CN109508743B - KNN improved algorithm-based method for repairing abnormal energy consumption data of lighting socket - Google Patents
KNN improved algorithm-based method for repairing abnormal energy consumption data of lighting socket Download PDFInfo
- Publication number
- CN109508743B CN109508743B CN201811342483.8A CN201811342483A CN109508743B CN 109508743 B CN109508743 B CN 109508743B CN 201811342483 A CN201811342483 A CN 201811342483A CN 109508743 B CN109508743 B CN 109508743B
- Authority
- CN
- China
- Prior art keywords
- data
- characteristic line
- energy consumption
- slope
- utilization characteristic
- 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
- 238000005265 energy consumption Methods 0.000 title claims abstract description 43
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 18
- 230000005611 electricity Effects 0.000 claims abstract description 33
- 238000005286 illumination Methods 0.000 claims abstract description 4
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 abstract description 13
- 238000012545 processing Methods 0.000 abstract description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for repairing abnormal energy consumption data of an illumination socket based on a KNN improved algorithm, and belongs to the technical field of data processing of building energy consumption monitoring platforms. The basic idea of the algorithm is shown in an abstract chart, historical data can be roughly divided into different energy utilization modes, the energy utilization modes of data to be repaired are compared, and the data in the energy utilization modes are used for repairing missing data. The data repairing process by using the KNN improved algorithm comprises the following steps: s1, establishing an energy utilization mode according to the historical data; s2, aiming at any energy utilization mode, establishing an electricity utilization characteristic database; and S3, selecting data with similar energy modes from the historical data to repair the missing data based on the KNN improved algorithm. The method has the advantages that the slope of the electricity utilization characteristic line is combined, the KNN improved algorithm is used for repairing abnormal energy consumption data by using data with similar energy consumption modes in historical data, and the quality of the building energy consumption data is greatly improved.
Description
Technical Field
The invention relates to a data processing method, in particular to a method for repairing abnormal data of building energy consumption monitoring based on a KNN improved algorithm, and belongs to the technical field of data processing of building energy consumption monitoring platforms.
Background
In recent years, a public building energy consumption monitoring platform is established, and the realization of building energy consumption item measurement is an important content for the construction of a public building energy supervision system. At present, building energy consumption monitoring platforms of different scales are established in many provinces and cities of China, and rich data is accumulated. However, due to some technical obstacles, the overall quality of building monitoring data is low, and data abnormal problems such as data loss, data mutation and the like occur, through investigation and analysis, the abnormal data proportion of the existing building energy consumption monitoring platform is generally up to 20%, so that the energy consumption monitoring data is far from the real energy consumption of the building, and a large amount of energy consumption data cannot be fully utilized. With the continuous development of green buildings in China, people not only acquire more and more types and larger data quantity of building energy consumption data, but also have higher and higher requirements on data quality. In order to improve the data quality of the building energy consumption monitoring platform, in recent years, many scientific researchers at home and abroad focus on the problems existing in the energy consumption monitoring platform data, and the missing or abnormal data of the platform is solved and corrected by adopting methods such as a data mining technology, cluster analysis and the like. The invention discloses a construction energy consumption monitoring abnormal data repairing method based on a KNN improved algorithm, which aims at solving the problem of repairing abnormal data. The energy consumption of building electricity is generally divided into four items of energy consumption of lighting sockets, energy consumption of heating ventilation air conditioners, energy consumption of power equipment and energy consumption of special electricity. The patent provides a method for repairing abnormal energy consumption data of the lighting socket based on a KNN improved algorithm by adopting a big data analysis method aiming at the power consumption characteristics of the energy consumption of the lighting socket so as to improve the data quality of a building energy consumption monitoring platform.
Disclosure of Invention
The invention aims to provide a method for repairing abnormal energy consumption data of an illumination socket based on a KNN improved algorithm, which can effectively improve the quality of building energy consumption monitoring data.
