CN117933493B - Multi-time granularity data coordinated carbon emission factor prediction method and system - Google Patents
Multi-time granularity data coordinated carbon emission factor prediction method and system Download PDFInfo
- Publication number
- CN117933493B CN117933493B CN202410323416.0A CN202410323416A CN117933493B CN 117933493 B CN117933493 B CN 117933493B CN 202410323416 A CN202410323416 A CN 202410323416A CN 117933493 B CN117933493 B CN 117933493B
- Authority
- CN
- China
- Prior art keywords
- prediction
- carbon emission
- matrix
- coordination
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 185
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 184
- 238000000034 method Methods 0.000 title claims abstract description 32
- 239000011159 matrix material Substances 0.000 claims abstract description 159
- 238000012549 training Methods 0.000 claims description 77
- 238000012360 testing method Methods 0.000 claims description 66
- 238000005457 optimization Methods 0.000 claims description 45
- 238000012795 verification Methods 0.000 claims description 36
- 238000010200 validation analysis Methods 0.000 claims description 18
- 238000011156 evaluation Methods 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 13
- 230000000903 blocking effect Effects 0.000 claims description 3
- 235000019580 granularity Nutrition 0.000 description 101
- 238000010586 diagram Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 230000002123 temporal effect Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 230000008878 coupling Effects 0.000 description 5
- 238000010168 coupling process Methods 0.000 description 5
- 238000005859 coupling reaction Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 3
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Accounting & Taxation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Finance (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a carbon emission factor prediction method and a system for multi-time granularity data coordination, comprising the following steps: collecting first carbon emission data of a region to be predicted, and establishing a plurality of first prediction models for predicting carbon emission factors according to time granularity of the first carbon emission data; wherein each time granularity corresponds to a first prediction model; establishing a time hierarchy matrix according to the time granularity and the period to be predicted, and fusing a plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model; parameter updating is carried out on the first coordination prediction model according to the first carbon emission data, so that a trained second coordination prediction model is obtained, and carbon emission factors are predicted through the second coordination prediction model; the method and the device can improve the prediction accuracy of the carbon emission factor.
Description
Technical Field
The invention relates to the technical field of carbon emission, in particular to a carbon emission factor prediction method and system with coordinated multi-time granularity data.
Background
Currently, in the process of accelerating the construction of the advanced carbon market, enterprises and organizations are pushed to reduce carbon emission through an economic incentive mechanism, so that the double-carbon target is accelerated to be realized. In this market, this numerical index of carbon emission factors provides quantification and assessment of carbon emission levels, providing fundamental data and guidance for the design and operation of the carbon market. And the prediction of the carbon emission factors can enable market managers to know the carbon emission factor change trend of different activities and industries, and reasonable carbon emission limits and pricing are set, so that emission reduction is guided and promoted.
However, the data for carbon emission factors present a time granularity problem: published carbon emission factors are recorded in "years"; and places may hold carbon emission factor data of finer granularity (e.g. "month"). Therefore, if the existing carbon emission factor prediction method is used to predict the data of different time particles, the generated prediction result is often difficult to be self-consistent, i.e. the aggregated fine-grained prediction result is inconsistent with the coarse-grained prediction. This results in low data utilization for different time granularity for the existing carbon emission factor prediction method, and often ignores the time-level coupling relationship, resulting in low accuracy of the obtained carbon emission factor.
Disclosure of Invention
The invention aims to provide a carbon emission factor prediction method and a system with coordinated multi-time granularity data, aiming at the defects of the prior related technology, and can improve the accuracy of the carbon emission factor.
In a first aspect, the present invention provides a method for predicting a carbon emission factor coordinated by multi-time granularity data, including:
collecting first carbon emission data of a region to be predicted, and establishing a plurality of first prediction models for predicting carbon emission factors according to time granularity of the first carbon emission data; wherein each time granularity corresponds to a first prediction model;
Establishing a time hierarchy matrix according to the time granularity and the period to be predicted, and fusing a plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model;
And updating parameters of the first coordination prediction model according to the first carbon emission data to obtain a trained second coordination prediction model so as to predict carbon emission factors through the second coordination prediction model.
According to the method, the prediction model is respectively built for different time granularities, so that the utilization rate of multi-element time sequence data can be improved, a time hierarchy matrix is built according to the time granularities and the period to be predicted, the time hierarchy matrix is used for fusing the plurality of prediction models, the prediction results of different fine granularities can be aggregated, the implicit influence of the coupling relation of the time hierarchy on the prediction of the carbon emission factors is enhanced, the self-consistent coordinated carbon emission factor coordination prediction model is built, the consistent multi-time hierarchy prediction result can be formed, the prediction precision of the carbon emission factors is improved, and a foundation is provided for the insight analysis of the change trend of the carbon emission factors under different time scales.
Further, the fusing the plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model includes:
And constructing an initial coordination matrix according to the quantity of the carbon emission factor predicted values at the finest time granularity, acquiring an output matrix formed by the outputs of the plurality of first prediction models, and constructing an initial first coordination prediction model according to the point multiplication sequence of the time hierarchy matrix, the initial coordination matrix and the output matrix.
According to the invention, the output matrix formed by outputting a plurality of prediction models is fused by using the time hierarchy matrix and the coordination matrix, so that the prediction results of different prediction models can be aggregated, the implicit influence of the time hierarchy coupling relation on the prediction of the carbon emission factor is enhanced, and a self-consistent coordinated carbon emission factor coordination prediction model is constructed, thereby improving the prediction precision of the carbon emission factor.
