Carbon dioxide concentration prediction method, device, equipment and medium

Abstract

The present application discloses a method, device, equipment and medium for predicting carbon dioxide concentration, the method comprising: obtaining first carbon emission data corresponding to a first model boiler and second carbon emission data corresponding to a second model boiler; performing feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set and a second carbon emission sample set; calculating correction optimization parameters between the first carbon emission sample set and the second carbon emission sample set; constructing a carbon dioxide concentration prediction model based on a training data set and correction optimization parameters corresponding to the first carbon emission device; predicting the carbon dioxide concentration emitted by the second carbon emission device based on a data set to be predicted, correction optimization parameters and a carbon dioxide concentration prediction model corresponding to the second carbon emission device. The present application solves the technical problem in the prior art that the prediction accuracy of carbon dioxide concentration is low due to the model differences of carbon emission devices.

Description

Technical Field

[0001] The present application relates to the field of machine learning technology, and in particular to a method, device, equipment and medium for predicting carbon dioxide concentration. Background Art

[0002] An effective way to achieve carbon neutrality is carbon trading, and the premise of carbon trading is accurate carbon monitoring. For carbon emission equipment that is not equipped with carbon monitoring equipment, existing data can be used to predict the concentration of carbon dioxide emitted by the carbon emission equipment, so it is necessary to accurately establish a prediction model to predict the concentration of carbon dioxide. However, for different types of carbon emission equipment, the data distribution of carbon emission data may change greatly, so the generalization of the carbon dioxide concentration prediction model directly constructed based on randomly collected training samples is poor, which will lead to a large difference in data distribution between the training sample set used in the training process and the prediction sample set in the prediction process, affecting the accuracy of carbon dioxide concentration prediction. Summary of the invention

[0003] The main purpose of this application is to provide a method, device, equipment and medium for predicting carbon dioxide concentration, aiming to solve the technical problem of low accuracy in carbon dioxide concentration prediction due to model differences in carbon emission equipment.

[0004] To achieve the above object, the present application provides a method for predicting carbon dioxide concentration, which comprises:

[0005] Acquire first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the first carbon emission device and the second carbon emission device have different models;

[0006] Performing feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data;

[0007] Calculating a correction optimization parameter between the first carbon emission sample set and the second carbon emission sample set, wherein the correction optimization parameter is used to correct the data distribution of the carbon emission data generated by the first carbon emission device and the data distribution of the carbon emission data generated by the second carbon emission device to be consistent;

[0008] Acquire a training data set corresponding to the first carbon emission device, and construct a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters;

[0009] The to-be-predicted data set corresponding to the second carbon emission device is obtained, and the carbon dioxide concentration emitted by the second carbon emission device is predicted according to the to-be-predicted data set, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

[0010] The present application also provides a device for predicting carbon dioxide concentration, the device comprising:

[0011] An acquisition module, configured to acquire first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the first carbon emission device and the second carbon emission device have different models;

[0012] Performing feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data;

[0013] A correction parameter determination module, used to calculate a correction optimization parameter between the first carbon emission sample set and the second carbon emission sample set;

[0014] A model building module, used to obtain a training data set corresponding to the first carbon emission device, and build a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters;

[0015] The prediction module is used to obtain the data set to be predicted corresponding to the second carbon emission device, and predict the carbon dioxide concentration emitted by the second carbon emission device according to the data set to be predicted, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

[0016] The present application also provides an electronic device, comprising: a memory, a processor, and a program of the carbon dioxide concentration prediction method stored in the memory and executable on the processor. When the program of the carbon dioxide concentration prediction method is executed by the processor, the steps of the carbon dioxide concentration prediction method as described above can be implemented.

[0017] The present application also provides a computer-readable storage medium, on which is stored a program for implementing a method for predicting carbon dioxide concentration. When the program for the method for predicting carbon dioxide concentration is executed by a processor, the steps of the method for predicting carbon dioxide concentration as described above are implemented.

[0018] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned carbon dioxide concentration prediction method when executed by a processor.

