CN117634678A - Low-carbon park carbon emission prediction method based on actual operation scene - Google Patents
Low-carbon park carbon emission prediction method based on actual operation scene Download PDFInfo
- Publication number
- CN117634678A CN117634678A CN202311579838.6A CN202311579838A CN117634678A CN 117634678 A CN117634678 A CN 117634678A CN 202311579838 A CN202311579838 A CN 202311579838A CN 117634678 A CN117634678 A CN 117634678A
- Authority
- CN
- China
- Prior art keywords
- carbon emission
- carbon
- data
- park
- low
- 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.)
- Pending
Links
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 175
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 145
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 33
- 238000005457 optimization Methods 0.000 claims abstract description 22
- 238000007637 random forest analysis Methods 0.000 claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000005096 rolling process Methods 0.000 claims abstract description 10
- 238000012216 screening Methods 0.000 claims abstract description 4
- 238000000605 extraction Methods 0.000 claims description 21
- 238000004088 simulation Methods 0.000 claims description 14
- 238000005265 energy consumption Methods 0.000 claims description 11
- 238000005259 measurement Methods 0.000 claims description 11
- 238000003066 decision tree Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000009966 trimming Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000002620 method output Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 14
- 238000011160 research Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a low-carbon park carbon emission prediction method based on actual operation scenes, which comprises the steps of constructing an initial carbon emission scene of a low-carbon park to obtain an initial carbon emission prediction model; dynamically training the initial carbon emission prediction model, training the processed data by using a random forest algorithm, and updating the carbon emission prediction model; performing multi-stage rolling optimization on the updated carbon emission prediction model and the updated influence factors, and giving an optimization target; constructing a sample pair by the enhanced data and the original training set data, comparing the similarity between the sample pair by using a similarity function, and calculating a comparison loss; screening important influencing factors according to the contrast loss and the related indexes; the method outputs the carbon prediction result, adopts the low-carbon park carbon emission prediction technology based on the dynamic training set, has better universality and prediction precision, and can well predict the park carbon emission characteristic.
Description
Technical Field
The invention relates to the field of carbon emission prediction, in particular to a low-carbon park carbon emission prediction method based on actual operation scenes.
Background
The park is taken as a basic component unit of urban development, is a main bearing platform and a human activity place of urban economy, and gathers a plurality of population and industry. The campus has multiple forms such as industry garden, education garden, scientific research garden, and quantity is huge. The carbon emission of the park reaches 31% of the total carbon emission of the whole country, wherein the carbon emission of the operation accounts for more than 21%.
Along with the development of the park comprehensive energy system, the development of the park comprehensive energy system tends to be intelligent, comprehensive and fine, the prediction and accounting of the building carbon emission become one of the main carbon reduction measures of the park, and the accuracy of the system operation optimization configuration and the investment and operation scheduling conditions of the whole energy system are directly influenced. The method relates to the counting of a large number of metering devices and data acquisition devices and the analysis and comparison of mass data in the actual operation process of the low-carbon park, and provides interaction trace conditions between the park and municipal information for the park. The analysis process of the carbon emission of the park involves real-time data conversion of a plurality of units, energy media and the like of the park, so that the analysis process is continuously corrected according to data calculation such as carbon measurement, standard coal calculation value and the like, energy form change and enterprise change. How to predict the corresponding carbon emission according to the historical data and the parks without the historical data, so that the basis data can be provided for optimizing the basic energy distribution of the parks and optimizing and controlling the energy transmission and distribution, and the minimum carbon emission of the parks is achieved.
Currently, there is considerable research on content such as carbon emission prediction in parks. For example, the traditional algorithms such as the artificial neural network method based on the statistical regression, and the artificial intelligent algorithm which is represented by intelligent learning are widely applied, such as particle swarm algorithm, deep convolutional neural network, long-short-term memory network, random forest algorithm, and the like. Existing research is typically based on a particular building or type of building and cannot cover the campus level. A small amount of researches establish a specific dynamic building carbon emission simulation model, and cannot be well adapted to the carbon emission characteristics of the building diversity of a park. Because the low-carbon park has higher precision requirements on low-carbon operation indexes, the existing research content cannot adapt to the refined operation requirements of the park. Moreover, the park construction planning belongs to expansibility planning, most of the planning belongs to new buildings or buildings to be operated, the requirements cannot be met through historical data training and prediction, and the prediction precision is poor, so that a low-carbon park carbon emission prediction method based on actual operation scenes is needed.