Further, the constructing an initial coordination matrix according to the predicted value quantity of the carbon emission factors under the granularity of the finest time comprises the following steps:
And constructing a unit square matrix by taking the quantity of the carbon emission factors predicted under the finest time granularity as a row and taking the difference between the total quantity of the carbon emission factors predicted under all time granularity and the quantity of the carbon emission factors predicted as a column, and constructing a zero matrix according to the initial coordination matrix of the block matrix formed by the zero matrix and the unit square matrix.
Further, the establishing a time hierarchy matrix according to the time granularity and the period to be predicted includes:
If the time granularity comprises: years, seasons and months, then the time granularity If the period to be predicted is 12 months in the future, the time hierarchy matrix is expressed as:
Wherein, ，/>；/>I is an identity matrix.
According to the method, 3 time granularities are established through year, season and month, and a time hierarchy matrix consisting of the 3 time granularities is established through the period to be predicted, so that the prediction results of different fine granularities are convenient to aggregate, the prediction accuracy of the carbon emission factors can be improved, and the reliability of the carbon emission factor prediction are improved.
Further, the updating parameters of the first coordination prediction model according to the first carbon emission data includes:
sequentially dividing second carbon emission data of the plurality of first prediction models into a training set and a verification set according to a proportion; the first carbon emission data includes: second carbon emission data corresponding to each of the first predictive models;
Updating initial parameters and an initial coordination matrix of the first coordination prediction model according to the training set, the verification set and a pre-constructed double-layer optimization model; wherein the bilayer optimization model comprises: a lower layer optimization model and an upper layer optimization model; the lower optimization model optimizes the initial parameters according to the training set to obtain optimal parameters; and the upper optimization model optimizes the initial coordination matrix according to the verification set and the optimal parameters.
Further, the double-layer optimization model is expressed as:
，
Wherein, Is an underlying optimization model,/>Expressed as/>Number of training set samples of the first predictive model,/>；/>Is/>The true load of the individual training samples; /(I)Is the initial coordination matrix/>Is/>Initial parameter sets of the first prediction models; /(I)Is the prediction result of the training set input to the first coordinated prediction model; /(I)，/>Is the prediction result from the test set to the first coordination prediction model; /(I)Expressed as/>The number of validation set samples of the first predictive model; /(I)Is/>The true load of the sample is verified.
Further, the sequentially dividing the second carbon emission data of the plurality of first prediction models into a training set and a validation set, includes:
Sequentially screening the characteristics of the second carbon emission data of the plurality of first prediction models to obtain characteristic screening results, dividing the characteristic screening results into a training set, a verification set and a test set according to a proportion, normalizing the training set, the verification set and the test set according to columns respectively to obtain a normalized training set, a normalized verification set and a normalized test set respectively, and training, verifying and testing according to the normalized training set, the normalized verification set and the normalized test set finally; wherein each first predictive model corresponds to a normalized training set, a normalized validation set, and a normalized test set.
Further, testing according to the normalized test set, comprising:
Respectively inputting a plurality of normalized test sets into a corresponding optimized second prediction model, respectively outputting corresponding prediction results according to optimized optimal parameters, inputting a plurality of prediction results into the second prediction model, and outputting total prediction results according to an optimized optimal coordination matrix;
And according to the real load of each normalized test set and the total prediction result, performing prediction accuracy evaluation on each second prediction model.
Further, the evaluating the prediction accuracy of each second prediction model includes:
Taking the average absolute percentage error as an evaluation index of the prediction precision of the second prediction model, wherein the evaluation index is expressed as:
，
Wherein, For testing sample sequence number,/>For the time point sequence number of the period to be predicted,/>For normalizing the prediction results of the test set,/>For testing the real load corresponding to the sample,/>For/>Number of test set samples of the second predictive model,/>Is the total number of time points of the period to be predicted.
In a second aspect, the present invention provides a multi-time granularity data coordinated carbon emission factor prediction system, comprising: the method comprises the steps of establishing a prediction model module, establishing a coordination prediction model module and a training module; wherein,
The prediction model establishment module is used for acquiring first carbon emission data of a region to be predicted, and establishing a plurality of first prediction models for predicting carbon emission factors according to time granularity; wherein each time granularity corresponds to a first prediction model;
The coordination prediction model establishing module is used for establishing a time hierarchy matrix according to the time granularity and the period to be predicted, and fusing a plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model;
And the training module is used for carrying out parameter updating on the first coordination prediction model according to the first carbon emission data to obtain a trained second coordination prediction model so as to predict the carbon emission factor through the second coordination prediction model.
Drawings
FIG. 1 is a flow chart of a method for predicting carbon emission factors with coordinated multi-time granularity data according to the present embodiment;
FIG. 2 is a schematic diagram of a coordinated prediction model provided by the present embodiment;
FIG. 3 is a flow chart of a complete method for predicting carbon emission factors with coordinated multi-time granularity data according to the present embodiment;
Fig. 4 is a schematic structural diagram of a carbon emission factor prediction system with coordinated multi-time granularity data according to the present embodiment.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It is worth noting that, in order to overcome the shortcomings of the prior art, the invention provides a carbon emission factor prediction method and system coordinated by multi-time granularity data, firstly, the carbon emission factor condition of a region to be predicted is researched, different time granularities to be predicted are determined, and a time hierarchy matrix is established; then, establishing a prediction model and a time hierarchy coordination model of the carbon emission factors; then, preprocessing the data set, and updating a prediction model and a coordination matrix to obtain a fitted prediction model of the multi-time granularity carbon emission factor; finally, calculating prediction accuracy of the carbon emission factor prediction under different time granularity, and evaluating the prediction effect of the carbon emission factor prediction; the method and the device help to solve the problem that the carbon emission factors are difficult and uncooled to predict under different time granularities, thereby improving the prediction accuracy of the carbon emission factors. In order to better explain the technical aspects of the present invention, the following examples will be described in detail.
