Method for predicting full-life-cycle carbon emission of transformer substation
By performing phase division and data preprocessing of the entire life cycle of the substation and building and training a carbon emission prediction model, the problem of difficulty in accurately predicting the carbon emissions of the substation in the existing technology is solved, and rapid and accurate carbon emission prediction is achieved, supporting effective energy-saving and emission reduction strategies.
Patent Information
- Application Number
- CN202510246008.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-04
- Publication Date
- 2025-04-11
AI Technical Summary
The existing technology is difficult to accurately predict the carbon emissions in the entire life cycle of the substation, which leads to difficulties in formulating and implementing energy-saving and emission reduction strategies.
By dividing the entire life cycle of the substation, obtaining historical data and operation data of each carbon emission stage, pre-processing, building and training a carbon emission prediction model, and using a BP neural network for prediction.
It realizes rapid and accurate prediction of carbon emissions throughout the life cycle of the substation, improves the accuracy of the prediction model, and supports the formulation and implementation of effective energy-saving and emission reduction strategies.
Smart Images
Figure CN119809899A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of carbon emission prediction, and more specifically to a method for predicting carbon emissions of a transformer substation over its entire life cycle. Background Art
[0002] As global warming becomes increasingly prominent, the issue of carbon emissions has risen to a critical level on the international agenda. In the power industry, substations, as hub facilities of the power grid, generate carbon emissions during their operation that cannot be underestimated.
[0003] Accurately predicting the carbon emissions of a substation throughout its life cycle is the basis for power companies to formulate and implement energy-saving and emission reduction strategies. By deeply analyzing the sources and influencing factors of carbon emissions at each stage of the substation, companies can design targeted measures to reduce consumption and increase efficiency, such as optimizing equipment configuration, improving operating efficiency, and promoting green energy, thereby effectively reducing carbon emissions and improving corporate economic and social benefits; for government decision-making, accurately grasping the carbon emission data of substations will help formulate more scientific and reasonable environmental protection policies and industry standards. Based on these data, the government can evaluate the total carbon emissions of the power industry and its impact on the environment, and then formulate effective regulations and incentive policies to promote the entire power industry and even the entire society to develop in a low-carbon and green direction.
[0004] Substations not only consume a lot of energy in the transmission and distribution of electricity, but also may produce greenhouse gas emissions to varying degrees during the entire life cycle of equipment manufacturing, installation, operation, maintenance and waste disposal. Therefore, it is particularly important to accurately predict the carbon emissions of substations throughout their life cycle.
[0005] Therefore, it is an urgent problem for those skilled in the art to propose a method for predicting carbon emissions of a substation throughout its life cycle to solve the difficulties existing in the prior art. Summary of the invention
[0006] In view of this, the present invention provides a method for predicting carbon emissions of a substation over its entire life cycle, which is used to solve the technical problems existing in the prior art.
[0007] In order to achieve the above object, the present invention provides the following technical solutions:
[0008] A method for predicting carbon emissions of a substation throughout its life cycle comprises the following steps:
[0009] The carbon emissions of the substation throughout its life cycle are divided into stages to obtain the carbon emission stages of the substation throughout its life cycle;
[0010] Obtain historical carbon emission data and substation operation data corresponding to each carbon emission stage;
[0011] Preprocess the historical carbon emission data to obtain effective historical carbon emission data;
[0012] Use effective historical carbon emission data to build and train a carbon emission prediction model;
[0013] The substation operation data is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction values corresponding to each carbon emission stage of the substation's entire life cycle are obtained.
[0014] In the above method, optionally, the carbon emission stages of the entire life cycle of the substation include: building material production stage, equipment manufacturing stage, transportation stage, construction stage, operation stage, maintenance stage and demolition stage.
[0015] The above method optionally obtains the carbon emission historical data and substation operation data corresponding to each carbon emission stage through smart meters and sensor networks.
[0016] In the above method, optionally, the substation operation data includes: energy consumption data, operation time, load rate, ambient temperature and humidity of various types of power equipment in the substation.
[0017] The above method optionally pre-processes the carbon emission historical data to obtain the specific content of valid carbon emission historical data: cleaning, denoising and normalizing the carbon emission historical data to obtain valid carbon emission historical data; the cleaning process includes eliminating invalid data, filling missing values, and correcting outliers; denoising is to reduce the random fluctuations of carbon emission historical data through filtering methods; normalization is the process of converting carbon emission historical data into the same dimension.
