Carbon emission prediction method, device, terminal and storage medium

Abstract

The present invention provides a carbon emission prediction method, device, terminal and storage medium. The method includes: obtaining the energy consumption structure of the area to be predicted and the power consumption of the area to be predicted in the time period to be predicted; inputting the power consumption of the area to be predicted in the time period to be predicted into a pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the area to be predicted in the time period to be predicted; the other energy in the area to be predicted includes energy other than electric energy in the energy consumption structure of the area to be predicted; according to the power consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy in the area to be predicted in the time period to be predicted, determine the predicted carbon emissions of the area to be predicted in the time period to be predicted. The present invention can predict the usage of other energy in the energy consumption structure through power consumption, and predict carbon emissions based on power consumption and other energy usage, and can realize carbon emission prediction without collecting historical carbon emissions.

Description

Technical Field

[0001] The present invention relates to the field of carbon emission technology, and in particular to a carbon emission prediction method, device, terminal and storage medium. Background Art

[0002] With the rapid development of the global economy, carbon emissions have also increased significantly. This is because economic growth has increased the demand for energy such as electricity, oil, and natural gas, and the use of fossil energy such as electricity, oil, and natural gas will produce a large amount of carbon emissions. Excessive carbon emissions can easily lead to the destruction of the ozone layer and global warming, which in turn increases the probability of extreme weather. In order to cope with climate change, it is necessary to establish an accurate carbon emissions prediction method.

[0003] At present, the carbon emissions monitoring system is not perfect. Authoritative organizations only publish annual carbon emissions data. Monthly and other higher-resolution carbon emissions data are missing, which brings great difficulties to the short-term prediction of carbon emissions. Summary of the invention

[0004] The embodiments of the present invention provide a carbon emissions prediction method, device, terminal and storage medium to solve the problem in the prior art that short-term carbon emissions prediction tasks are difficult to carry out effectively due to the lack of higher-resolution carbon emissions data such as monthly data.

[0005] In a first aspect, an embodiment of the present invention provides a carbon emission prediction method, comprising:

[0006] Obtaining the energy consumption structure of the area to be predicted and the power consumption of the area to be predicted in the time period to be predicted;

[0007] The power consumption of the area to be predicted in the time period to be predicted is input into the pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the area to be predicted in the time period to be predicted; the other energy in the area to be predicted includes energy other than electric energy in the energy consumption structure of the area to be predicted; the other energy usage prediction model is trained using the historical energy consumption data of the area to be predicted, and the historical energy consumption data includes historical power consumption and historical other energy usage;

[0008] The predicted carbon emissions of the area to be predicted in the time period to be predicted are determined according to the power consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy sources in the area to be predicted in the time period to be predicted.

[0009] In a possible implementation, the predicted carbon emissions of the area to be predicted in the time period to be predicted are determined according to the power consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy sources in the area to be predicted in the time period to be predicted, including:

[0010] Determining a first carbon emission amount according to the electricity consumption of the area to be predicted in the time period to be predicted;

[0011] Determining a second carbon emission amount according to the predicted usage of other energy sources in the area to be predicted during the time period to be predicted;

[0012] The first carbon emission and the second carbon emission are summed to obtain the predicted carbon emission of the area to be predicted in the time period to be predicted.

[0013] In a possible implementation, determining the first carbon emission amount according to the electricity consumption of the area to be predicted in the time period to be predicted includes:

[0014] according to Calculating the first carbon emissions

[0015] Among them, P e is the power consumption of the area to be predicted in the time period to be predicted; n is the number of coal-fired power units in the area to be predicted; P i is the power generation of the i-th coal-fired power unit in the area to be predicted during the time period to be predicted, P i，c，t，s T is the power generation of the i-th coal-fired power unit in the area to be predicted during the s-th power monitoring in the t-th hour in the c-th day in the time period to be predicted, m is the interval between two adjacent power monitorings, and D is the number of days in the time period to be predicted; T i is the raw coal consumption corresponding to the unit electricity produced by the i-th coal-fired power unit in the predicted area; K1 is the conversion coefficient between raw coal and standard coal; K2 is the carbon dioxide emission factor of standard coal.

[0016] In a possible implementation, determining the second carbon emission amount according to the predicted usage of other energy sources in the area to be predicted in the time period to be predicted includes:

[0017] according to Calculating the Second Carbon Emissions

[0018] Among them, f j The other energy sources in the area to be predicted j The predicted usage of the energy in the predicted time period; K2 is the carbon dioxide emission factor of standard coal; K′ i is the standard coal coefficient of the jth energy in other energy sources in the area to be predicted; J is the energy type in other energy sources in the area to be predicted.

[0019] In one possible implementation, the other energy usage prediction model is a Stacking-based prediction model;

[0020] Other energy usage prediction models include a first-layer prediction model using an XGBoost model, a random forest model, a long short-term memory network model, and a time-domain convolutional network model as base learners and a second-layer prediction model using a ridge regression model as a meta-learner.

[0021] In a possible implementation, before inputting the power consumption of the area to be predicted in the time period to be predicted into a pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the area to be predicted in the time period to be predicted, the carbon emission prediction method further includes:

[0022] Obtain historical energy consumption data for the area to be predicted;

[0023] The K-fold cross-validation method is used to train each base learner in other energy usage prediction models based on historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved;

[0024] The prediction results of each base learner for the validation set after each round of training are concatenated, and the concatenated data is used as the training data of the meta-learner;

[0025] The meta-learner is trained according to the training data of the meta-learner to obtain a trained prediction model for other energy usage.