[0019] The present application provides a method, device, equipment and medium for predicting carbon dioxide concentration. Compared with the technical means used in the prior art to predict carbon dioxide concentration by directly constructing a carbon dioxide concentration prediction model based on randomly collected training samples, the present application first obtains first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the models of the first carbon emission device and the second carbon emission device are different; performs feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data; calculates correction optimization parameters between the first carbon emission sample set and the second carbon emission sample set, wherein the correction optimization parameters are used to correct the data distribution of the carbon emission data generated by the first carbon emission device and the data distribution of the carbon emission data generated by the second carbon emission device to be consistent; obtains a training data set corresponding to the first carbon emission device, and constructs a carbon dioxide concentration prediction model based on the training data set and the correction optimization parameters. Prediction model, that is, according to the correction optimization parameters, after the training data set is corrected to the preset data distribution, the corresponding carbon dioxide concentration prediction model is constructed; the to-be-predicted data set corresponding to the second carbon emission device is obtained, and the carbon dioxide concentration emitted by the second carbon emission device is predicted according to the to-be-predicted data set, the correction optimization parameters and the carbon dioxide concentration prediction model, to obtain the carbon dioxide concentration prediction result, that is, after the to-be-predicted data set is corrected to the preset data distribution, it is input into the carbon dioxide concentration prediction model for prediction, so that the training sample set in the training process and the prediction sample set in the prediction process are consistent or similar in data distribution, so it is avoided that for different models of carbon emission equipment, the data distribution of carbon emission data may change greatly, so the generalization of the carbon dioxide concentration prediction model directly constructed according to the randomly collected training samples is poor, which will lead to a large difference in data distribution between the training sample set used in the training process and the prediction sample set in the prediction process, affecting the technical defects of the accuracy of carbon dioxide concentration prediction, and improving the accuracy of carbon dioxide concentration prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0021] In order to more clearly illustrate the embodiments of the present application 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, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0022] Figure 1 This is a flow chart of the first embodiment of the carbon dioxide concentration prediction method of the present application;

[0023] Figure 2 Schematic diagram of the device structure of the hardware operating environment involved in the carbon dioxide concentration prediction method in the embodiment of the present application.

[0024] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of this application.

[0026] The present application embodiment provides a method for predicting carbon dioxide concentration, which is applied to a federal participant. In the first embodiment of the method for predicting carbon dioxide concentration of the present application, refer to Figure 1 , the carbon dioxide concentration prediction method comprises:

[0027] Step S10, obtaining first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the first carbon emission device and the second carbon emission device have different models;

[0028] Step S20, performing feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data;

[0029] Step S30, calculating a correction optimization parameter between the first carbon emission sample set and the second carbon emission sample set, wherein the correction optimization parameter is used to correct the data distribution of the carbon emission data generated by the first carbon emission device and the data distribution of the carbon emission data generated by the second carbon emission device to be consistent;

[0030] Step S40, obtaining a training data set corresponding to the first carbon emission device, and constructing a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters;

[0031] Step S50, obtaining a data set to be predicted corresponding to the second carbon emission device, and predicting the carbon dioxide concentration emitted by the second carbon emission device according to the data set to be predicted, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

[0032] In this embodiment, it should be noted that the first carbon emission device and the second carbon emission device can be boilers of a power plant, that is, the first carbon emission device can be a first-model boiler, and the second carbon emission device can be a second-model boiler. The first-model boiler and the second-model boiler are different in model, so there is a large difference in data distribution between the first carbon emission data generated by the first-model boiler and the second carbon emission data generated by the second-model boiler. The first carbon emission data and the second carbon emission data can both be temperature data, humidity data, and coal consumption data of the boiler, etc. The label corresponding to the first carbon emission data or the second carbon emission data can be a scalar, which is used to identify the concentration of carbon dioxide.