Example 1
Referring to fig. 1, a flow chart of a carbon emission factor prediction method coordinated by multi-time granularity data provided in this embodiment includes: the steps S11 to S13 specifically include:
Step S11, collecting first carbon emission data of a region to be predicted, and establishing a plurality of first prediction models for predicting carbon emission factors according to time granularity; wherein each temporal granularity corresponds to a first predictive model.
In some embodiments, collecting carbon emission data of a region to be predicted, analyzing the change condition of carbon emission factors, and determining k different time granularities to be predicted and time periods to be predicted by combining expert advice; wherein k is a positive integer.
It is worth noting that the first carbon emission data is historical carbon emission data of the area to be predicted, and the first prediction model is an initial prediction model corresponding to time granularity.
In some embodiments, establishing a time hierarchy matrix according to the time granularity and the period to be predicted includes: if the time granularity comprises: years, seasons and months, then the time granularityIf the period to be predicted is 12 months in the future, the time hierarchy matrix is expressed as:
Wherein, ，/>；/>I is an identity matrix.
In some embodiments, a temporal hierarchy matrix is established based on the selected temporal granularity and periodWherein/>The number of predicted values for all time granularity periods; /(I)Is the number of predictors at the finest temporal granularity. Time hierarchy matrix/>Can be written as/>Individual blocking matrix/>Each block matrix representation constructs the/>, with the predicted value at the finest temporal granularityWeight of individual target temporal granularity predictions. If the carbon emission factor at the finest time granularity is known to be predicted as/>Wherein/>The total prediction at all time granularity can be expressed as/>, which is the total number of prediction points at the finest granularity，/>Is transposed.
In some embodiments, according to the selected time granularity and the duration to be predicted, the time granularity and the duration to be predicted can be respectivelyPredictive models of individual time-granularity carbon emission factors determine corresponding input feature spaces/>And output space/>。
Thus, the predictive model of carbon emission factor at different time granularities can be expressed as:
，
Wherein, For/>Input and output data of a prediction model with a plurality of time granularity; /(I)For/>A parameterized neural network regression model with trainable parameters/>；/>Is an error term.
In some embodiments, if respectively toAdjust its parameter/>Fitting to finally obtain/>The predicted value of the obtained carbon emission factor is:
。
And step S12, a time hierarchy matrix is established according to the time granularity and the period to be predicted, and a plurality of first prediction models are fused according to the time hierarchy matrix to obtain an initial first coordination prediction model.
In some embodiments, the fusing the plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordinated prediction model includes: and constructing an initial coordination matrix according to the quantity of the carbon emission factor predicted values at the finest time granularity, acquiring an output matrix formed by the outputs of the plurality of first prediction models, and constructing an initial first coordination prediction model according to the point multiplication sequence of the time hierarchy matrix, the initial coordination matrix and the output matrix.
It should be noted that the first coordination prediction model is an initial coordination prediction model established according to a time hierarchy matrix, a plurality of prediction models and a coordination matrix, and the initial coordination matrix and the optimal coordination matrix are respectively the initial coordination matrix and the optimal coordination matrix after optimization.
In some embodiments, referring to fig. 2, a schematic diagram of a coordinated prediction model provided in this embodiment is shown. In fig. 2, a prediction model is respectively established for each time granularity, a time hierarchy matrix is established according to the time granularity and the period to be predicted, a coordination prediction model is established according to the prediction model, the coordination matrix and the time hierarchy matrix, and parameters of the prediction model and the coordination matrix are substantially updated when the coordination prediction model is updated.
In some embodiments, the polymerizationThe prediction results of the prediction models can obtain an output matrix/>, which is composed of the carbon emission factors and is output by the plurality of prediction models，/>For/>Prediction results of carbon emission factors of each prediction model,/>Is transposed. Then, output matrix/>Time hierarchy matrix/>And a coordination matrix, constructing a coordination prediction model, which can be expressed as:
，
Wherein, For coordination matrix, if/>Meaning that only the finest granularity of the prediction result/>Layer-by-layer aggregation is carried out until the prediction of the coarsest granularity is carried out, and the method is the simplest initial coordination matrix. /(I)Can be understood as superparameters for adjusting the output matrix/>, of a plurality of predictive models; The total prediction result after final coordination is/>。
In some embodiments, constructing the initial coordination matrix based on the number of carbon emission factor predictions at the finest time granularity includes: and constructing a unit square matrix by taking the quantity of the carbon emission factors predicted under the finest time granularity as a row and taking the difference between the total quantity of the carbon emission factors predicted under all time granularity and the quantity of the carbon emission factors predicted as a column, and constructing a zero matrix according to the initial coordination matrix of the block matrix formed by the zero matrix and the unit square matrix.
And step S13, carrying out parameter updating on the first coordination prediction model according to the first carbon emission data to obtain a trained second coordination prediction model so as to predict the carbon emission factor through the second coordination prediction model.