[0026] In a possible implementation, after obtaining the historical energy consumption data of the area to be predicted, the carbon emission prediction method further includes:

[0027] Perform missing value filling and normalization processing on historical energy consumption data to obtain preprocessed historical energy consumption data;

[0028] Accordingly, the K-fold cross-validation method is used to train each base learner in other energy usage prediction models according to historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved, including:

[0029] The K-fold cross-validation method is used to train each base learner in other energy usage prediction models according to the preprocessed historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved.

[0030] In a second aspect, an embodiment of the present invention provides a carbon emission prediction device, comprising:

[0031] An acquisition module is used to acquire the energy consumption structure of the area to be predicted and the power consumption of the area to be predicted in the time period to be predicted;

[0032] The energy prediction module is used to input the power consumption of the area to be predicted in the time period to be predicted into the pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the area to be predicted in the time period to be predicted; the other energy in the area to be predicted includes energy other than electric energy in the energy consumption structure of the area to be predicted; the other energy usage prediction model is trained using the historical energy consumption data of the area to be predicted, and the historical energy consumption data includes historical power consumption and historical other energy usage;

[0033] The carbon emission prediction module is used to determine the predicted carbon emissions of the area to be predicted during the predicted time period based on the electricity consumption of the area to be predicted during the predicted time period and the predicted usage of other energy sources in the area to be predicted during the predicted time period.

[0034] In a third aspect, an embodiment of the present invention provides a terminal comprising a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the carbon emission prediction method as described in the first aspect or any possible implementation of the first aspect.

[0035] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the carbon emission prediction method described in the first aspect or any possible implementation method of the first aspect are implemented.

[0036] The embodiment of the present invention provides a carbon emissions prediction method, device, terminal and storage medium. The method inputs the electricity consumption of the area to be predicted in the time period to be predicted into a pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the energy usage structure of the area to be predicted in the time period to be predicted, and determines the predicted carbon emissions of the area to be predicted in the time period to be predicted based on the electricity consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy in the area to be predicted. The method can predict the usage of other energy in the energy usage structure through electricity consumption, and predict carbon emissions based on electricity consumption and other energy usage. The other energy usage prediction model is trained using historical energy consumption data of the area to be predicted, and the historical energy consumption data includes historical electricity consumption and historical other energy usage. Therefore, when collecting data, it is only necessary to collect electricity consumption and other energy usage data without collecting historical carbon emissions, so as to realize carbon emissions prediction, thereby solving the problem that carbon emissions prediction cannot be performed due to the lack of historical data on carbon emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 is a schematic diagram of a carbon emission prediction method provided by an embodiment of the present invention;

[0039] Figure 2 is a schematic diagram of the evolution relationship of "electricity-energy-carbon" provided by an embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of a Stacking integrated learning framework provided by an embodiment of the present invention;

[0041] Figure 4 Schematic diagram of the LSTM neuron structure provided by an embodiment of the present invention;

[0042] Figure 5 Schematic diagram of the network structure of the TCN model provided by the embodiment of the present invention;

[0043] Figure 6 is a schematic diagram of the structure of the dilated convolution provided by an embodiment of the present invention;

[0044] Figure 7 is a schematic diagram of other energy usage prediction models and training thereof provided by an embodiment of the present invention;

[0045] Figure 8 is a structural schematic diagram of a carbon emission prediction device provided by an embodiment of the present invention;

[0046] Fig. 9 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0048] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0049] See also Figure 1, which shows a flow chart of the implementation of the carbon emission prediction method provided by an embodiment of the present invention. The execution subject of the carbon emission prediction method may be a terminal.

[0050] See also Figure 1 , the above carbon emission prediction method includes:

[0051] In S101, the energy consumption structure of the area to be predicted and the power consumption of the area to be predicted in the time period to be predicted are obtained.

[0052] The energy structure of the area to be predicted is used to characterize the energy used in the area to be predicted, for example, it may include electricity, coal, oil, and natural gas, etc. Since the embodiment of the present application is used to predict carbon emissions, the energy structure of the area to be predicted only considers energy that will cause carbon emissions.

[0053] The area to be predicted can be set as the area whose carbon emissions need to be predicted according to actual needs. The area to be predicted can be a province, a city, a county or a district, etc., and is not specifically limited here. Among them, the embodiment of the present invention can also predict the carbon emissions of a certain industry in a certain area, that is, the energy consumption structure of the industry to be predicted in the area to be predicted and the electricity consumption of the industry to be predicted in the area to be predicted in the time period to be predicted are obtained, and the electricity consumption is used for subsequent prediction, and finally the predicted carbon emissions of the industry to be predicted in the area to be predicted in the prediction time period are obtained.

[0054] The time period to be predicted can also be set according to actual needs, for example, it can be this month, last month, multiple consecutive days, etc.

[0055] The energy consumption structure of the area to be predicted and the electricity consumption of the area to be predicted in the time period to be predicted can be obtained by existing means, and no specific restrictions are made here.

[0056] In S102, the electricity consumption of the area to be predicted in the time period to be predicted is input into a pre-trained prediction model for other energy usage to obtain the predicted usage of other energy in the area to be predicted in the time period to be predicted; the other energy in the area to be predicted includes energy other than electricity in the energy consumption structure of the area to be predicted; the prediction model for other energy usage is trained using historical energy consumption data of the area to be predicted, and the historical energy consumption data includes historical electricity consumption and historical other energy usage.

[0057] This embodiment constructs a mapping relationship between electricity consumption and other energy usage through other energy usage prediction models.