[0018] The above method optionally uses effective carbon emission historical data to construct and train the carbon emission prediction model as follows:
[0019] Determine the structure of the BP neural network based on the effective carbon emission historical data, including the number of nodes in the input layer, hidden layer and output layer, and the connection matrix between each layer;
[0020] Through simulation, determine the optimal BP neural network structure;
[0021] The BP neural network is trained using the gradient descent method or the conjugate gradient method. During the training process, the weights between the layers of the BP neural network are adjusted according to the error back propagation algorithm and the gradient descent method, thereby optimizing the BP neural network and obtaining a carbon emissions prediction model.
[0022] The above method optionally inputs the substation operation data into the trained carbon emission prediction model for prediction, and obtains the specific contents of the carbon emission prediction values corresponding to each carbon emission stage of the substation's entire life cycle:
[0023] The substation operation data corresponding to the production volume of building materials is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction value of the building material production stage is obtained;
[0024] The substation operation data corresponding to the production volume of the production equipment is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction value of the equipment manufacturing stage is obtained;
[0025] The substation operation data corresponding to the transportation mode is input into the trained carbon emission prediction model to make predictions, and the predicted value of carbon emissions in the transportation stage is obtained;
[0026] The substation operation data corresponding to the construction activities is input into the trained carbon emission prediction model to obtain the predicted value of carbon emissions in the construction phase;
[0027] Based on the substation operation data corresponding to the equipment operation, the trained carbon emission prediction model is input for prediction to obtain the predicted value of carbon emissions in the operation stage;
[0028] The substation operation data corresponding to the maintenance activities is input into the trained carbon emission prediction model to make predictions, and the predicted carbon emission value in the maintenance phase is obtained;
[0029] The substation operation data corresponding to the demolition activity is input into the trained carbon emission prediction model for prediction, and the predicted value of carbon emissions in the demolition stage is obtained.
[0030] The above method may optionally further include: verifying the accuracy of the carbon emission prediction model by comparing and analyzing the predicted carbon emission values at each stage with the actual carbon emission values at each stage.
[0031] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a method for predicting carbon emissions of a substation over its entire life cycle, which has the following beneficial effects:
[0032] 1) Using effective carbon emission historical data to build and train a carbon emission prediction model can improve the accuracy of the carbon emission prediction model;
[0033] 2) The technical solution of the present invention can quickly and accurately predict the carbon emissions of substation buildings throughout their life cycle;
[0034] 3) Preprocessing of historical carbon emission data solves the technical problem that only short-term predictions can be achieved due to missing data and the accuracy of prediction data is low, thereby improving the prediction effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0036] Figure 1 A flow chart of a method for predicting carbon emissions of a substation throughout its life cycle provided by the present invention;
[0037] Figure 2 This is a structural diagram of the BP neural network provided by the present invention. DETAILED DESCRIPTION
[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0039] See also Figure 1 As shown, the present invention discloses a method for predicting carbon emissions of a substation throughout its life cycle, comprising the following steps:
[0040] The carbon emissions of the substation throughout its life cycle are divided into stages to obtain the carbon emission stages of the substation throughout its life cycle;
[0041] Obtain historical carbon emission data and substation operation data corresponding to each carbon emission stage;
[0042] Preprocess the historical carbon emission data to obtain effective historical carbon emission data;
[0043] Use effective historical carbon emission data to build and train a carbon emission prediction model;
[0044] The substation operation data is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction values corresponding to each carbon emission stage of the substation's entire life cycle are obtained.
[0045] Furthermore, the carbon emission stages of the entire life cycle of a substation include: building material production stage, equipment manufacturing stage, transportation stage, construction stage, operation stage, maintenance stage, and demolition stage.
[0046] Specifically, before the construction of the substation, the carbon emissions of the entire life cycle of the substation construction are predicted. This can ensure that environmental factors are taken into account during the design stage, so that low-carbon and environmentally friendly building materials and equipment can be selected, transportation methods can be optimized, etc., to reduce carbon emissions from the source.
[0047] Furthermore, the carbon emission historical data and substation operation data corresponding to each carbon emission stage are obtained through smart meters and sensor networks.
[0048] Specifically, ensure the timeliness, accuracy and completeness of data.
[0049] Furthermore, the substation operation data includes: energy consumption data, operation time, load rate, ambient temperature and humidity of various power equipment in the substation.