In some embodiments, parameter updating the first coordinated prediction model according to the first carbon emission data comprises: sequentially dividing second carbon emission data of the plurality of first prediction models into a training set and a verification set according to a proportion; the first carbon emission data includes: second carbon emission data corresponding to each of the first predictive models; updating initial parameters and an initial coordination matrix of the first coordination prediction model according to the training set, the verification set and a pre-constructed double-layer optimization model; wherein the bilayer optimization model comprises: a lower layer optimization model and an upper layer optimization model; the lower optimization model optimizes the initial parameters according to the training set to obtain optimal parameters; and the upper optimization model optimizes the initial coordination matrix according to the verification set and the optimal parameters.
It is worth to say that the second carbon emission data is carbon emission data used by each prediction model, time scales among the second carbon emission data are different according to different time granularity, and the second coordination prediction model is a coordination prediction model after the initial first coordination prediction model is trained.
In some embodiments, the sequentially scaling the second carbon emission data of the plurality of first predictive models into a training set and a validation set comprises: sequentially screening the characteristics of the second carbon emission data of the plurality of first prediction models to obtain characteristic screening results, dividing the characteristic screening results into a training set, a verification set and a test set according to a proportion, normalizing the training set, the verification set and the test set according to columns respectively to obtain a normalized training set, a normalized verification set and a normalized test set respectively, and training, verifying and testing according to the normalized training set, the normalized verification set and the normalized test set finally; wherein each first predictive model corresponds to a normalized training set, a normalized validation set, and a normalized test set.
In some embodiments, according to the input space and the output spaceThe/>, can be constructed from the carbon emission factor datasetTotal data set/>, at individual time granularity; Then/>According to the time sequence, according to 8:2: the ratio of 2 is divided into training set/>Verification set/>And test set/>; Then, the input data needs to be normalized, and the normalization method is as follows:
，
Wherein, Is the minimum on the column vector,/>Is the maximum value on the column vector,/>And/>The elements of the column vector before and after normalization, respectively.
In some embodiments, the above normalization method is applied to the variables of each column in the global training set, and the maximum and minimum values of each column in the training set are memorized and applied to the verification set and the test set by the same normalization method. Thus, a global dataset after data preprocessing can be obtained.
In some embodiments, training set loss functions are defined separatelyAnd validation set loss function/>Establishing a double-layer optimization model:
，
Wherein the method comprises the steps of Expressed as/>Number of training set samples at a granularity of time. When the matrix/>, is coordinatedGiven, the corresponding optimal predictive model parameters/>Can be expressed as:
。
In some embodiments, the matrix is coordinated Can be interpreted as a hyper-parameter, so that the loss function/>, according to the validation set, can be performed on the validation setAnd (3) adjusting:
，
Wherein, Expressed as/>Number of validation set samples at a single time granularity.
It is noted that the carbon emission data is divided into the following groups according to the time granularityA prediction model is established for each carbon emission data set, namely, one time granularity corresponds to one prediction model, and the/>The granularity of the time refers to the/>And a prediction model.
In some embodiments, the two-layer optimization model is expressed as:
，
Wherein, Is an underlying optimization model,/>Expressed as/>Number of training set samples of the first predictive model,/>；/>Is/>The true load of the individual training samples; /(I)Is the initial coordination matrix/>Is/>Initial parameter sets of the first prediction models; /(I)Is the prediction result of the training set input to the first coordinated prediction model; /(I)，/>Is the prediction result from the test set to the first coordination prediction model; /(I)Expressed as/>The number of validation set samples of the first predictive model; /(I)Is/>The true load of the sample is verified.
It should be noted that the underlying optimization aims at giving a given coordination matrix based on training set lossOptimal carbon emission factor prediction model parameter set/>; The upper layer optimization aims at adjusting the coordination matrix/>, according to the verification set lossThereby jointly optimizing the predictive model parameters and the coordination matrix.
In some embodiments, testing according to the normalized test set includes: respectively inputting a plurality of normalized test sets into a corresponding optimized second prediction model, respectively outputting corresponding prediction results according to optimized optimal parameters, inputting a plurality of prediction results into the second prediction model, and outputting total prediction results according to an optimized optimal coordination matrix; and according to the real load of each normalized test set and the total prediction result, performing prediction accuracy evaluation on each second prediction model.
In some embodiments, the optimal coordination matrix is obtainedOptimal carbon emission factor prediction model parameters/>, under each time granularityLater, at various times, the set of particle size tests/>Predicting and coordinating the carbon emission factors to obtain a total prediction result/>Which can be disassembled into prediction results/>, under various time granularity, according to a time hierarchy structure。
In some embodiments, the evaluating the prediction accuracy of each second prediction model includes: taking the average absolute percentage error as an evaluation index of the prediction precision of the second prediction model, wherein the evaluation index is expressed as:
，
Wherein, For testing sample sequence number,/>For the time point sequence number of the period to be predicted,/>For normalizing the prediction results of the test set,/>For testing the real load corresponding to the sample,/>For/>Number of test set samples of the second predictive model,/>Is the total number of time points of the period to be predicted.
It is worth noting that the smaller the MAPE value, the higher the accuracy of the carbon emission factor prediction. Thus, it can beAnd evaluating the carbon emission factor prediction results of the individual time granularity.