[0058] In this embodiment, other energy sources in the area to be predicted may be energy sources other than electricity in the energy consumption structure of the area to be predicted, specifically, other energy sources other than electricity among all energy sources that generate carbon emissions in the energy consumption structure of the area to be predicted, for example, may include at least one of coal, oil, and natural gas.

[0059] The other energy usage prediction models may be one or more.

[0060] When there is only one other energy usage prediction model, its input is the electricity consumption of the area to be predicted in the time period to be predicted, and its output is the predicted usage of various other energy sources in the area to be predicted in the time period to be predicted, for example, including the predicted usage of coal in the area to be predicted in the time period to be predicted, the predicted usage of oil in the area to be predicted in the time period to be predicted, and the predicted usage of natural gas in the area to be predicted in the time period to be predicted. When training it, the training sample data is the historical energy consumption data of the area to be predicted, which can be expressed as {(x1,y1,),(x2,y2),(x3,y3),…,(x g ,y g )}, where x g is the power consumption of the g-th sample, y g =(y g 1 ,y g 2 ,…,y g h ), represents the other energy usage of the g-th sample, where y g h is the usage of the hth energy among other energy sources of the gth sample.

[0061] When there are multiple other energy usage prediction models, each of the other energy sources in the area to be predicted may correspond to one other energy usage prediction model.

[0062] For each type of other energy in the area to be predicted, the input of the other energy usage prediction model corresponding to this type of energy is the electricity consumption of the area to be predicted in the time period to be predicted, and the output is the predicted usage of this type of energy in the area to be predicted in the time period to be predicted. When the other energy usage prediction model corresponding to this type of energy is trained, the training sample data is the historical energy consumption data of the area to be predicted, which can be expressed as {(x1,y1,),(x2,y2),(x3,y3),…,(x g ,y g )}, where x g is the power consumption of the g-th sample, y g is the energy usage of the g-th sample.

[0063] In this embodiment, the pre-built other energy usage prediction model can be trained by the historical energy consumption data of the area to be predicted to obtain a pre-trained other energy usage prediction model, and the power consumption of the area to be predicted in the time period to be predicted is input into the pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the area to be predicted in the time period to be predicted. The predicted usage of other energy in the area to be predicted in the time period to be predicted includes the predicted usage of various energy sources in the other energy in the area to be predicted in the time period to be predicted.

[0064] In S103, the predicted carbon emissions of the area to be predicted in the time period to be predicted are determined according to the power consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy sources in the area to be predicted in the time period to be predicted.

[0065] In this embodiment, the electricity consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy in the area to be predicted in the time period to be predicted can be converted into carbon emissions, thereby obtaining the predicted carbon emissions of the area to be predicted in the time period to be predicted.

[0066] This embodiment inputs the electricity consumption of the area to be predicted in the time period to be predicted into a pre-trained prediction model for other energy usage, thereby obtaining the predicted usage of other energy in the energy usage structure of the area to be predicted in the time period to be predicted, and determines the predicted carbon emissions of the area to be predicted in the time period to be predicted based on the electricity consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy in the area to be predicted. This enables prediction of other energy usage in the energy usage structure through electricity consumption, and prediction of carbon emissions based on electricity consumption and other energy usage. The other energy usage prediction model is trained using historical energy consumption data of the area to be predicted, and the historical energy consumption data includes historical electricity consumption and historical other energy usage. Therefore, when collecting data, it is only necessary to collect electricity consumption and other energy usage data, without collecting historical carbon emissions, to achieve carbon emissions prediction, thereby solving the problem of the inability to predict carbon emissions due to the difficulty in collecting carbon emissions data and the lack of historical carbon emissions data.

[0067] In some embodiments, the above S103 may include:

[0068] Determining a first carbon emission amount according to the electricity consumption of the area to be predicted in the time period to be predicted;

[0069] Determining a second carbon emission amount according to the predicted usage of other energy sources in the area to be predicted during the time period to be predicted;

[0070] The first carbon emission and the second carbon emission are summed to obtain the predicted carbon emission of the area to be predicted in the time period to be predicted.

[0071] Among them, the first carbon emission is the carbon emission corresponding to the electricity consumption of the predicted area in the time period to be predicted. The second carbon emission is the sum of the carbon emissions corresponding to the predicted usage of various energy sources in the predicted area in the time period to be predicted. The sum of the first carbon emission and the second carbon emission is the predicted carbon emission of the predicted area in the time period to be predicted.

[0072] In some embodiments, determining the first carbon emissions according to the electricity consumption of the area to be predicted in the time period to be predicted includes:

[0073] according to Calculating the first carbon emissions

[0074] Among them, P e is the power consumption of the area to be predicted in the time period to be predicted; n is the number of coal-fired power units in the area to be predicted; P i is the power generation of the i-th coal-fired power unit in the area to be predicted during the time period to be predicted, P i，c，t，s is the power generation of the fth coal-fired power unit in the area to be predicted during the sth power monitoring in the tth hour of the cth day in the time period to be predicted; m is the interval between two adjacent power monitorings, which represents the sampling frequency of the coal-fired power unit output data; D is the number of days in the time period to be predicted; T i is the raw coal consumption corresponding to the unit electricity produced by the i-th coal-fired power unit in the predicted area; K1 is the conversion coefficient between raw coal and standard coal; K2 is the carbon dioxide emission factor of standard coal, which is 2.66 tons CO2 / ton of standard coal.