Disclosure of Invention
The invention aims to provide a low-carbon park carbon emission prediction method based on actual operation scenes.
In order to achieve the above purpose, the invention is implemented according to the following technical scheme:
the invention comprises the following steps:
constructing an initial carbon emission scene of a low-carbon park to obtain an initial carbon emission prediction model;
b, dynamically training the initial carbon emission prediction model, training the processed data by using a random forest algorithm, and updating the carbon emission prediction model;
the random forest algorithm divides historical carbon emission data into a training set and a testing set, data enhancement is carried out on the training set data, and influence factors are extracted;
c, carrying out multi-stage rolling optimization on the updated carbon emission prediction model and the influence factor, and giving an optimization target; constructing a sample pair by the enhanced data and the original training set data, comparing the similarity between the sample pair by using a similarity function, and calculating a comparison loss;
wherein the number of samples is 2N, the number of samples is x, the training set data is i, the similarity function is sim (·), the indication function is ifn (·), and the enhanced data is j;
screening important influencing factors according to the contrast loss and the related indexes; and outputting a carbon prediction result.
Further, the method for constructing the initial carbon emission scene of the low carbon park in the step A comprises the following steps:
1) Collecting carbon emission data of a low-carbon park, wherein the carbon emission data comprises building carbon emission data, production process carbon emission data and traffic carbon emission data;
2) A carbon emission source is identified based on the carbon emission data,
3) And establishing an initial carbon emission prediction model according to the carbon emission data and the carbon emission source.
Further, continuing to preprocess the carbon emission data comprises detecting missing values of the historical carbon emission data, and deleting data of missing points in continuous days.
Further, the specific steps of the random forest algorithm are as follows:
a) Normalization initialization is carried out on the analog data;
b) Drawing n samples by adopting a random forest Bootstrap method, and forming ni sample sets by ni times of sampling to form ni decision trees;
c) Randomly selecting M characteristic values from M characteristics of the ni-th tree to perform node splitting so as to obtain optimal splitting capacity characteristics, and continuously and deeply splitting the decision tree to leaf nodes;
d) Forming random forest by ni decision tree, predicting resultTaking the predicted value of each tree and taking an average value;
e) By passing throughThe operation of the building, the actual measurement data are increased, the training set is dynamic by replacing or trimming the simulation historical data with the actual measurement data of the operation of the building, and the prediction is performed again by using the data of the first three days to obtain a prediction resultFor checking contrast and correcting->
F) Dynamically superposing random forest outputs according to the steps B) to E) to obtainThe final result is the most.
Further, in step C, the multi-stage scrolling gives an optimization objective, the expression of which is:
wherein the loss weight of the ith extraction task is eta i The parameter set of the extraction task head network is phi, and the loss function of the ith extraction task is V i (. Cndot.) the extraction label of the ith extraction task is y i The classification head of the ith extraction task isThe number of extraction tasks is g, the feature parameters of the pre-trained feature extractor are +.>The expression for fine-tuning the use loss function is:
wherein the model predicts that the sample set data features belong to the category with a probability ofClass number s, sampleWhether the set belongs to the ith category y i If the sample set data belongs to the category i, the sample set data is 1, otherwise, the sample set data is 0;
and continuously adjusting parameters until the function value of the loss function is smaller than 0.23, and outputting the data characteristics predicted by the model.
Further, the related indexes include carbon emission intensity, emission amount, growth rate and energy consumption amount.
Further, the method for extracting the influence factors comprises the steps of obtaining a carbon emission data set, analyzing the characteristics of periodic intensity and the like of carbon emission, and extracting the initial influence factors by adopting the methods of principal component analysis and the like to obtain the influence factors.