[0033] As an example, steps S10 to S50 include: obtaining temperature data, humidity data, and coal consumption data of a first model boiler as first carbon emission data, and obtaining temperature data, humidity data, and coal consumption data of a second model boiler as second carbon emission data, wherein the first model boiler and the second model boiler are of different models; performing feature transformation on the first carbon emission data and the second carbon emission data, respectively, to map samples in the first carbon emission data and samples in the second carbon emission data to preset sample dimensions, respectively, to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data, wherein the feature transformation method can be a random mapping of an extreme learning machine; calculating the first carbon emission sample set and the second carbon emission sample set. Put correction optimization parameters between sample sets, wherein the correction optimization parameters are used to correct the data distribution of the carbon emission data generated by the first carbon emission device and the data distribution of the carbon emission data generated by the second carbon emission device to be consistent; obtain the training data set corresponding to the first carbon emission device, use the correction optimization parameters to calibrate and optimize the training data set, and use the corrected and optimized training data set and the corresponding data labels to build a carbon dioxide concentration prediction model; obtain the data set to be predicted corresponding to the second carbon emission data, use the correction optimization parameters to calibrate and optimize the data set to be predicted, input the corrected and optimized data set to be predicted into the carbon dioxide concentration prediction model, predict the carbon dioxide concentration emitted by the second carbon emission device, and obtain the carbon dioxide concentration prediction value, which is the carbon dioxide concentration prediction result.

[0034] The first carbon emission device includes a first model boiler, and the step of obtaining first carbon emission data corresponding to the first carbon emission device includes:

[0035] Step S11, obtaining the temperature, humidity, coal consumption and carbon dioxide concentration of the first model boiler at a preset sampling time;

[0036] Step S12, combining the temperature, humidity and the coal consumption into a carbon emission sample feature vector, and generating a corresponding carbon emission sample label according to the carbon dioxide concentration;

[0037] Step S13: taking the carbon emission sample feature vector, the carbon emission sample label and the model label of the first model boiler together as the first carbon emission data.

[0038] In this embodiment, it should be noted that the boiler can be sampled once every time period, the coal consumption is the amount of coal consumed by the boiler within the time period between two sampling time points, and the first carbon emission data can be composed of a carbon emission sample feature vector, a carbon emission sample label and a model label of a first model boiler.

[0039] As an example, steps S11 to S13 include: obtaining the temperature, humidity, coal consumption and carbon dioxide concentration of the first model boiler at a preset sampling time; concatenating the temperature, humidity and coal consumption as characteristic values ​​into a vector to obtain a carbon emission sample characteristic vector; generating a corresponding carbon emission sample label according to the size of the carbon dioxide concentration; and using the carbon emission sample characteristic vector, the carbon emission sample label and the model label of the first model boiler as the first carbon emission data.

[0040] Similarly, as an example, the second carbon emission data is obtained as follows:

[0041] Obtain the temperature, humidity, coal consumption and carbon dioxide concentration of the second model boiler at the preset sampling time; use the temperature, humidity and coal consumption of the second model boiler as characteristic values ​​to splice into a vector to obtain a second carbon emission sample characteristic vector; generate a corresponding second carbon emission sample label according to the size of the carbon dioxide concentration corresponding to the second model boiler; use the second carbon emission sample characteristic vector, the second carbon emission sample label and the model label of the second model boiler together as the second carbon emission data.

[0042] The calculating of the correction optimization parameters between the first carbon emission sample set and the second carbon emission sample set includes:

[0043] Step S31, merging the first carbon emission sample set and the second carbon emission sample set to obtain a merged sample set;

[0044] Step S32, calculating the correction optimization parameters by constructing a preset optimization problem corresponding to the combined sample set.

[0045] As an example, step S31 to step S32 include:

[0046] The first carbon emission sample set and the second carbon emission sample set are concatenated to obtain a combined sample set, and the preset optimization problem corresponding to the combined sample set is constructed using the following formula:

[0047]

[0048] st T β=I

[0049] Where A=H T H+λ1H T LH+λ0I, L is the Laplace matrix calculated from two boiler sample sets, I is the unit matrix, β is the weight matrix, H is the merged sample set, λ0 and λ1 are balance coefficients, which are constants greater than 0; by performing eigenvalue decomposition on the preset optimization problem, the eigenvalues ​​corresponding to the Laplace matrix are obtained; the target weight matrix is ​​determined according to each of the eigenvalues, and the target weight matrix is ​​used as the correction optimization parameter, wherein the merged sample set can be a matrix, and the matrix size of the target weight matrix is ​​consistent with the matrix size corresponding to the merged sample set.