[0075] In this embodiment, the carbon emissions generated by coal-fired power generation are considered to be carbon emissions from the power production and supply industry, and the energy source is coal. Therefore, the control center can collect statistics to obtain the details of the coal-fired power units in the area to be predicted, and the output of each coal-fired power unit can be monitored in real time to obtain the power generation of each coal-fired power unit in the time period to be predicted, and the coal consumption per unit of electricity produced by each coal-fired power unit is counted and updated monthly. According to the power generation and coal consumption per unit of electricity of each coal-fired power unit in the time period to be predicted, the coal consumption of each coal-fired power unit is converted, and then the carbon emissions of electricity produced by each coal-fired power unit are calculated through coal consumption, the conversion coefficient of raw coal and standard coal, and the carbon dioxide emission molecule of standard coal.

[0076] In some embodiments, the second carbon emission amount is determined according to the predicted usage of other energy sources in the area to be predicted in the time period to be predicted, including:

[0077] according to Calculating the Second Carbon Emissions

[0078] Among them, f j is the predicted usage of the jth energy in the predicted area during the predicted period; K2 is the carbon dioxide emission factor of standard coal; K′ i is the standard coal coefficient of the jth energy in other energy sources in the area to be predicted; J is the energy type in other energy sources in the area to be predicted.

[0079] Since carbon emissions mainly come from electricity production and the use of fossil energy, the embodiment of the present invention adopts the method of "electricity-carbon conversion" to convert carbon emissions, that is, by constructing the evolution relationship of "electricity-energy-carbon" (see Figure 2 ) to predict carbon emissions. Specifically, it is to count the use of electricity and fossil energy such as coal, oil, and natural gas, and convert them into carbon emissions according to the corresponding formula.

[0080] In some embodiments, the other energy usage prediction model is a Stacking-based prediction model;

[0081] Other energy usage prediction models include a first-layer prediction model using an XGBoost model, a random forest model, a long short-term memory network (LSTM) model, and a temporal convolutional network (TCN) model as base learners and a second-layer prediction model using a ridge regression model as a meta-learner.

[0082] The stacking ensemble learning framework generally consists of two layers, such as Figure 3 As shown, the first-layer prediction model includes multiple base learners, and the original data set is processed to obtain several data sets, which are respectively used as training sets for each base learner in the first-layer prediction model. Each base learner outputs its own prediction result, which will be used as the input of the second-layer prediction model. The meta-learner of the second-layer prediction model is trained, and the meta-learner outputs the final prediction result. The Stacking prediction model generalizes the prediction results of multiple base learners through the meta-learner to obtain higher prediction accuracy. In order to achieve high-precision prediction, the performance of the selected base learners is required to be as good as possible and the difference between the base learners is large. Therefore, the embodiment of the present invention is based on the Stacking ensemble learning idea, selects XGBoost, LSTM, random forest, and TCN as base learners to construct the first-layer prediction model, and selects the ridge regression model as the meta-learner to construct the second-layer prediction model.

[0083] Using the XGBoost model as the base learner, XGboost optimizes the boosting algorithm. Its main task is to integrate weak classifiers into a strong classifier. The XGBoost algorithm generates a new tree to fit the residual of the previous tree through continuous iteration. As the number of iterations increases, the accuracy continues to improve. The tree model used by XGBoost is the CART (classification and regression tree) model, which is represented as follows:

[0084]

[0085] f u (x) = w q(x)

[0086] in, is the predicted value of the model, N is the number of trees, F is the set of N trees, x i represents the i-th data input, w q(x) represents the weight of each leaf node corresponding to the u-th tree, and q(x) represents the sequence number of the output leaf node.

[0087] The objective function of XGBoost is as follows:

[0088]

[0089]

[0090] Where: is a measure of the true value y i and predicted values The loss function of is a regularization term, which represents the sum of the complexity of all trees and is used to control the complexity of the model to prevent overfitting. t ), where T represents the number of leaf nodes, represents the square of the leaf node weight; μ and ν correspond to the coefficients of the penalty term.

[0091] The objective function of XGBoost can also be expressed as follows:

[0092]

[0093] In f t = 0, the objective function is approximated by the second-order Taylor expansion, and the objective function can be expressed as follows:

[0094]

[0095] Where: represents the loss function of the learning model composed of the first t-1 trees; g i and h i They represent the first-order and second-order derivatives of the loss function for the current model. Expanding the regularization term yields the following formula:

[0096]

[0097] In the above formula: I j ={i|q(x i )=j}, which means the sample set on the leaf node with sequence number j. And it can be known that when The optimal solution is obtained when The loss function is the minimization objective function of the XGBoost model, which can be used to judge the quality of the XGBoost model. The smaller the loss function, the better the model training.

[0098] We can enumerate all possibilities to build a CART tree with the minimum objective function, and then compare the structure scores to obtain the optimal tree structure, but this method requires a large amount of calculation. In order to reduce the amount of calculation and ensure accuracy to a certain extent, the greedy algorithm is often used to simplify the above process. That is, only the current node is considered optimal, and each attempt is made to split the current node, and the segmentation gain C g The calculation expression is as follows:

[0099]

[0100] in, Respectively represent the gains generated by splitting the left and right subtrees, Indicates the gain of not splitting the subtree. The split gain can be used as the split gain, and the method with the largest split gain is finally selected as the split point of the CART tree.

[0101] The XGBoost model improves the model's fitting ability through the boosting method. Compared with other models, it has better results in calculation speed and prediction accuracy.

[0102] The random forest model is used as the base learner. The random forest model is an integrated model composed of the CART tree model, the bagging method and the random feature subspace.