The carbon emission factor prediction method for coordinating the multi-time granularity data provided by the real-time example can efficiently utilize the different-time granularity data to construct a coordinated and high-performance carbon emission factor prediction model, thereby providing a basis for the insight analysis of the change trend of the carbon emission factor under different time scales.
Example 2
Referring to fig. 3, a flow chart of a complete carbon emission factor prediction method coordinated by multi-time granularity data provided in this embodiment includes: the steps S21-S26 specifically comprise:
S21, collecting carbon emission factors of a region to be predicted, and determining And establishing a time hierarchy matrix according to different time granularities to be predicted.
S22, combining expert opinion, screening input characteristics for each time granularity, and respectively establishing a carbon emission factor prediction model。
S23, constructing a coordinated prediction model according to the prediction model and the time hierarchy matrix under different time granularities。
And step S24, dividing and normalizing the data set to obtain a training set, a verification set and a test set after data preprocessing.
And S25, inputting a training set and a verification set, and iteratively updating the prediction model and the coordination matrix to obtain a fitted multi-time granularity carbon emission factor prediction model. The method specifically comprises two substeps: substep S251 and substep S252.
Sub-step S251 of defining training set loss function respectivelyAnd validation set loss function/>Establishing a double-layer optimization model:
。
Sub-step S252, to train the model, iteratively updating model parameters using gradient descent method And a coordination matrix G, obtaining an optimal coordination matrix/>Optimal carbon emission factor prediction model parameters/>, under each time granularity。
And S26, inputting a test set, calculating prediction accuracy of the carbon emission factor predictions at different time granularity, and evaluating the prediction effect. Mean absolute percent error (Mean Absolute Percentage Error, MAPE) was used as an evaluationIndex of prediction accuracy of individual time granularity, prediction result/>And true load/>The following formula is carried in for calculation:
。
the smaller the MAPE value, the higher the accuracy of the carbon emission factor prediction. Thus, it can be And evaluating the carbon emission factor prediction results of the individual time granularity.
According to the embodiment, the prediction models are respectively built for different time granularities, so that the utilization rate of multi-element time sequence data can be improved, a time hierarchy matrix is built according to the time granularities and the period to be predicted, the time hierarchy matrix is used for fusing the prediction models, prediction results of different fine granularities can be aggregated, the implicit influence of a time hierarchy coupling relation on the prediction of the carbon emission factors is enhanced, a self-consistent coordinated carbon emission factor coordination prediction model is built, and consistent multi-time hierarchy prediction results can be formed, so that the prediction precision of the carbon emission factors is improved, and a foundation is provided for the insight analysis of the change trend of the carbon emission factors under different time scales.
Example 3
Referring to fig. 4, a schematic structural diagram of a carbon emission factor prediction system coordinated by multi-time granularity data according to the present embodiment includes: a build prediction model module 31, a build coordination prediction model module 32, and a training module 33.
It should be noted that, the model building module 31 mainly divides the carbon emission data according to the time granularity, builds a prediction model for each time granularity, and transmits the number of the time granularity and a plurality of prediction models to the model building module 32; after the coordination prediction model building module 32 receives the time granularity and the plurality of prediction models, a time hierarchy matrix is built in combination with the period to be predicted, the plurality of prediction models are fused according to the time hierarchy matrix, a coordination prediction model is built, and the coordination prediction model is transmitted to the training module 33; the training module 33 trains the received coordinated prediction model, and uses the trained coordinated prediction model to conduct coordinated prediction of the carbon emission factor.
The prediction model building module 31 is configured to collect first carbon emission data of an area to be predicted, and build a plurality of first prediction models for predicting carbon emission factors according to time granularity of the first carbon emission data; wherein each temporal granularity corresponds to a first predictive model.
In some embodiments, establishing a time hierarchy matrix according to the time granularity and the period to be predicted includes: if the time granularity comprises: years, seasons and months, then the time granularityIf the period to be predicted is 12 months in the future, the time hierarchy matrix is expressed as:
Wherein, ，/>；/>I is an identity matrix.
And the coordination prediction model establishing module 32 is configured to establish a time hierarchy matrix according to the time granularity and the period to be predicted, and fuse a plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model.
In some embodiments, the fusing the plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordinated prediction model includes: and constructing an initial coordination matrix according to the quantity of the carbon emission factor predicted values at the finest time granularity, acquiring an output matrix formed by the outputs of the plurality of first prediction models, and constructing an initial first coordination prediction model according to the point multiplication sequence of the time hierarchy matrix, the initial coordination matrix and the output matrix.
In some embodiments, constructing the initial coordination matrix based on the number of carbon emission factor predictions at the finest time granularity includes: and constructing a unit square matrix by taking the quantity of the carbon emission factors predicted under the finest time granularity as a row and taking the difference between the total quantity of the carbon emission factors predicted under all time granularity and the quantity of the carbon emission factors predicted as a column, and constructing a zero matrix according to the initial coordination matrix of the block matrix formed by the zero matrix and the unit square matrix.
And the training module 33 is configured to update parameters of the first coordination prediction model according to the first carbon emission data to obtain a trained second coordination prediction model, so that the carbon emission factor is predicted by the second coordination prediction model.