[0050] As an example, the step of determining the target weight matrix according to each of the eigenvalues ​​includes:

[0051] A maximum preset number of target eigenvalues ​​are selected from each of the eigenvalues; and the eigenvectors corresponding to each of the target eigenvalues ​​are concatenated into a matrix to obtain the target weight matrix.

[0052] Wherein, the step of constructing a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters includes:

[0053] Step S41, performing feature transformation on the training data set to obtain a training sample set;

[0054] Step S42, performing correction optimization on the training sample set according to the correction optimization parameters to obtain a corrected training sample set;

[0055] Step S43: constructing the carbon dioxide concentration prediction model according to the corrected training sample set and the sample labels corresponding to the training sample set.

[0056] In this embodiment, it should be noted that the feature change method may be a random mapping of an extreme learning machine.

[0057] As an example, steps S41 to S43 include: randomly mapping each training sample in the training data set to map each training sample to a preset sample dimension to obtain a training sample set; performing correction optimization on the training sample set according to the correction optimization parameter to convert each training sample in the training sample set to a preset data distribution to obtain a correction training sample set; selecting a correction training sample from the correction training sample set, inputting the correction training sample into an initialized carbon dioxide concentration prediction model, and predicting the output carbon dioxide concentration corresponding to the correction training sample; calculating the model loss according to the output carbon dioxide concentration and the sample label corresponding to the correction training sample; judging whether the model training has converged, if the model loss has converged, judging that the iterative training is completed, and using the initialized carbon dioxide concentration prediction model as the carbon dioxide concentration prediction model; if the model loss has not converged, judging that the iterative training is not completed, back-propagating and updating the initialized carbon dioxide concentration prediction model according to the gradient calculated by the model loss, and returning to the execution step: selecting a correction training sample from the correction training sample set until the calculated model loss converges.

[0058] The step of predicting the carbon dioxide concentration emitted by the second carbon emission equipment according to the data set to be predicted, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain the carbon dioxide concentration prediction result includes:

[0059] Step S51, performing feature transformation on the data set to be predicted to obtain a sample to be predicted;

[0060] Step S52, performing correction optimization on the sample to be predicted according to the correction optimization parameter to obtain a corrected prediction sample;

[0061] Step S53, predicting the carbon dioxide concentration emitted by the second carbon emission equipment by inputting the corrected prediction sample into the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

[0062] In this embodiment, it should be noted that the data set to be predicted is an unlabeled data set.

[0063] As an example, it should be noted that the data set to be predicted includes one or more samples, and steps S51 to S53 include: randomly mapping each sample in the data set to be predicted, so as to map each sample in the data set to be predicted to a preset sample dimension, and obtain the sample to be predicted; according to the correction optimization parameter, the sample to be predicted is corrected and optimized to convert the sample to be predicted to a preset data distribution, that is, the sample to be predicted belongs to the preset data distribution, and the corrected prediction sample is obtained; by inputting the corrected prediction sample into the carbon dioxide concentration prediction model, the carbon dioxide concentration emitted by the second carbon emission equipment is predicted, and the carbon dioxide concentration prediction value is obtained, that is, the carbon dioxide concentration prediction result. Thereby, the purpose of accurately predicting the carbon dioxide concentration by using the carbon dioxide concentration prediction model constructed with the labeled data of the first model boiler is achieved when the carbon emission data of the second model boiler is unlabeled, and the carbon dioxide concentration prediction is accurately predicted while considering the influence of different boiler models on the carbon dioxide concentration prediction.

[0064] As an example, the implementation process of the carbon dioxide concentration prediction method in the embodiment of the present application is as follows:

[0065] Step 1: Carbon Emissions Data Collection

[0066] Collect the carbon emission data set of the power plant. The temperature, humidity, coal consumption and other data at a certain sampling time form a feature vector. The coal consumption here refers to the total amount of coal used in the two sampling time intervals before and after.