[0103] In random forest, bagging method is to select a part of the sample data set with replacement as the basis for building a CART tree. The remaining samples that are not selected are called out-of-bag samples. These samples are used as validation samples to evaluate the model internally. The final model prediction result is based on the voting method, taking the average of all tree model prediction results as the prediction result of the random forest regression model.

[0104] In random forest, when each decision regression tree branches, the optimal feature is selected from the subspace of the total feature set for branching. This method ensures the independence and diversity of each decision tree, thereby avoiding overfitting to a certain extent. The random forest model effectively improves its generalization ability through bagging methods and random feature subspaces.

[0105] The training process of the random forest model is as follows: the random forest algorithm uses the bootstrap method to randomly extract n samples from the original training set with replacement and construct n decision trees; assuming that there are m features in the training sample data, the best feature is selected for each split, and each tree continues to split in this way until all training samples of the node belong to the same category; then each decision tree is allowed to grow to the maximum without any pruning; finally, the generated multiple classification trees are combined into a random forest, and the final prediction result is determined by the mean of the predicted values ​​of multiple trees.

[0106] Taking the LSTM model as the base learner, the LSTM neuron structure is as follows: Figure 4 As shown in the figure, the LSTM model consists of a forget gate, an input gate, and an output gate. The key point of the model is the transmission of the cell state. t-1 Transmission is via a horizontal line similar to a conveyor belt.

[0107] (1) Forget Gate

[0108] The function of the forget gate is to determine the forgotten part of the memory unit. First, the output h of the network at the previous moment is t-1 With the current input x t Splicing, and then input into the sigmoid module, the sigmoid function formula is as follows:

[0109]

[0110] The sigmoid module processes the data of the entire sequence into the (0,1) interval, and then compares the activated sequence with s t-1 The corresponding elements in are multiplied point by point to complete the deletion and update of data. If a data in the sequence after Sigmoid activation is close to 1, it means that the state s at the previous moment t-1 The information at the corresponding position is approximately preserved completely. If it is close to 0, it means that the information is approximately deleted. The working principle of the forget gate can be expressed by the following formula.

[0111] f t =σ(W f [h t-1 ,x t ]+bf )

[0112] Among them, σ represents the sigmoid function, W f Represents the weight coefficient, which needs to be obtained through training. t-1 ,x t ] is h t-1 By and x t The new vector obtained by concatenation, b f Represents the bias coefficient of the forget gate.

[0113] (2) Input Gate

[0114] The function of the input gate is to determine the retention vector in the memory unit. This part is divided into two steps. The first step is to select the information i that needs to be updated through the sigmoid module. t And generate the updated content g through the tanh module t The second step is to multiply the corresponding positions of the two parts point by point to complete the information update. The formula is as follows:

[0115] i t =σ(W i [h t-1 ,x t ]+b i )

[0116] g t =tanh(W g [h t-1 ,x t ]+b g )

[0117] Among them, i t Indicates updated information, g t represents the updated content, σ represents the sigmoid function, tanh represents the tanh function, W i and W g Respectively represent the weights of the corresponding inputs, b i and b g They represent the bias coefficients of the corresponding functions respectively.

[0118] (3) Output gate

[0119] The function of the output gate is mainly to determine the output of the network. Figure 4 The position of the output gate is indicated in . The output h at the previous moment t-1 and the current input x t After being processed by the sigmoid module, we get o t , and then update the input gate to get s tInput to the tanh module for adjustment, and then adjust it with o t Multiply point by point to get the output h at the current moment t The formula corresponding to the output gate is as follows:

[0120] o t =σ(W o [h t-1 ,x t ]+b o )

[0121] s t =g t ⊙i t +s t-1 ⊙f t

[0122] h t =tanh(s t )⊙o t

[0123] Among them, W o represents the weight of the output gate, b o represents the bias of the output gate, ⊙ represents the bitwise multiplication of the elements in the vector, and s t and t-1 Respectively represent the state of the state unit at the previous moment and the current moment.

[0124] The LSTM model can better fit the time series characteristics of the data, thereby improving the accuracy of the prediction model.

[0125] The TCN model is used as the base learner. The TCN model consists of causal convolution, dilated convolution and residual modules. The network structure of the TCN model is shown in the figure below. Figure 5 As shown in the figure, Dropout means that during the process of neuron propagation, the activation value of a certain neuron stops working with a certain probability, so as to avoid overfitting of the model and enhance the generalization ability of the model; Relu means the linear rectification function, which is used as the activation function of the neural network; Weight Norm is used to normalize the weight; Dilated CasualConv means the dilated convolution module. The structural diagram of the dilated convolution is shown in the figure. Figure 6 shown. Figure 6 The specific structure of the dilated convolution is shown when the dilation coefficient d is [1, 2, 4, 8]. It can be intuitively found that the field of view of each neuron in the dilated convolution increases, and as the number of layers increases, the field of view of the neuron increases rapidly.

[0126] Causal convolution is suitable for extracting sequence information. Compared with traditional CNN and LSTM models, dilated convolution is better at memorizing longer historical data and has stronger memory capacity. The residual module can effectively improve the accuracy of the prediction model. Each TCN layer contains L convolution layers, and the calculation formula for dilated convolution is:

[0127]

[0128] Among them, d represents the dilation coefficient. For multi-layer convolution, the dilation coefficient of the nth layer is generally taken as 2 n-1 . In addition, the residual module can effectively avoid the problem of deep neural network degradation. The following two formulas represent the activation function of TCN:

[0129]

[0130]

[0131] Where W (1) , W (2) is the weight matrix corresponding to the input, b is the bias vector, S (i,j) represents the activation function of the i-th layer of the j-th block, the formula It is the result of adding the residual to the dilated convolution at time t.