In some embodiments, parameter updating the first coordinated prediction model according to the first carbon emission data comprises: sequentially dividing second carbon emission data of the plurality of first prediction models into a training set and a verification set according to a proportion; the first carbon emission data includes: second carbon emission data corresponding to each of the first predictive models; updating initial parameters and an initial coordination matrix of the first coordination prediction model according to the training set, the verification set and a pre-constructed double-layer optimization model; wherein the bilayer optimization model comprises: a lower layer optimization model and an upper layer optimization model; the lower optimization model optimizes the initial parameters according to the training set to obtain optimal parameters; and the upper optimization model optimizes the initial coordination matrix according to the verification set and the optimal parameters.
In some embodiments, the two-layer optimization model is expressed as:
，
Wherein, Is an underlying optimization model,/>Expressed as/>Number of training set samples of the first predictive model,/>；/>Is/>The true load of the individual training samples; /(I)Is the initial coordination matrix/>Is/>Initial parameter sets of the first prediction models; /(I)Is the prediction result of the training set input to the first coordinated prediction model; /(I)，/>Is the prediction result from the test set to the first coordination prediction model; /(I)Expressed as/>The number of validation set samples of the first predictive model; /(I)Is/>The true load of the sample is verified.
In some embodiments, the sequentially scaling the second carbon emission data of the plurality of first predictive models into a training set and a validation set comprises: sequentially screening the characteristics of the second carbon emission data of the plurality of first prediction models to obtain characteristic screening results, dividing the characteristic screening results into a training set, a verification set and a test set according to a proportion, normalizing the training set, the verification set and the test set according to columns respectively to obtain a normalized training set, a normalized verification set and a normalized test set respectively, and training, verifying and testing according to the normalized training set, the normalized verification set and the normalized test set finally; wherein each first predictive model corresponds to a normalized training set, a normalized validation set, and a normalized test set.
In some embodiments, testing according to the normalized test set includes: respectively inputting a plurality of normalized test sets into a corresponding optimized second prediction model, respectively outputting corresponding prediction results according to optimized optimal parameters, inputting a plurality of prediction results into the second prediction model, and outputting total prediction results according to an optimized optimal coordination matrix; and according to the real load of each normalized test set and the total prediction result, performing prediction accuracy evaluation on each second prediction model.
In some embodiments, the evaluating the prediction accuracy of each second prediction model includes: taking the average absolute percentage error as an evaluation index of the prediction precision of the second prediction model, wherein the evaluation index is expressed as:
，
Wherein, For testing sample sequence number,/>For the time point sequence number of the period to be predicted,/>For normalizing the prediction results of the test set,/>For testing the real load corresponding to the sample,/>For/>Number of test set samples of the second predictive model,/>Is the total number of time points of the period to be predicted.
In the embodiment, the prediction model is respectively built for different time granularities by the prediction model building module 31, so that the utilization rate of multi-element time sequence data can be improved, the time hierarchy matrix is built by the coordination prediction model building module 32 according to the time granularities and the period to be predicted, the time hierarchy matrix is used for fusing the plurality of prediction models, the prediction results of different fine granularities can be aggregated, the implicit influence of the coupling relation of the time hierarchy on the prediction of the carbon emission factors is enhanced, the coordination prediction model of the self-coordination carbon emission factors is built, the training is performed by the training module 33, the optimal and consistent multi-time hierarchy prediction results can be formed, so that the prediction precision of the carbon emission factors is improved, and a basis is provided for the insight analysis of the change trend of the carbon emission factors under different time scales.
It will be appreciated by those skilled in the art that embodiments of the present application may also be provided including a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (5)
1. A multi-time granularity data coordinated carbon emission factor prediction method, comprising:
collecting first carbon emission data of a region to be predicted, and establishing a plurality of first prediction models for predicting carbon emission factors according to time granularity of the first carbon emission data; wherein each time granularity corresponds to a first prediction model;
Establishing a time hierarchy matrix according to the time granularity and the period to be predicted, and fusing a plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model;
Parameter updating is carried out on the first coordination prediction model according to the first carbon emission data, so that a trained second coordination prediction model is obtained, and carbon emission factors are predicted through the second coordination prediction model;
the fusing the plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model comprises the following steps:
Constructing an initial coordination matrix according to the quantity of the carbon emission factor predicted values at the finest time granularity, acquiring an output matrix formed by the output of a plurality of first prediction models, and constructing an initial first coordination prediction model according to the point multiplication sequence of the time hierarchy matrix, the initial coordination matrix and the output matrix;
the method for constructing the initial coordination matrix according to the quantity of the carbon emission factor predicted values under the granularity of the finest time comprises the following steps:
constructing a unit square matrix by taking the quantity of the carbon emission factors predicted value at the finest time granularity as a row and taking the difference between the total quantity of the carbon emission factors predicted value and the quantity of the carbon emission factors predicted value at all time granularity as a column, constructing a zero matrix, and taking a blocking matrix formed by the zero matrix and the unit square matrix as an initial coordination matrix;
the establishing a time hierarchy matrix according to the time granularity and the period to be predicted comprises the following steps:
If the time granularity comprises: years, seasons and months, the time granularity is If the period to be predicted is 12 months in the future, the time hierarchy matrix is expressed as:
Wherein, ，/>；/>I is an identity matrix;
the parameter updating of the first coordination prediction model according to the first carbon emission data comprises:
sequentially dividing second carbon emission data of the plurality of first prediction models into a training set and a verification set according to a proportion; the first carbon emission data includes: second carbon emission data corresponding to each of the first predictive models;
Updating initial parameters and an initial coordination matrix of the first coordination prediction model according to the training set, the verification set and a pre-constructed double-layer optimization model; wherein the bilayer optimization model comprises: a lower layer optimization model and an upper layer optimization model; the lower optimization model optimizes the initial parameters according to the training set to obtain optimal parameters; the upper optimization model optimizes the initial coordination matrix according to the verification set and the optimal parameters;
The double-layer optimization model is expressed as:
，
Wherein, Is an underlying optimization model,/>Expressed as/>Number of training set samples of the first predictive model,/>；/>Is/>Carbon emission true values for the individual training samples; /(I)Is the initial coordination matrix/>Is/>Initial parameter sets of the first prediction models; /(I)Is the prediction result of the training set input to the first coordinated prediction model; /(I)，/>Is the prediction result from the test set to the first coordination prediction model; /(I)Expressed as/>The number of validation set samples of the first predictive model; /(I)Is/>The actual value of carbon emissions of the samples was verified.