[0067] A feature vector is a sample x, where x is a d-dimensional horizontal quantity and y is a scalar, i.e., the label corresponding to the sample, indicating the carbon dioxide concentration. By collecting data for a period of time, we can obtain a set of labeled samples {(x i ,y i ), i=1, ..., n}, different types of boilers can be distinguished, and a set of labeled samples corresponding to each type of boiler can be obtained.

[0068] Step 2: Carbon emission data preprocessing

[0069] The samples in the labeled sample set are randomly mapped. The random mapping can adopt the random mapping method from the input layer to the hidden layer in the extreme learning machine. A d-dimensional sample becomes m-dimensional after random mapping, which is represented by the horizontal quantity h.

[0070] Step 3: Carbon emission data correction

[0071] Assume boiler 1 has n s samples, boiler 2 has n t samples, then the random mapping matrix obtained by boiler 1 after step 2 is H s , the matrix is ​​ns ×m-dimensional, with one row representing the lateral volume generated by random mapping of a sample; correspondingly, the random mapping matrix of the data of boiler 2 is H t , the matrix is ​​n t ×m-dimensional;

[0072] H s and H t Merge to get H = [H s ;H t ], H is an n×m matrix;

[0073] Define the following optimization problem:

[0074]

[0075] st T β=I

[0076] Where A=H T H+λ1H T LH+λ0I, L is the Laplace matrix calculated from the two boiler sample sets, I is the identity matrix, β is the m×c weight matrix, λ0 and λ1 are balance coefficients, which are constants greater than 0;

[0077] To solve the above optimization problem: First, obtain the eigenvalues ​​of A by eigenvalue decomposition and find the eigenvectors a1,...,a corresponding to the largest c eigenvalues ​​among these eigenvalues. c , these eigenvectors are column vectors, and finally the optimized β=[a1,...,a c ].

[0078] Step 4: Train the carbon emission prediction model

[0079] Use random mapping in extreme learning to transform the data {x i , i=1,...,n s} is projected into a high-dimensional space for feature transformation to make it linearly separable, and we get {h i , i=1,...,n s}, and then obtain the corrected sample set {h i β, i = 1, ..., n s}, label these samples and train a regression model f, which can be a model such as support vector regression (SVR);

[0080] The data of boiler 2 Perform random mapping to obtain Then we get the corrected sample set Substitute it into f, That is the predicted value of the carbon dioxide concentration emitted by boiler 2.

[0081] The embodiments of the present application provide a method, apparatus, device and medium for predicting carbon dioxide concentration. Compared with the technical means used in the prior art to predict carbon dioxide concentration by directly constructing a carbon dioxide concentration prediction model based on randomly collected training samples, the embodiments of the present application first obtain first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the models of the first carbon emission device and the second carbon emission device are different; perform feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data; calculate the correction optimization parameters between the first carbon emission sample set and the second carbon emission sample set, wherein the correction optimization parameters are used to correct the data distribution of the carbon emission data generated by the first carbon emission device and the data distribution of the carbon emission data generated by the second carbon emission device to be consistent; obtain a training data set corresponding to the first carbon emission device, and construct a carbon dioxide concentration prediction model based on the training data set and the correction optimization parameters. The carbon concentration prediction model is a method for predicting the carbon dioxide concentration of the second carbon emission device, that is, according to the correction optimization parameters, the training data set is corrected to the preset data distribution, and then the corresponding carbon dioxide concentration prediction model is constructed; the to-be-predicted data set corresponding to the second carbon emission device is obtained, and the carbon dioxide concentration emitted by the second carbon emission device is predicted according to the to-be-predicted data set, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result, that is, after the to-be-predicted data set is corrected to the preset data distribution, it is input into the carbon dioxide concentration prediction model for prediction, so that the training sample set in the training process and the prediction sample set in the prediction process are consistent or similar in data distribution, so as to avoid that the data distribution of carbon emission data may change greatly for different models of carbon emission devices, so the carbon dioxide concentration prediction model directly constructed according to the randomly collected training samples has poor generalization, which will lead to a large difference in data distribution between the training sample set used in the training process and the prediction sample set in the prediction process, which affects the technical defects of the accuracy of carbon dioxide concentration prediction and improves the accuracy of carbon dioxide concentration prediction.