[0132] TCN can change the receptive field of view by increasing the number of layers, changing the expansion coefficient and the size of the filter. It is more flexible in extracting historical information. The existence of the residual module enables TCN to avoid gradient diffusion and gradient explosion during training. It can more effectively extract high-dimensional time series features. At the same time, it takes up less memory for training long sequences.

[0133] Excessive model complexity will cause deep overfitting. Since the complexity of the base learner model in the embodiment of the present invention is relatively high, in order to avoid the complexity of the overall prediction model being too high, the meta-learner of the second-layer prediction model in the embodiment of the present invention selects the ridge regression model. Ridge regression is the most commonly used regularization method for regression analysis of ill-posed problems. It is a biased estimation regression specifically used for collinear data analysis. It is essentially an improved least squares estimation method. By abandoning the unbiasedness of the least squares method, the regression coefficient is obtained at the cost of losing some information and reducing precision. A more practical and reliable regression method can effectively avoid overfitting of the model. The ridge regression layer learns the input meta-features and outputs the final prediction results.

[0134] According to the above prediction model framework, the embodiment of the present invention can construct models for predicting the use of fossil energy such as oil, natural gas, and coal based on electricity consumption. After inputting electricity consumption into each prediction model, the prediction results of the use of other fossil energy are obtained and converted into carbon emissions respectively.

[0135] In some embodiments, before the above S102, the carbon emission prediction method may further include:

[0136] Obtain historical energy consumption data for the area to be predicted;

[0137] The K-fold cross-validation method is used to train each base learner in other energy usage prediction models based on historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved;

[0138] The prediction results of each base learner for the validation set after each round of training are concatenated, and the concatenated data is used as the training data of the meta-learner;

[0139] The meta-learner is trained according to the training data of the meta-learner to obtain a trained prediction model for other energy usage.

[0140] The embodiment of the present invention trains the four types of base learners separately based on the K-fold cross-validation method. Referring to FIG7 , the details are as follows: the historical energy consumption data is divided into 10 equal parts, and a total of 10 rounds of training are performed. In each round of training, 1 part is selected in sequence as a validation set, and the other 9 parts are selected as training sets. Each type of base learner is trained using the training set. After each round of training, the corresponding XGBoost, random forest, LSTM, and TCN models are obtained. The prediction results of these four models for the validation set are saved. The prediction results of the general base learner for the validation set are called meta-features, and the meta-features of 10 rounds of training are spliced ​​as the training data for the second-layer meta-learner. The reason why this model uses the validation set prediction results is to avoid repeated learning of the same data by similar base learners and overfitting of the meta-learner.

[0141] In some embodiments, after obtaining the historical energy consumption data of the area to be predicted, the carbon emissions prediction method further includes:

[0142] Perform missing value filling and normalization processing on historical energy consumption data to obtain preprocessed historical energy consumption data;

[0143] Accordingly, the K-fold cross-validation method is used to train each base learner in other energy usage prediction models according to historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved, including:

[0144] The K-fold cross-validation method is used to train each base learner in other energy usage prediction models according to the preprocessed historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved.

[0145] In the embodiment of the present invention, firstly, missing values ​​are filled in the historical energy consumption data. The missing values ​​are filled using linear interpolation. Taking the power consumption as an example, the formula is as follows:

[0146]

[0147] Among them, x n Indicates the power consumption of the nth sample, which is the default value; x n-1 represents the power consumption of the n-1th sample, x n+1 Indicates the power consumption of the n+1th sample.

[0148] The filled historical energy consumption data is normalized, and the normalization formula is as follows:

[0149]

[0150] in, is the normalized data; x represents a certain type of energy usage data in the original data; x min Represents the minimum value in the original data, x max It represents the maximum value in the original data.

[0151] It should be understood that the order of execution of the steps in the above embodiments does not imply a precedence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0152] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0153] Figure 8 The structural schematic diagram of the carbon emission prediction device provided by the embodiment of the present invention is shown. For the convenience of explanation, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0154] like Figure 8 As shown, the carbon emission prediction device 30 includes: an acquisition module 31 , an energy prediction module 32 and a carbon emission prediction module 33 .

[0155] An acquisition module 31 is used to acquire the energy consumption structure of the area to be predicted and the power consumption of the area to be predicted in the time period to be predicted;

[0156] The energy prediction module 32 is used to input the power consumption of the area to be predicted in the time period to be predicted into the pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the area to be predicted in the time period to be predicted; the other energy in the area to be predicted includes energy other than electric energy in the energy consumption structure of the area to be predicted; the other energy usage prediction model is trained using the historical energy consumption data of the area to be predicted, and the historical energy consumption data includes historical power consumption and historical other energy usage;

[0157] The carbon emission prediction module 33 is used to determine the predicted carbon emission of the area to be predicted in the time period to be predicted based on the electricity consumption of the area to be predicted in the time period to be predicted and the predicted usage of other energy sources in the area to be predicted in the time period to be predicted.

[0158] In a possible implementation, the carbon emission prediction module 33 is specifically used to:

[0159] Determining a first carbon emission amount according to the electricity consumption of the area to be predicted in the time period to be predicted;

[0160] Determining a second carbon emission amount according to the predicted usage of other energy sources in the area to be predicted during the time period to be predicted;

[0161] The first carbon emission and the second carbon emission are summed to obtain the predicted carbon emission of the area to be predicted in the time period to be predicted.