[0050] Furthermore, the carbon emission historical data is preprocessed to obtain the specific contents of valid carbon emission historical data: the carbon emission historical data is cleaned, denoised and normalized to obtain valid carbon emission historical data; the cleaning process includes eliminating invalid data, filling missing values, and correcting outliers; denoising is to reduce the random fluctuation of carbon emission historical data through filtering methods; normalization is the process of converting carbon emission historical data into the same dimension.
[0051] Specifically, invalid data may be caused by equipment failure, system error, etc., and need to be eliminated or corrected. Missing values need to be filled according to the actual situation, including mean filling, median filling, mode filling, etc. Outliers may be caused by random fluctuations or measurement errors, and need to be corrected or eliminated.
[0052] Filtering methods include moving average filtering, exponential moving average filtering, Savitzky-Golay filtering, etc. These methods can select appropriate filters for denoising according to the characteristics and requirements of the data.
[0053] Normalization methods include minimum-maximum normalization, z-score normalization, etc.
[0054] Furthermore, the specific contents of constructing and training the carbon emission prediction model using effective carbon emission historical data are as follows:
[0055] Determine the structure of the BP neural network based on the effective carbon emission historical data, including the number of nodes in the input layer, hidden layer and output layer, and the connection matrix between each layer;
[0056] See also Figure 2 As shown, the optimal BP neural network structure is determined through simulation;
[0057] The BP neural network is trained using the gradient descent method or the conjugate gradient method. During the training process, the weights between the layers of the BP neural network are adjusted according to the error back propagation algorithm and the gradient descent method, thereby optimizing the BP neural network and obtaining a carbon emissions prediction model.
[0058] Specifically, in the forward propagation process of the error back propagation algorithm, the input information is processed layer by layer from the input layer through the hidden layer and transmitted to the output layer. If the expected output value is not obtained in the output layer, the difference between the output and the expectation is calculated as the loss value and the back propagation process is entered. Back propagation calculates the partial derivative of the objective function with respect to each neuron weight layer by layer to form the gradient of the objective function with respect to the weight vector as the basis for modifying the weight. The learning of the network is completed in the process of modifying the weight.
[0059] Specifically, model parameter optimization is required before training the model. Model parameter optimization is the key to improving prediction accuracy. By adjusting the hyperparameters of the carbon emission prediction model, such as learning rate, number of iterations, number of hidden layer nodes, etc., the performance of the carbon emission prediction model can be further improved. The selection of these hyperparameters is crucial to the training and prediction effect of the carbon emission prediction model. Therefore, it is necessary to use a suitable optimization algorithm to adjust these hyperparameters to find the optimal parameter combination.
[0060] In the process of model parameter optimization, methods such as cross-validation and grid search can be used. Cross-validation can divide the data set into multiple subsets, one part of which is used to train the carbon emission prediction model, and the other part of which is used to verify the carbon emission prediction model. Through multiple cross-validations, indicators such as the average performance and standard deviation of the carbon emission prediction model can be obtained. Grid search is an exhaustive search method that finds the optimal parameter combination by traversing all possible parameter combinations.
[0061] Furthermore, the substation operation data is input into the trained carbon emission prediction model for prediction, and the specific contents of the carbon emission prediction values corresponding to each carbon emission stage of the substation's entire life cycle are obtained as follows:
[0062] The substation operation data corresponding to the production volume of building materials is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction value of the building material production stage is obtained;
[0063] The substation operation data corresponding to the production volume of the production equipment is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction value of the equipment manufacturing stage is obtained;
[0064] The substation operation data corresponding to the transportation mode is input into the trained carbon emission prediction model to make predictions, and the predicted value of carbon emissions in the transportation stage is obtained;
[0065] The substation operation data corresponding to the construction activities is input into the trained carbon emission prediction model to obtain the predicted value of carbon emissions in the construction phase;
[0066] Based on the substation operation data corresponding to the equipment operation, the trained carbon emission prediction model is input for prediction to obtain the predicted value of carbon emissions in the operation stage;
[0067] The substation operation data corresponding to the maintenance activities is input into the trained carbon emission prediction model to make predictions, and the predicted carbon emission value in the maintenance phase is obtained;
[0068] The substation operation data corresponding to the demolition activity is input into the trained carbon emission prediction model for prediction, and the predicted value of carbon emissions in the demolition stage is obtained.