[0082] The embodiment of the present application also provides a device for predicting carbon dioxide concentration, the device comprising:

[0083] An acquisition module, configured to acquire first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the first carbon emission device and the second carbon emission device have different models;

[0084] Performing feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data;

[0085] A correction parameter determination module, used to calculate a correction optimization parameter between the first carbon emission sample set and the second carbon emission sample set;

[0086] A model building module, used to obtain a training data set corresponding to the first carbon emission device, and build a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters;

[0087] The prediction module is used to obtain the data set to be predicted corresponding to the second carbon emission device, and predict the carbon dioxide concentration emitted by the second carbon emission device according to the data set to be predicted, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

[0088] Optionally, the correction parameter determination module is further used to:

[0089] Merging the first carbon emission sample set and the second carbon emission sample set to obtain a merged sample set;

[0090] The correction optimization parameters are calculated by constructing a preset optimization problem corresponding to the combined sample set.

[0091] Optionally, the correction parameter determination module is further used to:

[0092] The preset optimization problem corresponding to the combined sample set is constructed using the following formula:

[0093]

[0094] st T β=I

[0095] Where A=H T H+λ1H T LH+λ0I, L is the Laplace matrix calculated from the two boiler sample sets, I is the identity matrix, β is the weight matrix, H is the combined sample set, λ0 and λ1 are balance coefficients, which are constants greater than 0;

[0096] Obtaining eigenvalues ​​corresponding to the Laplace matrix by performing eigenvalue decomposition on the preset optimization problem;

[0097] A target weight matrix is ​​determined according to each of the eigenvalues, and the target weight matrix is ​​used as the correction optimization parameter.

[0098] Optionally, the correction parameter determination module is further used to:

[0099] Selecting a maximum preset number of target eigenvalues ​​from among the eigenvalues;

[0100] The eigenvectors corresponding to the target eigenvalues ​​are combined into the target weight matrix.

[0101] Optionally, the model building module is further used to:

[0102] Performing feature transformation on the training data set to obtain a training sample set;

[0103] According to the correction optimization parameters, the training sample set is corrected and optimized to obtain a corrected training sample set;

[0104] The carbon dioxide concentration prediction model is constructed according to the corrected training sample set and the sample labels corresponding to the training sample set.

[0105] Optionally, the prediction module is further used for:

[0106] Performing feature transformation on the data set to be predicted to obtain a sample to be predicted;

[0107] According to the correction optimization parameters, the sample to be predicted is corrected and optimized to obtain a corrected prediction sample;

[0108] The concentration of carbon dioxide emitted by the second carbon emission device is predicted by inputting the corrected prediction sample into the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

[0109] Optionally, the first carbon emission device includes a first model boiler, and the acquisition module is further used for:

[0110] Obtaining the temperature, humidity, coal consumption and carbon dioxide concentration of the first model boiler at a preset sampling time;

[0111] Combining the temperature, humidity and coal consumption into a carbon emission sample feature vector, and generating a corresponding carbon emission sample label according to the carbon dioxide concentration;

[0112] The carbon emission sample feature vector, the carbon emission sample label and the model label of the first model boiler are collectively used as the first carbon emission data.

[0113] The carbon dioxide concentration prediction device provided by the present application adopts the carbon dioxide concentration prediction method in the above embodiment, which solves the technical problem of low accuracy of carbon dioxide concentration prediction due to model differences in carbon emission equipment. Compared with the prior art, the beneficial effects of the carbon dioxide concentration prediction device provided by the embodiment of the present application are the same as the beneficial effects of the carbon dioxide concentration prediction method provided by the above embodiment, and the other technical features in the carbon dioxide concentration prediction device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0114] An embodiment of the present application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the carbon dioxide concentration prediction method in the above-mentioned embodiment 1.