[0162] In a possible implementation, the carbon emission prediction module 33 is specifically used to:

[0163] according to Calculating the first carbon emissions

[0164] Among them, P e is the power consumption of the area to be predicted in the time period to be predicted; n is the number of coal-fired power units in the area to be predicted; P i is the power generation of the i-th coal-fired power unit in the area to be predicted during the time period to be predicted, P i，c，t，s T is the power generation of the i-th coal-fired power unit in the area to be predicted during the s-th power monitoring in the t-th hour in the c-th day in the time period to be predicted, m is the interval between two adjacent power monitorings, and D is the number of days in the time period to be predicted; T i is the raw coal consumption corresponding to the unit electricity produced by the i-th coal-fired power unit in the predicted area; K1 is the conversion coefficient between raw coal and standard coal; K2 is the carbon dioxide emission factor of standard coal.

[0165] In a possible implementation, the carbon emission prediction module 33 is specifically used to:

[0166] according to Calculating the Second Carbon Emissions

[0167] Among them, f j is the predicted usage of the jth energy in the predicted area during the predicted period; K2 is the carbon dioxide emission factor of standard coal; K′ j is the standard coal coefficient of the jth energy in other energy sources in the area to be predicted; J is the energy type in other energy sources in the area to be predicted.

[0168] In one possible implementation, the other energy usage prediction model is a Stacking-based prediction model;

[0169] Other energy usage prediction models include a first-layer prediction model using an XGBoost model, a random forest model, a long short-term memory network model, and a time-domain convolutional network model as base learners and a second-layer prediction model using a ridge regression model as a meta-learner.

[0170] In a possible implementation, the carbon emission prediction device 30 further includes: a training module.

[0171] The training module is used to:

[0172] Obtain historical energy consumption data for the area to be predicted;

[0173] The K-fold cross-validation method is used to train each base learner in other energy usage prediction models based on historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved;

[0174] The prediction results of each base learner for the validation set after each round of training are concatenated, and the concatenated data is used as the training data of the meta-learner;

[0175] The meta-learner is trained according to the training data of the meta-learner to obtain a trained prediction model for other energy usage.

[0176] In a possible implementation, the carbon emission prediction device 30 further includes: a preprocessing module.

[0177] The preprocessing module is used to:

[0178] Perform missing value filling and normalization processing on historical energy consumption data to obtain preprocessed historical energy consumption data;

[0179] Accordingly, the training module is specifically used for:

[0180] The K-fold cross-validation method is used to train each base learner in other energy usage prediction models according to the preprocessed historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved.

[0181] Fig. 9 is a schematic diagram of a terminal provided by an embodiment of the present invention. Fig. 9 As shown, the terminal 4 of this embodiment includes: a processor 40 and a memory 41. The memory 41 is used to store a computer program 42, and the processor 40 is used to call and run the computer program 42 stored in the memory 41 to perform the steps in the above-mentioned various carbon emission prediction method embodiments, such as Figure 1 Alternatively, the processor 40 is used to call and run the computer program 42 stored in the memory 41 to implement the functions of each module / unit in the above-mentioned device embodiments, for example Figure 8 Functions of the modules / units 31 to 33 are shown.

[0182] Exemplarily, the computer program 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program 42 in the terminal 4. For example, the computer program 42 may be divided into Figure 8 Modules / units 31 to 33 are shown.

[0183] The terminal 4 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will appreciate that Fig. 9 It is only an example of terminal 4 and does not constitute a limitation on terminal 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0184] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0185] The memory 41 may be an internal storage unit of the terminal 4, such as a hard disk or memory of the terminal 4. The memory 41 may also be an external storage device of the terminal 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the terminal 4. The memory 41 is used to store the computer program and other programs and data required by the terminal. The memory 41 may also be used to temporarily store data that has been output or is to be output.

[0186] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit, and the above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0187] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0188] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0189] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0190] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0191] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0192] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned carbon emission prediction method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable media does not include electrical carrier signals and telecommunication signals.

[0193] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A carbon emissions prediction method, characterized in that: include: Obtaining the energy consumption structure of the area to be predicted and the power consumption of the area to be predicted in the time period to be predicted; the power consumption includes electric energy and power generated by other energy sources; The amount of electricity generated by other energy sources in the area to be predicted during the time period to be predicted is input into a pre-trained other energy usage prediction model to obtain the predicted usage of other energy in the area to be predicted during the time period to be predicted; the other energy in the area to be predicted includes energy other than electric energy in the energy usage structure of the area to be predicted; the other energy usage prediction model is trained using historical energy usage data of the area to be predicted, and the historical energy usage data includes historical electricity generated by other energy sources and historical other energy usage; the other energy usage prediction model is a prediction model based on Stacking; the other energy usage prediction model includes a first-layer prediction model using an XGBoost model, a random forest model, a long short-term memory network model, and a time domain convolutional network model as base learners and a second-layer prediction model using a ridge regression model as a meta-learner; Determine the predicted carbon emissions of the area to be predicted in the period to be predicted according to the electric energy of the area to be predicted in the period to be predicted and the predicted usage of other energy sources in the area to be predicted in the period to be predicted; The step of determining the predicted carbon emissions of the area to be predicted in the time period to be predicted according to the electric energy of the area to be predicted in the time period to be predicted and the predicted usage of other energy sources in the area to be predicted in the time period to be predicted includes: Determining a first carbon emission amount according to the electric energy of the area to be predicted in the time period to be predicted; Determining a second carbon emission amount according to a predicted usage of other energy sources in the area to be predicted during the time period to be predicted; The first carbon emission and the second carbon emission are summed to obtain the predicted carbon emission of the to-be-predicted area in the to-be-predicted time period; The determining the first carbon emission amount according to the electric energy of the area to be predicted in the time period to be predicted includes: according to Calculating the first carbon emissions Among them, P e is the electric energy of the area to be predicted in the time period to be predicted; n is the number of coal-fired power units in the area to be predicted; P i is the power generation of the i-th coal-fired power unit in the area to be predicted during the time period to be predicted, P i,c,t,s is the power generation of the i-th coal-fired power unit in the area to be predicted during the s-th power monitoring in the t-th hour in the c-th day in the time period to be predicted, m is the interval between two adjacent power monitorings, and D is the number of days in the time period to be predicted; T i is the raw coal consumption corresponding to the unit electricity produced by the i-th coal-fired power unit in the predicted area; K1 is the conversion coefficient between raw coal and standard coal; K2 is the carbon dioxide emission factor of standard coal.