2. The method for predicting carbon emission factors with coordinated multi-time granularity data according to claim 1, wherein the sequentially dividing the second carbon emission data of the plurality of first prediction models into the training set and the validation set comprises:
Sequentially screening the characteristics of the second carbon emission data of the plurality of first prediction models to obtain characteristic screening results, dividing the characteristic screening results into a training set, a verification set and a test set according to a proportion, normalizing the training set, the verification set and the test set according to columns respectively to obtain a normalized training set, a normalized verification set and a normalized test set respectively, and training, verifying and testing according to the normalized training set, the normalized verification set and the normalized test set finally; wherein each first predictive model corresponds to a normalized training set, a normalized validation set, and a normalized test set.
3. The method for predicting carbon emission factors coordinated with multi-time granularity data according to claim 2, wherein the testing according to the normalized test set comprises:
Respectively inputting a plurality of normalized test sets into a corresponding optimized second prediction model, respectively outputting corresponding prediction results according to optimized optimal parameters, inputting a plurality of prediction results into the second prediction model, and outputting total prediction results according to an optimized optimal coordination matrix;
And according to the carbon emission true value of each normalized test set and the total prediction result, performing prediction accuracy evaluation on each second prediction model.
4. The method for predicting carbon emission factors coordinated with multi-time granularity data according to claim 3, wherein said evaluating the prediction accuracy of each second prediction model comprises:
Taking the average absolute percentage error as an evaluation index of the prediction precision of the second prediction model, wherein the evaluation index is expressed as:
，
Wherein, For testing sample sequence number,/>For the time point sequence number of the period to be predicted,/>For normalizing the prediction results of the test set,/>For testing the corresponding carbon emission true value of the sample,/>For/>Number of test set samples of the second predictive model,/>Is the total number of time points of the period to be predicted.
5. A multi-time granularity data coordinated carbon emission factor prediction system, comprising: the method comprises the steps of establishing a prediction model module, establishing a coordination prediction model module and a training module; wherein,
The prediction model establishment module is used for acquiring first carbon emission data of a region to be predicted, and establishing a plurality of first prediction models for predicting carbon emission factors according to time granularity; wherein each time granularity corresponds to a first prediction model;
The coordination prediction model establishing module is used for establishing a time hierarchy matrix according to the time granularity and the period to be predicted, and fusing a plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model;
the training module is used for carrying out parameter updating on the first coordination prediction model according to the first carbon emission data to obtain a trained second coordination prediction model so as to predict carbon emission factors through the second coordination prediction model;
the fusing the plurality of first prediction models according to the time hierarchy matrix to obtain an initial first coordination prediction model comprises the following steps:
Constructing an initial coordination matrix according to the quantity of the carbon emission factor predicted values at the finest time granularity, acquiring an output matrix formed by the output of a plurality of first prediction models, and constructing an initial first coordination prediction model according to the point multiplication sequence of the time hierarchy matrix, the initial coordination matrix and the output matrix;
the method for constructing the initial coordination matrix according to the quantity of the carbon emission factor predicted values under the granularity of the finest time comprises the following steps:
constructing a unit square matrix by taking the quantity of the carbon emission factors predicted value at the finest time granularity as a row and taking the difference between the total quantity of the carbon emission factors predicted value and the quantity of the carbon emission factors predicted value at all time granularity as a column, constructing a zero matrix, and taking a blocking matrix formed by the zero matrix and the unit square matrix as an initial coordination matrix;
the establishing a time hierarchy matrix according to the time granularity and the period to be predicted comprises the following steps:
If the time granularity comprises: years, seasons and months, the time granularity is If the period to be predicted is 12 months in the future, the time hierarchy matrix is expressed as:
Wherein, ，/>；/>I is an identity matrix;
the parameter updating of the first coordination prediction model according to the first carbon emission data comprises:
sequentially dividing second carbon emission data of the plurality of first prediction models into a training set and a verification set according to a proportion; the first carbon emission data includes: second carbon emission data corresponding to each of the first predictive models;
Updating initial parameters and an initial coordination matrix of the first coordination prediction model according to the training set, the verification set and a pre-constructed double-layer optimization model; wherein the bilayer optimization model comprises: a lower layer optimization model and an upper layer optimization model; the lower optimization model optimizes the initial parameters according to the training set to obtain optimal parameters; the upper optimization model optimizes the initial coordination matrix according to the verification set and the optimal parameters;