[0069] Specifically, this application divides the various stages in the substation construction process into the building materials production stage, equipment manufacturing stage, transportation stage, construction stage, operation stage, maintenance stage and demolition stage, and predicts the carbon emissions of the substation over its entire life cycle; this application can quickly predict the carbon emissions of the substation building over its entire life cycle, ensuring the low-carbon design and energy conservation and emission reduction of the substation.
[0070] Furthermore, it also includes: verifying the accuracy of the carbon emission prediction model by comparing and analyzing the predicted carbon emission values at each stage with the actual carbon emission values at each stage.
[0071] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0072] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for predicting carbon emissions during the entire life cycle of a substation, characterized in that: The following steps are involved: The carbon emissions of the substation throughout its life cycle are divided into stages to obtain the carbon emission stages of the substation throughout its life cycle; Obtain historical carbon emission data and substation operation data corresponding to each carbon emission stage; Preprocess the historical carbon emission data to obtain effective historical carbon emission data; Use effective historical carbon emission data to build and train a carbon emission prediction model; The substation operation data is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction values corresponding to each carbon emission stage of the substation's entire life cycle are obtained.
2. A method for predicting carbon emissions during the entire life cycle of a substation according to claim 1, characterized in that: The carbon emission stages of the entire life cycle of a substation include: building material production stage, equipment manufacturing stage, transportation stage, construction stage, operation stage, maintenance stage and demolition stage.
3. A method for predicting carbon emissions during the entire life cycle of a substation according to claim 1, characterized in that: The historical carbon emission data and substation operation data corresponding to each carbon emission stage are obtained through smart meters and sensor networks.
4. A method for predicting carbon emissions during the entire life cycle of a substation according to claim 1, characterized in that: Substation operation data includes: energy consumption data, operating time, load rate, ambient temperature and humidity of various power equipment in the substation.
5. The method for predicting carbon emissions during the entire life cycle of a substation according to claim 1, characterized in that: The specific contents of preprocessing the historical carbon emission data to obtain valid historical carbon emission data are as follows: cleaning, denoising and normalizing the historical carbon emission data to obtain valid historical carbon emission data; the cleaning process includes eliminating invalid data, filling missing values and correcting outliers; Denoising is to reduce the random fluctuations of historical carbon emission data through filtering; Normalization is the process of converting historical carbon emission data into the same dimension.
6. A method for predicting carbon emissions during the entire life cycle of a substation according to claim 1, characterized in that: The specific contents of building and training the carbon emission prediction model using effective carbon emission historical data are as follows: Determine the structure of the BP neural network based on the effective carbon emission historical data, including the number of nodes in the input layer, hidden layer and output layer, and the connection matrix between each layer; Through simulation, determine the optimal BP neural network structure; The BP neural network is trained using the gradient descent method or the conjugate gradient method. During the training process, the weights between the layers of the BP neural network are adjusted according to the error back propagation algorithm and the gradient descent method, thereby optimizing the BP neural network and obtaining a carbon emissions prediction model.
7. A method for predicting carbon emissions during the entire life cycle of a substation according to claim 2, characterized in that: The substation operation data is input into the trained carbon emission prediction model for prediction, and the specific contents of the carbon emission prediction values corresponding to each carbon emission stage of the substation's entire life cycle are obtained as follows: The substation operation data corresponding to the production volume of building materials is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction value of the building material production stage is obtained; The substation operation data corresponding to the production volume of the production equipment is input into the trained carbon emission prediction model for prediction, and the carbon emission prediction value of the equipment manufacturing stage is obtained; The substation operation data corresponding to the transportation mode is input into the trained carbon emission prediction model to make predictions, and the predicted value of carbon emissions in the transportation stage is obtained; The substation operation data corresponding to the construction activities is input into the trained carbon emission prediction model to obtain the predicted value of carbon emissions in the construction phase; Based on the substation operation data corresponding to the equipment operation, the trained carbon emission prediction model is input for prediction to obtain the predicted value of carbon emissions in the operation stage; The substation operation data corresponding to the maintenance activities is input into the trained carbon emission prediction model to make predictions, and the predicted carbon emission value in the maintenance phase is obtained; The substation operation data corresponding to the demolition activity is input into the trained carbon emission prediction model for prediction, and the predicted value of carbon emissions in the demolition stage is obtained.
8. The method for predicting carbon emissions during the entire life cycle of a substation according to claim 1, characterized in that: Also includes: By comparing and analyzing the predicted carbon emissions at each stage with the actual carbon emissions at each stage, the accuracy of the carbon emissions prediction model is verified.