2. The carbon emission prediction method according to claim 1, characterized in that: The determining the second carbon emission amount according to the predicted usage of other energy sources in the area to be predicted in the time period to be predicted includes: according to Calculating the Second Carbon Emissions Among them, f j is the predicted usage of the jth energy in the other energy sources in the area to be predicted during the time period to be predicted; K2 is the carbon dioxide emission factor of standard coal; K j ′ is the standard coal coefficient of the jth energy in other energy sources in the area to be predicted; J is the energy type in other energy sources in the area to be predicted.

3. The carbon emission prediction method according to claim 1, characterized in that: Before inputting the power generated by other energy sources in the area to be predicted in the time period to be predicted into a pre-trained other energy usage prediction model to obtain the predicted usage of other energy sources in the area to be predicted in the time period to be predicted, the carbon emission prediction method further includes: Obtaining historical energy consumption data of the area to be predicted; Using the K-fold cross-validation method, each base learner in the other energy usage prediction model is trained according to the historical energy usage data, and the prediction results of each base learner for the validation set after each round of training are saved; The prediction results of each base learner for the validation set after each round of training are concatenated, and the concatenated data is used as the training data of the meta-learner; The meta-learner is trained according to the training data of the meta-learner to obtain a trained prediction model for other energy usage.

4. The carbon emission prediction method according to claim 3, characterized in that: After acquiring the historical energy consumption data of the area to be predicted, the carbon emission prediction method further includes: Performing missing value filling and normalization processing on the historical energy consumption data to obtain preprocessed historical energy consumption data; Accordingly, the K-fold cross validation method is used to train each base learner in the other energy usage prediction model according to the historical energy usage data, and the prediction results of each base learner for the validation set after each round of training are saved, including: The K-fold cross validation method is adopted to train each base learner in the other energy usage prediction model according to the preprocessed historical energy consumption data, and the prediction results of each base learner for the validation set after each round of training are saved.

5. A carbon emission prediction device, characterized in that: include: An acquisition module, used to acquire the energy consumption structure of the area to be predicted and the power consumption of the area to be predicted in the time period to be predicted; the power consumption includes electric energy and power generated by other energy sources; An energy prediction module, used for inputting the amount of electricity generated by other energy sources in the area to be predicted during the time period to be predicted into a pre-trained other energy usage prediction model, and obtaining the predicted usage of other energy sources in the area to be predicted during the time period to be predicted; the other energy sources in the area to be predicted include energy other than electric energy in the energy usage structure of the area to be predicted; the other energy usage prediction model is trained using the historical energy usage data of the area to be predicted, and the historical energy usage data includes the amount of electricity generated by other energy sources in history and the amount of other energy usage in history; the other energy usage prediction model is a prediction model based on Stacking; the other energy usage prediction model includes a first-layer prediction model using an XGBoost model, a random forest model, a long short-term memory network model, and a time-domain convolutional network model as base learners, and a second-layer prediction model using a ridge regression model as a meta-learner; A carbon emission prediction module, used to determine the predicted carbon emission of the area to be predicted in the time period to be predicted based on the electric energy of the area to be predicted in the time period to be predicted and the predicted usage of other energy sources in the area to be predicted in the time period to be predicted; The carbon emission prediction module is specifically used for: Determining a first carbon emission amount according to the electric energy of the area to be predicted in the time period to be predicted; Determining a second carbon emission amount according to a predicted usage of other energy sources in the area to be predicted during the time period to be predicted; The first carbon emission and the second carbon emission are summed to obtain the predicted carbon emission of the to-be-predicted area in the to-be-predicted time period; In the carbon emission prediction module, determining the first carbon emission amount according to the electric energy of the area to be predicted in the time period to be predicted includes: according to Calculating the first carbon emissions Among them, P e is the electric energy of the area to be predicted in the time period to be predicted; n is the number of coal-fired power units in the area to be predicted; P i is the power generation of the i-th coal-fired power unit in the area to be predicted during the time period to be predicted, P i,c,t,s is the power generation of the i-th coal-fired power unit in the area to be predicted during the s-th power monitoring in the t-th hour in the c-th day in the time period to be predicted, m is the interval between two adjacent power monitorings, and D is the number of days in the time period to be predicted; T i is the raw coal consumption corresponding to the unit electricity produced by the i-th coal-fired power unit in the predicted area; K1 is the conversion coefficient between raw coal and standard coal; K2 is the carbon dioxide emission factor of standard coal.

6. A terminal, characterized in that: It comprises a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the carbon emission prediction method according to any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the carbon emission prediction method as described in any one of claims 1 to 4 are implemented.

Images