[0115] Reference below Figure 2 , which shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. The electronic device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 2 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0116] like Figure 2 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). In the RAM, various programs and data required for the operation of the electronic device are also stored. The processing device, ROM, and RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0117] Typically, the following systems can be connected to the I / O interface: input devices including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices including, for example, a magnetic tape, a hard disk, etc.; and communication devices. The communication device can allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows an electronic device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0118] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0119] The electronic device provided by the present application adopts the carbon dioxide concentration prediction method in the above embodiment to solve the technical problem of low accuracy of carbon dioxide concentration prediction due to model differences in carbon emission equipment. Compared with the prior art, the beneficial effects of the electronic device provided by the embodiment of the present application are the same as the beneficial effects of the carbon dioxide concentration prediction method provided by the above embodiment, and other technical features in the electronic device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0120] It should be understood that the various parts of the present disclosure can be implemented with hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0121] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0122] This embodiment provides a computer-readable storage medium having computer-readable program instructions stored thereon, and the computer-readable program instructions are used to execute the prediction method based on multimodal extreme learning in the above-mentioned embodiment 1.

[0123] The computer-readable storage medium provided in the embodiment of the present application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a system or a device or used in combination with it. The program code contained on the computer-readable storage medium can be transmitted with any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0124] The computer-readable storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0125] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by an electronic device, the electronic device: obtains first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the first carbon emission device and the second carbon emission device have different models; performs feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data; calculates correction optimization parameters between the first carbon emission sample set and the second carbon emission sample set, wherein the correction optimization parameters are used to correct the data distribution of the carbon emission data generated by the first carbon emission device and the data distribution of the carbon emission data generated by the second carbon emission device to be consistent; obtains a training data set corresponding to the first carbon emission device, and constructs a carbon dioxide concentration prediction model based on the training data set and the correction optimization parameters; obtains a to-be-predicted data set corresponding to the second carbon emission device, and predicts the carbon dioxide concentration emitted by the second carbon emission device based on the to-be-predicted data set, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

[0126] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0127] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0128] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of the module does not limit the unit itself in some cases.

[0129] The computer-readable storage medium provided by the present application stores computer-readable program instructions for executing the above-mentioned carbon dioxide concentration prediction method, which solves the technical problem of low accuracy of carbon dioxide concentration prediction due to model differences in carbon emission equipment. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the embodiment of the present application are the same as the beneficial effects of the carbon dioxide concentration prediction method provided in the above-mentioned embodiment, and will not be repeated here.

[0130] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned carbon dioxide concentration prediction method when executed by a processor.

[0131] The computer program product provided by the present application solves the technical problem of low accuracy in carbon dioxide concentration prediction due to differences in the models of carbon emission equipment. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiment of the present application are the same as the beneficial effects of the carbon dioxide concentration prediction method provided by the above embodiment, which will not be described in detail here.

[0132] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A method for predicting carbon dioxide concentration, characterized in that: The carbon dioxide concentration prediction method comprises: Acquire first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the first carbon emission device and the second carbon emission device have different models; Performing feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data; Calculating a correction optimization parameter between the first carbon emission sample set and the second carbon emission sample set, wherein the correction optimization parameter is used to correct the data distribution of the carbon emission data generated by the first carbon emission device and the data distribution of the carbon emission data generated by the second carbon emission device to be consistent; Acquire a training data set corresponding to the first carbon emission device, and construct a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters; Acquire a data set to be predicted corresponding to the second carbon emission device, and predict the carbon dioxide concentration emitted by the second carbon emission device according to the data set to be predicted, the correction optimization parameter and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result; The calculating of the correction optimization parameter between the first carbon emission sample set and the second carbon emission sample set includes: Merging the first carbon emission sample set and the second carbon emission sample set to obtain a merged sample set; The correction optimization parameter is calculated by constructing a preset optimization problem corresponding to the combined sample set; the preset optimization problem corresponding to the combined sample set is constructed using the following formula: stβ T β=I Where A=H T H+λ1H T LH+λ0I, L is the Laplace matrix calculated from the two boiler sample sets, I is the identity matrix, β is the weight matrix, H is the combined sample set, λ0 and λ1 are balance coefficients, which are constants greater than 0; Obtaining eigenvalues ​​corresponding to the Laplace matrix by performing eigenvalue decomposition on the preset optimization problem; A target weight matrix is ​​determined according to each of the eigenvalues, and the target weight matrix is ​​used as the correction optimization parameter.

