data processing module is used to preprocess the received data to obtain the training dataset and upload the data to the model building module. The model building module is used to build a basic carbon emission prediction model based on the received data and the support vector machine regression model, and upload the data to the model training module. The model training module is used to revise the constructed carbon emission prediction base model using a genetic algorithm based on the received data, and to train the revised carbon emission prediction base model using a training dataset to obtain the carbon emission prediction model. The data is then uploaded to the carbon emission prediction module. The carbon emission prediction module is used to predict carbon emissions from the power industry in the target region based on the received data and the carbon emission prediction model.
Citation Information
Patent Citations
Carbon emission prediction method for oil field transfer station system
CN115879638A
Data processing method and apparatus, electronic device and storage medium
WO2020140913A1