The beneficial effects of the invention are as follows:
compared with the prior art, the method for predicting the carbon emission of the low-carbon park based on the actual operation scene has the following technical effects:
(1) The invention designs a multi-stage rolling optimization carbon emission prediction technology based on a low-carbon park, which can well predict park carbon emission in different construction stages of the low-carbon park and has certain superiority.
(2) Compared with other deep machine learning, the method has better universality and prediction precision by adopting the low-carbon garden carbon emission prediction technology based on the dynamic training set, and can well predict the carbon emission characteristic of the garden by continuously replacing and continuously correcting the measured data in the first three days.
(3) The regional carbon emission prediction method based on the invention has the advantages of high safety and reliability, simple and convenient operability, large data capacity and strong calling performance, reduces the technical requirements on carbon emission prediction personnel, reduces the huge workload of simulation and simulation, forms a replicable and generalized park carbon emission prediction mode, and has certain universality.
Drawings
FIG. 1 is a flow chart of the steps of the method for predicting carbon emission in a low-carbon park based on actual operation scene;
fig. 2 is a schematic diagram of a prediction curve of a certain time point in an embodiment of a low carbon park carbon emission prediction method based on an actual operation scenario.
Detailed Description
The invention is further described by the following specific examples, which are presented to illustrate, but not to limit, the invention.
The invention discloses a low-carbon park carbon emission prediction method based on actual operation scenes, which comprises the following steps of:
as shown in fig. 1, in the present embodiment, a constructs an initial carbon emission scene of a low carbon park to obtain an initial carbon emission prediction model;
the park planning scenes are researched in stages, so that park construction contents in different stages are obtained, and key carbon emission sources such as buildings, traffic and industry are obtained. The building aspect researches the contents of important building envelope thermal parameters, internal heat interference (personnel, office equipment, illumination), cold and heat source equipment and the like, and related building operation timetables such as personnel's room rate, illumination timetables, office equipment operation timetables and building cold and heat source, domestic hot water and other equipment operation control strategies, so that detailed modeling is performed through dynamic simulation software, and dynamic carbon emission data is obtained; in the traffic aspect, the carbon emission of the fuel automobiles and the charged automobiles in the park is predicted and estimated by researching the equal proportion of the fuel automobiles and the charged automobiles in the park and estimating the number of travelable mileage in the park; and similarly, obtaining industrial carbon emission data.
1) Collecting carbon emission data of a low-carbon park, wherein the carbon emission data comprises building carbon emission data, production process carbon emission data and traffic carbon emission data;
2) A carbon emission source is identified based on the carbon emission data,
3) And establishing an initial carbon emission prediction model according to the carbon emission data and the carbon emission source.
Further, continuing to preprocess the carbon emission data comprises detecting missing values of the historical carbon emission data, and deleting data of missing points in continuous days.
B, dynamically training the initial carbon emission prediction model, training the processed data by using a random forest algorithm, and updating the carbon emission prediction model;
according to the method, a random algorithm forest model is adopted to predict carbon emission of a park, the random algorithm forest model generates a training set through a Bagging method, classified regression trees are used as unit classifiers to be combined into an integrated classifier, and a prediction result is an arithmetic average value of all the integrated classifiers. The dynamic training set takes the simulation data or the estimation data as initial prediction, and the actual operation of the park is continuously provided with data, so that the data content of the original training set is gradually corrected or replaced, and the training set data is more close to the actual project operation effect.
The random forest algorithm comprises the following specific steps:
a) Normalization initialization is carried out on the analog data;
b) Drawing n samples by adopting a random forest Bootstrap method, and forming ni sample sets by ni times of sampling to form ni decision trees;
c) Randomly selecting M characteristic values from M characteristics of the ni-th tree to perform node splitting so as to obtain optimal splitting capacity characteristics, and continuously and deeply splitting the decision tree to leaf nodes;
d) Forming random forest by ni decision tree, predicting resultTaking the predicted value of each tree and taking an average value;
e) Through the operation of the building, the actual measurement data is increased, the training set is dynamic by replacing or trimming the simulation historical data with the actual measurement data of the operation of the building, and the prediction is performed again by using the data of the previous three days to obtain a prediction resultFor checking contrast and correcting->
F) Dynamically superposing random forest outputs according to the steps B) to E) to obtainThe final result is the most.
The random forest algorithm divides historical carbon emission data into a training set and a testing set, data enhancement is carried out on the training set data, and influence factors are extracted;
c, carrying out multi-stage rolling optimization on the updated carbon emission prediction model and the influence factor, and giving an optimization target; the expression of the optimization objective is:
wherein the loss weight of the ith extraction task is eta i The parameter set of the extraction task head network is phi, and the loss function of the ith extraction task is V i (. Cndot.) the extraction label of the ith extraction task is y i The classification head of the ith extraction task isThe number of extraction tasks is g, the feature parameters of the pre-trained feature extractor are +.>The expression for fine-tuning the use loss function is:
wherein the model predicts that the sample set data features belong to the category with a probability ofThe number of categories is s, and whether the sample set belongs to the ith category is y i If the sample set data belongs to the category i, the sample set data is 1, otherwise, the sample set data is 0;
and continuously adjusting parameters until the function value of the loss function is smaller than 0.23, and outputting the data characteristics predicted by the model.
Constructing a sample pair by the enhanced data and the original training set data, comparing the similarity between the sample pair by using a similarity function, and calculating a comparison loss;
wherein the number of samples is 2N, the number of samples is x, the training set data is i, the similarity function is sim (·), the indication function is ifn (·), and the enhanced data is j;
screening important influencing factors according to the contrast loss and the related indexes; and outputting a carbon prediction result.
The related indexes comprise carbon emission intensity, emission quantity, growth rate and energy consumption, and the method for extracting the influence factors comprises the steps of acquiring a carbon emission data set, analyzing the characteristics of periodic intensity of carbon emission and the like, and extracting the initial influence factors by adopting methods of principal component analysis and the like to obtain the influence factors.
The carbon emission prediction in the initial stage of the park is realized through the step A and the step C, and for the carbon emission prediction of the park in different later stages, the performance, the operation mode and the like of the park equipment are continuously updated by continuously updating new buildings, new processes or new communication modes and adopting a simulation or estimation method, and the original park operation data is subjected to supplementary correction according to the step I and the step II. Thereby realizing the multi-stage rolling optimization carbon emission prediction of the park.
As shown in fig. 2, the following is a case of a low carbon park carbon emission prediction method based on an actual operation scenario:
in the embodiment, actual measurement verification is carried out on a certain low-carbon park, the park is mainly built by hospitals, and system tests are carried out on the park. Because the carbon emission factors do not have uniformity at present and are required to be set according to actual conditions of projects, the actual measurement is verified by electricity consumption, and specific carbon emission can be converted by multiplying the corresponding carbon emission factors. By comparing the database simulation prediction data with the field actual measurement data, the reasonable accuracy of the prediction method is verified.
When the initial carbon emission scene is constructed, the historical conventional building energy consumption data, the hospital special equipment energy consumption data and the traffic energy consumption data are determined by analyzing the historical conventional building energy consumption data, the hospital special equipment energy consumption data and the traffic energy consumption data. The campus then trains the historical data using a machine learning algorithm, creating an initial carbon emission prediction model.
In the dynamic training stage, the park realizes the real-time monitoring of carbon emission sources such as energy consumption, hospital electricity consumption, transportation and the like through the Internet of things technology, and uses the processed data to perform model training and verification so as to ensure the accuracy of a prediction result.
In the multi-stage rolling optimization carbon prediction stage, optimization targets such as total carbon emission, carbon emission intensity and the like are set, and then operation data of different stages are collected for comparison analysis. According to the operation data of different stages, the park carries out rolling optimization on the carbon emission prediction model, and the prediction accuracy is improved. Finally, the park evaluates the optimized carbon emission prediction result to ensure that the requirements of the optimization target are met
The power consumption of the system is smaller by comparing the field actual measurement data with the simulation prediction data, and the change trend is basically consistent. The validity of the prediction method is effectively verified.
According to the method, a multi-stage rolling optimization prediction scene of the carbon emission data of the park is built, so that the carbon emission data of the low-carbon park is predicted, actual operation control optimization of the park is guided, for the initial planning situation of the park, a scene simulation method is adopted to simulate the operation situation of the park building, optimize equipment configuration and operation scheme, and perform real-time dynamic simulation on the energy consumption data of related industries all year by adopting an analogy method, so that the operation situation of the park in the initial stage is obtained, the carbon emission situation of the park in the initial stage is obtained, and then a model is built through basic algorithms such as random forests, artificial neural networks and the like, and the carbon emission of the park in future time is simulated and calculated; the simulation data is then gradually used to replace or trim the previous simulation data based on the actual operational data, thereby performing the stage of building data simulation. And similarly, in the subsequent expansion stage, the carbon emission data iteration of the park is carried out through the same method, and the multi-stage data and the equipment optimization control mode are updated. Thereby realizing the multi-stage rolling optimization prediction of the carbon emission in the park.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (7)
1. The low-carbon park carbon emission prediction method based on the actual operation scene is characterized by comprising the following steps of:
constructing an initial carbon emission scene of a low-carbon park to obtain an initial carbon emission prediction model;
b, dynamically training the initial carbon emission prediction model, training the processed data by using a random forest algorithm, and updating the carbon emission prediction model;
the random forest algorithm divides historical carbon emission data into a training set and a testing set, data enhancement is carried out on the training set data, and influence factors are extracted;
c, carrying out multi-stage rolling optimization on the updated carbon emission prediction model and the influence factor, and giving an optimization target; constructing a sample pair by the enhanced data and the original training set data, comparing the similarity between the sample pair by using a similarity function, and calculating a comparison loss;
wherein the number of samples is 2N, the number of samples is x, the training set data is i, the similarity function is sim (·), the indication function is ifn (·), and the enhanced data is j;
screening important influencing factors according to the contrast loss and the related indexes; and outputting a carbon prediction result.
2. The actual operating scenario-based low carbon park carbon emission prediction method according to claim 1, wherein the method of constructing the initial low carbon park carbon emission scenario in step a comprises:
1) Collecting carbon emission data of a low-carbon park, wherein the carbon emission data comprises building carbon emission data, production process carbon emission data and traffic carbon emission data;
2) A carbon emission source is identified based on the carbon emission data,
3) And establishing an initial carbon emission prediction model according to the carbon emission data and the carbon emission source.
3. The actual operating scenario-based low-carbon park carbon emission prediction method of claim 2, wherein continuing the preprocessing of the carbon emission data comprises detecting missing values of the historical carbon emission data, and deleting data of missing points on consecutive days.
4. The method for predicting carbon emission in a low-carbon park based on actual operation scene as set forth in claim 1, wherein the random forest algorithm comprises the following specific steps:
a) Normalization initialization is carried out on the analog data;
b) Drawing n samples by adopting a random forest Bootstrap method, and forming ni sample sets by ni times of sampling to form ni decision trees;
c) Randomly selecting M characteristic values from M characteristics of the ni-th tree to perform node splitting so as to obtain optimal splitting capacity characteristics, and continuously and deeply splitting the decision tree to leaf nodes;
d) A random forest is formed by ni decision trees, and the result y is predicted 1 Taking the predicted value of each tree and taking an average value;
e) Through the operation of the building, the actual measurement data is increased, the training set is dynamic through replacing or trimming the simulation historical data with the actual measurement data of the operation of the building, and meanwhile, the prediction is performed again by using the data of the previous three days to obtain a prediction result y 2 For checking contrast and correcting y 1 ；
F) Dynamically superposing random forest outputs according to the steps B) -E) to obtain y 3 The final result is the most.
5. The actual operating scenario-based low-carbon park carbon emission prediction method according to claim 1, wherein in step C, the multi-stage roll gives an optimization objective, the expression of which is:
wherein the loss weight of the ith extraction task is eta i The parameter set of the extraction task head network is phi, and the loss function of the ith extraction task is V i (. Cndot.) the extraction label of the ith extraction task is y i The classification head of the ith extraction task is w αi (. Cndot.) the number of extraction tasks is g, and the feature parameters of the pre-trained feature extractor areThe expression for fine-tuning the use loss function is:
wherein the model predicts that the sample set data features belong to the category with a probability ofThe number of categories is s, and whether the sample set belongs to the ith category is y i If the sample set data belongs to the category i, the sample set data is 1, otherwise, the sample set data is 0;
and continuously adjusting parameters until the function value of the loss function is smaller than 0.23, and outputting the data characteristics predicted by the model.
6. The method for predicting carbon emissions in a low-carbon park based on an actual operation scenario of claim 1, wherein the related indicators include carbon emission intensity, emission amount, growth rate and energy consumption amount.
7. The method for predicting carbon emissions in a low-carbon park based on an actual operation scenario of claim 1, wherein the method for extracting the influencing factors comprises obtaining a carbon emission data set, analyzing characteristics such as cycle intensity of carbon emissions, and extracting initial influencing factors by adopting methods such as principal component analysis to obtain the influencing factors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311579838.6A CN117634678A (en) | 2023-11-24 | 2023-11-24 | Low-carbon park carbon emission prediction method based on actual operation scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311579838.6A CN117634678A (en) | 2023-11-24 | 2023-11-24 | Low-carbon park carbon emission prediction method based on actual operation scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117634678A true CN117634678A (en) | 2024-03-01 |
Family
ID=90036926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311579838.6A Pending CN117634678A (en) | 2023-11-24 | 2023-11-24 | Low-carbon park carbon emission prediction method based on actual operation scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117634678A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117910657A (en) * | 2024-03-14 | 2024-04-19 | 杭州阿里云飞天信息技术有限公司 | Prediction method, model training method, computing device, storage medium, and program product for carbon shift factor |
-
2023
- 2023-11-24 CN CN202311579838.6A patent/CN117634678A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117910657A (en) * | 2024-03-14 | 2024-04-19 | 杭州阿里云飞天信息技术有限公司 | Prediction method, model training method, computing device, storage medium, and program product for carbon shift factor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Song et al. | Hourly heat load prediction model based on temporal convolutional neural network | |
| CN110705743B (en) | New energy consumption electric quantity prediction method based on long-term and short-term memory neural network | |
| CN103853106B (en) | A kind of energy consumption Prediction Parameters optimization method of building energy supplied equipment | |
| CN111080032A (en) | Load prediction method based on Transformer structure | |
| Liu et al. | Study on environment-concerned short-term load forecasting model for wind power based on feature extraction and tree regression | |
| CN113554466B (en) | Short-term electricity consumption prediction model construction method, prediction method and device | |
| CN117634678A (en) | Low-carbon park carbon emission prediction method based on actual operation scene | |
| CN105868887A (en) | Building comprehensive energy efficiency analysis method based on subentry measure | |
| CN117013527A (en) | Distributed photovoltaic power generation power prediction method | |
| CN109242136A (en) | A kind of micro-capacitance sensor wind power Chaos-Genetic-BP neural network prediction technique | |
| CN115034485A (en) | Wind power interval prediction method and device based on data space | |
| CN113033110A (en) | Important area personnel emergency evacuation system and method based on traffic flow model | |
| CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
| Qu et al. | Research on short‐term output power forecast model of wind farm based on neural network combination algorithm | |
| Zhang et al. | Ensemble learning-based approach for residential building heating energy prediction and optimization | |
| CN116826745B (en) | Layered and partitioned short-term load prediction method and system in power system background | |
| CN112365082A (en) | Public energy consumption prediction method based on machine learning | |
| CN112132344A (en) | Short-term wind power prediction method based on similar day and FRS-SVM | |
| CN117033923A (en) | Method and system for predicting crime quantity based on interpretable machine learning | |
| CN113837486B (en) | RNN-RBM-based distribution network feeder long-term load prediction method | |
| Sarmas et al. | Baseline energy modeling for improved measurement and verification through the use of ensemble artificial intelligence models | |
| CN110210642A (en) | A kind of city electric energy substitution amount prediction technique and device | |
| Lan et al. | Urban air quality prediction based on space-time optimization LSTM model | |
| Ali et al. | Deep learning networks for short term load forecasting | |
| CN114372691A (en) | Electric energy substitution potential estimation method based on holographic perception |
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 |