2. The method for predicting carbon dioxide concentration according to claim 1, characterized in that: The step of determining the target weight matrix according to each of the eigenvalues ​​comprises: Selecting a maximum preset number of target eigenvalues ​​from among the eigenvalues; The eigenvectors corresponding to the target eigenvalues ​​are combined into the target weight matrix.

3. The method for predicting carbon dioxide concentration according to claim 1, characterized in that: The step of constructing a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters comprises: Performing feature transformation on the training data set to obtain a training sample set; According to the correction optimization parameters, the training sample set is corrected and optimized to obtain a corrected training sample set; The carbon dioxide concentration prediction model is constructed according to the corrected training sample set and the sample labels corresponding to the training sample set.

4. The method for predicting carbon dioxide concentration according to claim 1, characterized in that: The step of predicting the carbon dioxide concentration emitted by the second carbon emission equipment according to the data set to be predicted, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain the carbon dioxide concentration prediction result comprises: Performing feature transformation on the data set to be predicted to obtain a sample to be predicted; According to the correction optimization parameters, the sample to be predicted is corrected and optimized to obtain a corrected prediction sample; The concentration of carbon dioxide emitted by the second carbon emission device is predicted by inputting the corrected prediction sample into the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

5. The method for predicting carbon dioxide concentration according to claim 1, characterized in that: The first carbon emission device includes a first model boiler, and the step of obtaining first carbon emission data corresponding to the first carbon emission device includes: Obtaining the temperature, humidity, coal consumption and carbon dioxide concentration of the first model boiler at a preset sampling time; Combining the temperature, humidity and coal consumption into a carbon emission sample feature vector, and generating a corresponding carbon emission sample label according to the carbon dioxide concentration; The carbon emission sample feature vector, the carbon emission sample label and the model label of the first model boiler are collectively used as the first carbon emission data.

6. A device for predicting carbon dioxide concentration, characterized in that: The carbon dioxide concentration prediction device comprises: An acquisition module, configured to acquire first carbon emission data corresponding to a first carbon emission device and second carbon emission data corresponding to a second carbon emission device, wherein the first carbon emission device and the second carbon emission device have different models; Performing feature transformation on the first carbon emission data and the second carbon emission data respectively to obtain a first carbon emission sample set corresponding to the first carbon emission data and a second carbon emission sample set corresponding to the second carbon emission data; A correction parameter determination module is used to calculate the correction optimization parameters between the first carbon emission sample set and the second carbon emission sample set. The correction parameter determination module is also used to merge the first carbon emission sample set and the second carbon emission sample set to obtain a merged sample set; by constructing a preset optimization problem corresponding to the merged sample set, the correction optimization parameters are calculated, and the preset optimization problem corresponding to the merged sample set is constructed using the following formula: stβ T β=I Where A=H T H+λ1H T LH+λ0I, L is the Laplace matrix calculated from the two boiler sample sets, I is the unit matrix, β is the weight matrix, H is the combined sample set, λ0 and λ1 are balance coefficients, which are constants greater than 0; by performing eigenvalue decomposition on the preset optimization problem, the eigenvalues ​​corresponding to the Laplace matrix are obtained; the target weight matrix is ​​determined according to each of the eigenvalues, and the target weight matrix is ​​used as the correction optimization parameter; A model building module, used to obtain a training data set corresponding to the first carbon emission device, and build a carbon dioxide concentration prediction model according to the training data set and the correction optimization parameters; The prediction module is used to obtain the data set to be predicted corresponding to the second carbon emission device, and predict the carbon dioxide concentration emitted by the second carbon emission device according to the data set to be predicted, the correction optimization parameters and the carbon dioxide concentration prediction model to obtain a carbon dioxide concentration prediction result.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the carbon dioxide concentration prediction method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program for implementing a method for predicting carbon dioxide concentration, and the program for implementing a method for predicting carbon dioxide concentration is executed by a processor to implement the steps of the method for predicting carbon dioxide concentration as claimed in any one of claims 1 to 5.

Images