The double-layer optimization model is expressed as:
，
Wherein, Is an underlying optimization model,/>Expressed as/>Number of training set samples of the first predictive model,/>；/>Is/>Carbon emission true values for the individual training samples; /(I)Is the initial coordination matrix/>Is/>Initial parameter sets of the first prediction models; /(I)Is the prediction result of the training set input to the first coordinated prediction model; /(I)，/>Is the prediction result from the test set to the first coordination prediction model; /(I)Expressed as/>The number of validation set samples of the first predictive model; /(I)Is/>The actual value of carbon emissions of the samples was verified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410323416.0A CN117933493B (en) | 2024-03-21 | 2024-03-21 | Multi-time granularity data coordinated carbon emission factor prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410323416.0A CN117933493B (en) | 2024-03-21 | 2024-03-21 | Multi-time granularity data coordinated carbon emission factor prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117933493A CN117933493A (en) | 2024-04-26 |
| CN117933493B true CN117933493B (en) | 2024-06-07 |
Family
ID=90754110
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410323416.0A Active CN117933493B (en) | 2024-03-21 | 2024-03-21 | Multi-time granularity data coordinated carbon emission factor prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117933493B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925914A (en) * | 2022-02-17 | 2022-08-19 | 宁波莲腾智慧能源科技有限公司 | Energy optimization scheduling method based on eigenspace and electric power trading method thereof |
| CN116822803A (en) * | 2023-08-29 | 2023-09-29 | 北京国网信通埃森哲信息技术有限公司 | Carbon emission data graph construction method, device and equipment based on intelligent algorithm |
| CN117010723A (en) * | 2023-08-21 | 2023-11-07 | 国网陕西省电力有限公司电力科学研究院 | Carbon emission prediction method, system, apparatus, and computer-readable storage medium |
| CN117010541A (en) * | 2022-08-31 | 2023-11-07 | 腾讯科技（深圳）有限公司 | Time sequence prediction method, device and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210272027A1 (en) * | 2020-02-28 | 2021-09-02 | Kinaxis Inc. | Systems, methods and apparatus for hierarchical forecasting |
| CN114519444A (en) * | 2020-11-18 | 2022-05-20 | 英业达科技有限公司 | Hierarchical time sequence prediction method |
-
2024
- 2024-03-21 CN CN202410323416.0A patent/CN117933493B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925914A (en) * | 2022-02-17 | 2022-08-19 | 宁波莲腾智慧能源科技有限公司 | Energy optimization scheduling method based on eigenspace and electric power trading method thereof |
| CN117010541A (en) * | 2022-08-31 | 2023-11-07 | 腾讯科技（深圳）有限公司 | Time sequence prediction method, device and storage medium |
| CN117010723A (en) * | 2023-08-21 | 2023-11-07 | 国网陕西省电力有限公司电力科学研究院 | Carbon emission prediction method, system, apparatus, and computer-readable storage medium |
| CN116822803A (en) * | 2023-08-29 | 2023-09-29 | 北京国网信通埃森哲信息技术有限公司 | Carbon emission data graph construction method, device and equipment based on intelligent algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 时间序列数据挖掘综述;贾澎涛;何华灿;刘丽;孙涛;;计算机应用研究;20071130(11);第15-18页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117933493A (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10606862B2 (en) | Method and apparatus for data processing in data modeling | |
| CN110717671B (en) | Method and device for determining contribution degree of participants | |
| CN102694800A (en) | Gaussian process regression method for predicting network security situation | |
| JP7304698B2 (en) | Water demand forecasting method and system | |
| CN112668822B (en) | Scientific and technological achievement transformation platform sharing system, method, storage medium and mobile phone APP | |
| CN106485348A (en) | A kind of Forecasting Methodology of transaction data and device | |
| Azimi et al. | Applying basic control theory principles to project control: Case study of off-site construction shops | |
| CN106295351A (en) | A kind of Risk Identification Method and device | |
| CN113268883A (en) | Method for predicting corrosion rate of submarine crude oil pipeline based on PCA-ABC-SVM model | |
| Felix et al. | Analysis of training parameters in the ANN learning process to mapping the concrete carbonation depth | |
| Aram et al. | Qualitative and quantitative cost estimation: a methodology analysis | |
| CN112184495A (en) | Low-efficiency land stock monitoring system and analysis platform applying same | |
| CN104408317A (en) | Metallurgy enterprise gas flow interval predicting method based on Bootstrap echo state network integration | |
| CN116776006B (en) | Customer portrait construction method and system for enterprise financing | |
| CN117933493B (en) | Multi-time granularity data coordinated carbon emission factor prediction method and system | |
| Wirawan et al. | Application of data mining to prediction of timeliness graduation of students (a case study) | |
| Wederhake et al. | On the surplus accuracy of data-driven energy quantification methods in the residential sector | |
| CN110827111A (en) | Real-time credit analysis method and system based on dynamic extension model | |
| WO2022222230A1 (en) | Indicator prediction method and apparatus based on machine learning, and device and storage medium | |
| CN117455681A (en) | Service risk prediction method and device | |
| CN114612231A (en) | Stock quantitative trading method and device, terminal device and readable storage medium | |
| CN113011748A (en) | Recommendation effect evaluation method and device, electronic equipment and readable storage medium | |
| Trigkas et al. | Assessing artificial neural networks (ANNS) adequacy against econometric models for decision making approaches in banking industry | |
| CN111612626A (en) | Method and device for preprocessing bond evaluation data | |
| Liang et al. | Hga-lstm: LSTM architecture and hyperparameter search by hybrid GA for air pollution prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |