CN111612031A - Regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search - Google Patents
Regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search Download PDFInfo
- Publication number
- CN111612031A CN111612031A CN202010257783.7A CN202010257783A CN111612031A CN 111612031 A CN111612031 A CN 111612031A CN 202010257783 A CN202010257783 A CN 202010257783A CN 111612031 A CN111612031 A CN 111612031A
- Authority
- CN
- China
- Prior art keywords
- load
- database
- building
- variable
- clustering
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013461 design Methods 0.000 claims abstract description 21
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 238000012360 testing method Methods 0.000 claims description 14
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 14
- 230000005611 electricity Effects 0.000 claims description 13
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 5
- 230000037237 body shape Effects 0.000 claims description 2
- 238000011160 research Methods 0.000 abstract description 4
- 238000005265 energy consumption Methods 0.000 abstract description 2
- 238000005259 measurement Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003050 experimental design method Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- 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
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search, which belongs to the field of regional building load prediction, and is characterized in that loads of different state buildings in different regions under different situations (orientation, thermal parameters of an enclosure structure, indoor load intensity, building use time tables and the like) are simulated through a building energy consumption calculation engine EnergyPlus widely recognized by the industry to form a database based on a forward model, and the database contains a large number of typical state building scenes, so that the actual engineering design condition can be better covered; on the basis of the database, a high-dimensional spatial clustering neighbor searching method is adopted to realize rapid prediction of the annual hourly load of the region. And overlapping the annual time-by-time cold and heat loads of all the ecological buildings in the area to obtain the annual peak cold load and peak heat load for guiding the model selection work of the equipment. The method is simple and convenient, and has positive significance to the theoretical research field and the practical application field.
Description
Technical Field
The invention belongs to the field of regional building load prediction, and relates to a regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search.
Background
In order to respond to government calls and adapt to rapid development of social economy and steady promotion of urbanization process, more and more projects are planned and constructed in a new district form and are provided with regional energy centers, and the regional centralized energy supply system becomes one of the preferred methods for realizing regional energy supply targets and sustainable development due to the fact that the regional centralized energy supply efficiency is higher than that of distributed cold and heat sources, the initial investment is small, and the maintenance and management are easy.
The regional loads in regional energy supply comprise a cold and hot load, an electric load and a domestic hot water load superposed on each building. The electric load prediction is of great importance to power grid planning, power generation equipment model selection, electric power overhaul and the like, and the cold and hot load prediction mainly influences the design and equipment model selection of a regional heat and cold supply system. In the system design and model selection in the planning stage, a certain margin coefficient is usually considered for various loads to ensure the safe operation of the system. However, the research on the established energy station project discovers that the system equipment of the energy station operated at present is configured too much, the load rate is very low, and the original design purpose of energy conservation, high efficiency, environmental protection and economic operation cannot be realized. The most fundamental reason is that the load prediction accuracy is poor, the design load is large, and the load on the energy supply side of the energy station is far larger than the load on the actual user demand side.
Load prediction has important significance for regional energy supply, and scholars at home and abroad also provide a large number of prediction methods which can be roughly divided into three methods, namely an area load index method, a forward model method and a data driving method. The area load index method obtains a static load result and cannot reflect the time dynamic characteristics of the load in the region; the forward modeling method has long calculation time and high requirement on the professional of a modeler; the data-driven approach is completely based on statistics and does not take into account the physical properties of the building. And because of the limitation of the output quantity of the model, the existing research rarely discusses the annual time-by-time load, but replaces the annual time-by-time load/peak load by the typical daily time-by-time load/peak load, and the replacement causes the difference between the calculation result of the regional load peak value and the real result to be larger, thereby wasting the system capacity.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search, which realizes the annual time-by-time load prediction of regional building groups by designing a variable-load database of each business state of each region and combining the high-dimensional spatial clustering neighbor search.
The technical scheme adopted by the invention for solving the problems is as follows: a regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search is characterized by comprising the following steps:
variable-load database design;
designing a prediction algorithm based on high-dimensional spatial clustering neighbor search;
and combining the database with the regional scale load prediction of the algorithm.
A load database, namely a common typical building state (office/hotel/business/hospital/data center) variable of five cities of Beijing, Shanghai, Wuhan, Chongqing and Guangzhou, is established to obtain a cold, hot, electricity and domestic hot water load database, wherein the form of the database is sql.
Furthermore, parameters selected in the load prediction model in the variable-load database design stage are figure coefficients, comprehensive heat transfer coefficients, summer indoor set temperature, winter indoor set temperature, personnel density, equipment power density, illumination density and a timetable in sequence, wherein the first seven parameters are in four levels, and the timetable is in three levels;
preferably, a test design method based on a quasi-Monte Carlo method in a number theory, namely uniform design, is adopted to carry out high-dimensional space filling design, so that test points are uniformly filled with a parameter space in a test range, and a Monte Carlo method is used for extracting 10% of the test points as variables, namely a load calculation example set.
Further, sampling and generating a variable-load example set for each business state of each city, and using the variable-load example set as a variable set in a load database, wherein each variable in the database represents a building in the business state reality of the city corresponding to the database.
Preferably, the load results in the cold, hot, electricity and domestic hot water load databases of all the cities in all the business states are calculated by energy plus, the load is 8760 hours per hour all year round, and the units are W/square meter.
Preferably, the load calculation and the database establishment of each business state of each city are realized by writing Python codes to automatically calculate energy plus and automatically establish the database.
Furthermore, Kmeans clustering is carried out on the variables in the variable-load database, the variable set is clustered into k clusters, the city is different from the business state, and the value of k is also different.
Preferably, the metric selected in the aforementioned Kmeans cluster is a block distance, and the block distance is:
further, for Kmeans clustering, the Kmeans clustering is divided into k clusters in advance, k is preset to be 1, then k is k +1 in each clustering, k is plotted with the clustering performance to obtain a 'elbow' line with the clustering performance changing along with k, and when k is increased and the increase rate of the clustering performance is reduced, k at the moment is considered as the optimal classification number of the variables in the database.
Preferably, the performance index of the Kmeans cluster is DB index and Dunn index;
DB Index (Davies-Bouldin Index, DBI for short):
dunn Index (Dunn Index, DI for short):
avg (C) corresponds to the average distance between samples within a cluster, diam (C) corresponds to the farthest distance between samples within a cluster C, dmin(Ci,Cj) Corresponding to the distance between the cluster and the closest sample of the cluster, dcen(Ci,Cj) Corresponding to the distance of the cluster from the cluster center point; the DBI compares the intra-cluster spacing to the cluster center spacing, and the DI compares the distance between the nearest samples in a cluster to the distance between the farthest sample points in a cluster.
Preferably, the parameters (form coefficient, comprehensive heat transfer coefficient, summer indoor set temperature, winter indoor set temperature, personnel density, equipment power density, lighting density and timetable) of the building to be predicted are taken as high-dimensional variables, and the annual time-by-time load prediction of cold, heat, electricity and domestic hot water is carried out by using a database and an algorithm of corresponding business states of the area where the building to be predicted is located.
Further, the variable to be predicted and N cluster centers in the corresponding database are subjected to block distance calculation and comparison, the cluster center closest to the variable to be predicted is found out, a KNN neighbor algorithm is carried out in the cluster, 3 neighbor variables closest to the variable to be predicted are found out, and the annual hourly load prediction of cold, heat, electricity and domestic hot water of the variable to be predicted is obtained through distance weighting.
Preferably, in the KNN nearest neighbor algorithm, the distance weighting function is selected to be an inverse proportional function:
furthermore, the annual hourly load prediction of cold, heat, electricity and domestic hot water of regional building groups can be carried out.
Preferably, parameters and building areas of all buildings in the area are collected to obtain variables representing all buildings, a database corresponding to each variable is found according to cities and states, algorithm prediction is carried out to obtain annual time-by-time loads corresponding to each variable, each load is multiplied by the corresponding area to obtain annual time-by-time total loads of all buildings, and the annual time-by-time total loads of all buildings are summed to obtain annual time-by-time loads of the building group in the area.
Compared with the prior art, the invention has the following advantages and effects: the invention simulates the loads of different industrial buildings under different situations (orientation, thermal parameters of an enclosure structure, indoor load intensity, a building use time schedule and the like) in different areas through a building energy consumption calculation engine EnergyPlus widely accepted by the industry to form a database based on a forward model, and the database contains a large number of typical industrial building situations and can better cover the actual engineering design situation; on the basis of the database, a high-dimensional spatial clustering neighbor searching method is adopted to realize rapid prediction of the annual hourly load of the region. And overlapping the annual time-by-time cold and heat loads of all the ecological buildings in the area to obtain the annual peak cold load and peak heat load for guiding the model selection work of the equipment. The method is simple and convenient, and has positive significance to the theoretical research field and the practical application field.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail below by way of examples with reference to the accompanying drawings, which are illustrative of the present invention and are not to be construed as limiting the present invention.
Examples are given.
As shown in fig. 1, in this embodiment, a method for predicting dynamic load of a regional building based on high-dimensional spatial clustering neighbor search includes the following steps:
102, generating a variable-load example set;
105, selecting high-dimensional space measurement;
106, designing a high-dimensional spatial clustering neighbor searching algorithm;
and step 107, forecasting the time-by-time load of the region.
Specifically, step 101, selecting load prediction model variables;
selecting parameters in a load prediction model in a variable load database design stage as a body shape coefficient, a comprehensive heat transfer coefficient, summer indoor set temperature, winter indoor set temperature, personnel density, equipment power density, lighting density and a timetable in sequence, wherein the first seven parameters take four average levels in a value range, and the standard range of the building in the area is expanded outwards by 10% to be determined as the value range of each parameter; the timetable is on three levels, wherein 1 is the first to friday 8:00-19:00, 2 is the first to friday 8:00-22:00, and 3 is the first to friday 8:00-22: 00.
102, generating a variable-load example set;
in this step, the 8 parameters in step 101 are orthogonalized, each group of 8 parameter combinations can be regarded as a test point, 49152 test points are generated by using Python, a test design method based on a quasi-monte carlo method in number theory-uniform design is adopted to perform high-dimensional space filling design, so that the test points uniformly fill parameter space in a test range, and a monte carlo method is used to extract 10% of the test points as a variable-load example set.
in the step, a building information file corresponding to the variable is generated in batches by using Python, the file type is idf, and meanwhile, the cold, hot, electricity and domestic hot water loads corresponding to the variable are obtained by using Python multithreading running EnergyPlus, so that a variable-load data set is formed.
the method comprises the steps of using Python to operate Mysql to quickly call and store data, enabling a storage engine to be INNODB, automatically reading a variable-load data set, writing the variable-load data set into an automatically established database in Mysql, establishing five-state databases of five cities in total, and reading variable and load information into the corresponding databases according to the city and state where the variable is located. Each database stores data in a sub-table mode, and consists of a general table and a large number of sub-tables, wherein the general table comprises 8 parameter information of all variables in the database and serial numbers of the variables; the branch table comprises the year-by-year load information of cold, heat, electricity and domestic hot water corresponding to the variables. When searching, the building is searched according to the basic information of the general table, and the load data is read from the hour-by-hour load sub-table of the building according to the external key guidance.
105, selecting high-dimensional space measurement;
in this step, the high-dimensional spatial clustering neighbor search method described in step 106 based on various high-dimensional spatial metrics is used to perform metric selection exploration, compare the predicted load under this metric with the actual load, select the spatial metric with the best prediction performance, and finally select the metric as the block distance.
106, designing a high-dimensional spatial clustering neighbor searching algorithm;
in this step, Kmeans clustering is performed on variables in the load database, which are variables of each business state of each city, and the variable set is clustered into k clusters, wherein the cities are different from the business states, the values of k are also different, and the measurement based on clustering is the measurement selected in the step 105.
Specifically, the method is characterized by comprising the steps of predetermining k clusters, presetting k as 1, then, when clustering is carried out each time, mapping k to clustering performance to obtain a 'elbow' line with clustering performance changing along with k, and when k is increased and the increase rate of the clustering performance is reduced, considering k at the moment as the optimal classification number of variables in the database; wherein the clustering performance indexes are DB index and Dunn index.
Regarding parameters (form coefficient, comprehensive heat transfer coefficient, summer indoor set temperature, winter indoor set temperature, personnel density, equipment power density, lighting density and timetable) of a building to be predicted as a high-dimensional variable, calculating and comparing the distance between the variable to be predicted and N cluster centers in a corresponding database, finding out the cluster center closest to the variable to be predicted, carrying out KNN nearest neighbor algorithm in the cluster, finding out 3 nearest neighbor variables away from the variable to be predicted, and obtaining year-round time-by-year load prediction of cold, heat, electricity and domestic hot water of the variable to be predicted through distance weighting; wherein the distance weighting function is an inverse proportional function.
collecting parameters and building areas of all buildings in the area to obtain variables representing all the buildings, finding a database corresponding to each variable according to cities and states, performing algorithm prediction to obtain annual time-by-time loads corresponding to each variable, multiplying each load by a corresponding area to obtain annual time-by-time total loads of all the buildings, and then summing the annual time-by-time total loads of all the buildings to obtain the annual time-by-time loads of the building group in the area.
The above detailed description of cities, states, parameter selections, experimental design methods, sampling methods, clustering indicators, search algorithms, and distance weighting functions is provided to facilitate an understanding and application of the present invention by those of ordinary skill in the art. It will be readily apparent to those skilled in the art that various modifications can be made to these embodiments and the generic principles described herein may be applied to other embodiments without the use of the inventive faculty. Therefore, the present invention is not limited to the specific description herein, and those skilled in the art should, in light of the present disclosure, appreciate that many changes and modifications can be made without departing from the spirit and scope of the invention.
Claims (10)
1. A regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search is characterized by comprising the following steps: the method comprises the steps of variable-load database design, prediction algorithm design based on high-dimensional space clustering neighbor search, and region scale load prediction combining a database and an algorithm.
2. The regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search is characterized in that a load database which is a common typical building state variable of an urban is established to obtain a cold, hot, electricity and domestic hot water load database, wherein the database is in the form of sql.
3. The method for predicting the dynamic load of the regional building based on the high-dimensional spatial clustering nearest neighbor search as claimed in claim 2, wherein the parameters selected in the load prediction model in the variable-load database design stage are body shape coefficient, comprehensive heat transfer coefficient, indoor set temperature in summer, indoor set temperature in winter, personnel density, equipment power density, lighting density and timetable in sequence, wherein the first seven parameters are at four levels, and the timetable is at three levels.
4. The method for predicting the dynamic load of the regional building based on the high-dimensional spatial clustering neighbor search as claimed in claim 1, wherein a test design method based on a quasi-Monte Carlo method in a number theory-uniform design is adopted for high-dimensional space filling design, so that test points uniformly fill parameter spaces in a test range, and a Monte Carlo method is adopted to extract 10% of the test points as a variable-load example set.
5. The method as claimed in claim 2, wherein the model of each city is sampled and generated as a set of variables in a load database, and each variable in the database represents a building in the model reality of the city corresponding to the database.
6. The regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search is characterized in that load results in cold, hot, electricity and domestic hot water load databases of various urban models are all calculated by EnergyPlus, the load is a load which is 8760 hours per hour all year round, and the unit is W/square meter; the automatic calculation of EnergyPlus and the automatic establishment of the database are realized by compiling Python codes.
7. The method for predicting the dynamic load of the regional building based on the high-dimensional spatial clustering neighbor search of claim 2, wherein Kmeans clustering is performed on variables in a variable-load database, a variable set is clustered into k clusters, cities are different from businesses, and values of k are different;
the metric selected in the aforementioned Kmeans clusters is the block distance, which is:
for Kmeans clustering, the Kmeans clustering is divided into k clusters in advance, k is preset to be 1, then k is k +1 when clustering is carried out each time, k and clustering performance are plotted to obtain a 'elbow' line with clustering performance changing along with k, and when k is increased and the clustering performance increasing rate is reduced, k at the moment is considered to be the optimal classification number of variables in a database;
the performance indexes of the Kmeans cluster are DB index and Dunn index;
DB index:
dunn index:
avg (C) corresponds to the average distance between samples within a cluster, diam (C) corresponds to the farthest distance between samples within a cluster C, dmin(Ci,Cj) Corresponding to the distance between the cluster and the closest sample of the cluster, dcen(Ci,Cj) Corresponding to the distance of the cluster from the cluster center point; the DBI compares the intra-cluster spacing to the cluster center spacing, and the DI compares the distance between the nearest samples in a cluster to the distance between the farthest sample points in a cluster.
8. The regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search is characterized in that parameters of a building to be predicted are regarded as high-dimensional variables, and a database and an algorithm of corresponding states of an area where the building to be predicted is located are used for carrying out annual time-by-time load prediction on cold, heat, electricity and domestic hot water.
9. The regional building dynamic load prediction method based on high-dimensional spatial clustering nearest neighbor search according to claim 8, characterized in that the calculation and comparison of the block distance are performed on the variables to be predicted and N cluster centers in the corresponding database, the cluster center closest to the variable to be predicted is found, the KNN nearest neighbor algorithm is performed in the cluster, the 3 nearest neighbor variables to the variable to be predicted are found, and the annual hourly load prediction of cold, hot, electricity and domestic hot water of the variables to be predicted is obtained through distance weighting; in the KNN nearest neighbor algorithm, the distance weighting function is selected as an inverse proportional function:
10. the dynamic load prediction method for the regional buildings based on the high-dimensional spatial clustering nearest neighbor search is characterized in that the annual hourly load prediction of cold, heat, electricity and domestic hot water of the regional building groups is carried out; collecting parameters and building areas of all buildings in the area to obtain variables representing all the buildings, finding a database corresponding to each variable according to cities and states, performing algorithm prediction to obtain annual hourly loads corresponding to each variable, multiplying each load by a corresponding area to obtain annual hourly total loads of all the buildings, and then summing the annual hourly total loads of all the buildings to obtain the annual hourly loads of the building group in the area.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010257783.7A CN111612031A (en) | 2020-04-03 | 2020-04-03 | Regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010257783.7A CN111612031A (en) | 2020-04-03 | 2020-04-03 | Regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111612031A true CN111612031A (en) | 2020-09-01 |
Family
ID=72201607
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010257783.7A Pending CN111612031A (en) | 2020-04-03 | 2020-04-03 | Regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111612031A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115293461A (en) * | 2022-09-28 | 2022-11-04 | 四川中电启明星信息技术有限公司 | Building heating carbon emission prediction method and system |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706778A (en) * | 2008-08-27 | 2010-05-12 | 王阳 | Method for predicting power load |
CN106022509A (en) * | 2016-05-07 | 2016-10-12 | 国网浙江省电力公司经济技术研究院 | Power distribution network space load prediction method taking region and load property dual differences into consideration |
CN107730027A (en) * | 2017-09-13 | 2018-02-23 | 深圳供电规划设计院有限公司 | A kind of load forecasting method and device based on region building property |
CN108304978A (en) * | 2018-05-08 | 2018-07-20 | 国网江西省电力有限公司经济技术研究院 | A kind of mid-term Electric Power Load Forecast method based on data clusters theory |
CN108491969A (en) * | 2018-03-16 | 2018-09-04 | 国家电网公司 | Spatial Load Forecasting model building method based on big data |
CN110334366A (en) * | 2019-03-14 | 2019-10-15 | 华电电力科学研究院有限公司 | A kind of instantaneous cooling load prediction method of building based on the Monte Carlo Method with Latin Hypercube Sampling |
CN110458340A (en) * | 2019-07-25 | 2019-11-15 | 天津大学 | Building air conditioning refrigeration duty Autoregressive Prediction Method based on pattern classification |
CN110472892A (en) * | 2019-08-27 | 2019-11-19 | 上海电气分布式能源科技有限公司 | Build flow of the people, cool and thermal power load evaluation method and system, equipment and medium |
CN110717619A (en) * | 2019-09-11 | 2020-01-21 | 国网浙江省电力有限公司经济技术研究院 | Multi-scale space-time load prediction method and system for bottom-up power distribution network |
-
2020
- 2020-04-03 CN CN202010257783.7A patent/CN111612031A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706778A (en) * | 2008-08-27 | 2010-05-12 | 王阳 | Method for predicting power load |
CN106022509A (en) * | 2016-05-07 | 2016-10-12 | 国网浙江省电力公司经济技术研究院 | Power distribution network space load prediction method taking region and load property dual differences into consideration |
CN107730027A (en) * | 2017-09-13 | 2018-02-23 | 深圳供电规划设计院有限公司 | A kind of load forecasting method and device based on region building property |
CN108491969A (en) * | 2018-03-16 | 2018-09-04 | 国家电网公司 | Spatial Load Forecasting model building method based on big data |
CN108304978A (en) * | 2018-05-08 | 2018-07-20 | 国网江西省电力有限公司经济技术研究院 | A kind of mid-term Electric Power Load Forecast method based on data clusters theory |
CN110334366A (en) * | 2019-03-14 | 2019-10-15 | 华电电力科学研究院有限公司 | A kind of instantaneous cooling load prediction method of building based on the Monte Carlo Method with Latin Hypercube Sampling |
CN110458340A (en) * | 2019-07-25 | 2019-11-15 | 天津大学 | Building air conditioning refrigeration duty Autoregressive Prediction Method based on pattern classification |
CN110472892A (en) * | 2019-08-27 | 2019-11-19 | 上海电气分布式能源科技有限公司 | Build flow of the people, cool and thermal power load evaluation method and system, equipment and medium |
CN110717619A (en) * | 2019-09-11 | 2020-01-21 | 国网浙江省电力有限公司经济技术研究院 | Multi-scale space-time load prediction method and system for bottom-up power distribution network |
Non-Patent Citations (5)
Title |
---|
刘思等: "考虑地域差异的配电网空间负荷聚类及一体化预测方法", vol. 41, no. 3, pages 70 - 75 * |
周鸣争等: "《中医药统计学与软件应用》", 中国中医药出版社, pages: 503 - 505 * |
徐伟等: "基于建筑特性的冷热电动态负荷预测方法研究", vol. 46, no. 9, pages 64 - 69 * |
林芳等: "基于均衡KNN算法的电力负荷短期并行预测", vol. 51, no. 10, pages 88 - 94 * |
高靖等: "基于特征聚类的多能源***负荷预测方法研究", vol. 37, no. 2, pages 232 - 236 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115293461A (en) * | 2022-09-28 | 2022-11-04 | 四川中电启明星信息技术有限公司 | Building heating carbon emission prediction method and system |
CN115293461B (en) * | 2022-09-28 | 2022-12-16 | 四川中电启明星信息技术有限公司 | Building heating carbon emission prediction method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xu et al. | Smart energy systems: A critical review on design and operation optimization | |
CN111291963B (en) | Park comprehensive energy system planning method for coordinating economy and reliability | |
CN109919480B (en) | Three-layer target energy Internet planning method and equipment | |
CN110363353A (en) | The optimization design and dispatching method and system of a kind of Distributed Integration energy resource system | |
CN112348276A (en) | Comprehensive energy system planning optimization method based on multiple elements and three levels | |
CN111709638B (en) | Combined cooling heating power system construction method and system based on graph theory and equivalent electric method | |
Liu et al. | Optimization of intelligent heating ventilation air conditioning system in urban building based on BIM and artificial intelligence technology | |
CN115115193A (en) | Low-carbon analysis and optimization planning method for industrial park | |
CN111612031A (en) | Regional building dynamic load prediction method based on high-dimensional spatial clustering neighbor search | |
CN117332971A (en) | Regional comprehensive energy planning method and system considering city morphology | |
CN115879652B (en) | Hierarchical collaborative planning method and device for energy network, electronic equipment and storage medium | |
CN116644941A (en) | Industrial energy planning system based on Internet of things | |
CN115222227A (en) | Database-based carbon emission determination method, system, medium and device | |
CN113222216B (en) | Cold and hot electric load prediction method, device and system | |
CN110765726A (en) | Intelligent generation system of energy network planning map | |
CN110135619B (en) | Method and system for predicting medium-and-long-term electric heating requirements | |
Meng et al. | Data center cooling load prediction and analysis based on weather data clustering | |
CN113902167A (en) | Comprehensive intelligent energy system configuration optimization method and device considering station house investment | |
CN111898846A (en) | Method for extracting power consumption missing data | |
Cao et al. | Optimal capacity configuration of battery storage system for zero energy office building on campus | |
Lian et al. | Research on the comprehensive evaluation method and index system of power network planning | |
CN118195279B (en) | Heating management scheduling method and system based on renewable energy intelligent coupling | |
Guo et al. | Research and application of energy consumption optimization design of electromechanical system based on genetic algorithms | |
CN113505939B (en) | Energy system random planning method and system considering uncertainty factors | |
Wang et al. | Energy-saving optimisation method for green space planning of urban gardens based on artificial bee colony algorithm |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200901 |