The technical scheme of the invention is as follows:
a method for repairing abnormal energy consumption data of an illumination socket based on a KNN improved algorithm comprises the following steps:
s1, calculating the slope of the historical data electricity utilization characteristic line: aiming at a certain type of energy consumption per unit time, calculating the slope of the historical data electricity utilization characteristic line per unit time, assuming that m historical data electricity utilization characteristic lines per unit time are provided, and for any historical data electricity utilization characteristic line per unit time i, the electricity consumptions at the j moment and the j +1 moment are respectively Ei,jAnd Ei,j+1The slope of the electricity utilization characteristic line from the j time to the j +1 time is Li,jThe calculation formula of (2) is as follows:
wherein, tjIs the unit time length;
s2, calculating the slope of the electricity utilization characteristic line of the abnormal data in unit time: calculating the abnormal data position by the same method as that of step S1Slope l of electricity utilization characteristic line at 24 hoursj,j=0,1,2,…,23;
S3, searching sample data: slope sequence l based on abnormal datajAnd historical data set slope sequence Li,jSearching the hour energy consumption data of the day with the minimum sequencing k days distance from the historical data set (m days); slope l of electrical characteristic line for abnormal datajThe slope of the electricity utilization characteristic line of the ith calendar history data is Li,jEuclidean distance p ofiThe calculation formula is as follows:
s4, determining a weighting coefficient: using the ratio of the inverse Euclidean distance to the sum of inverse Euclidean distances of k days as a weighting coefficient for weighting and summing abnormal data; the weighting coefficient formula is:
wherein q istThe weight p of the t day in k days nearest to the Euclidean distance calculated by the KNN algorithmtEuclidean distance of day t;
s5, repairing abnormal data: obtaining the repairing data of the abnormal data according to the k-day sample data and the weighting coefficient; assuming that the moment j is abnormal data, the data recovery formula is as follows:
wherein ejFor calculated repair energy consumption data, tjIs a unit time length, Ln,jIs the slope of the electricity utilization characteristic line of the nth day in the k-day calendar history data at the time j, qnThe weighting factor corresponding to the day.
The method is also suitable for repairing abnormal electricity utilization data of other power equipment with relatively regular electricity utilization loads.
Compared with the prior art, the method has the advantages that the method for repairing the abnormal data of the building energy consumption based on the KNN improved algorithm is provided, and the repairing quality of the data of the building energy consumption is greatly improved.
Drawings
FIG. 1 is a flow chart of an abnormal data repair operation.
FIG. 2 is a schematic diagram of the calculation of the slope of the electrical characteristic line.
Fig. 3 is a schematic diagram of sample data search.
Detailed Description
The following detailed description of the invention refers to the accompanying drawings that illustrate specific embodiments of the invention.
Referring to the attached figure 1, the method for repairing abnormal building energy consumption data based on the KNN improved algorithm comprises the following steps:
s1, referring to FIG. 2, calculating the slope of the historical data electricity utilization characteristic line in unit time, assuming a total of m historical data electricity utilization characteristic broken lines in unit time, and regarding any historical data electricity utilization characteristic line i in unit time, the electricity consumption at the j th moment and the j +1 th moment are respectively Ei,j、Ei,j+1The slope of the electricity utilization characteristic line from the j th moment to the j +1 th moment is Li,jThe calculation formula of (2) is as follows:
wherein t isjIs a unit time length.
S2, calculating the slope l of the electricity utilization characteristic line in the unit time of 24 hours in which the abnormal data are positioned by adopting the method of S1j(j=0,1,2,…,23);
S3, referring to the attached figure 3, calculating the slope sequence l of the electricity utilization characteristic line where the abnormal data are located by using an Euclidean distance calculation formulajAnd historical data set slope sequence Li,jAnd searching energy consumption data with the minimum day-hour distance of k days from the historical data set (m days). Slope l of electrical characteristic line for abnormal datajThe slope of the electricity utilization characteristic line of the ith calendar history data is Li,jEuclidean distance p ofiThe calculation formula is as follows:
and S4, using the ratio of the inverse Euclidean distance to the sum of inverse Euclidean distances of k days as a weighting coefficient for weighting and summing the abnormal data. The weighting coefficient formula is:
wherein q istThe weight p of the t day in k days nearest to the Euclidean distance calculated by the KNN algorithmtThe Euclidean distance of day t.
And S5, finally, obtaining the repairing data of the abnormal data according to the k-day sample data and the weighting coefficient. Assuming that the moment j is abnormal data, the data recovery formula is as follows:
wherein ejFor calculated repair energy consumption data, tjIs a unit time length, Ln,jIs the slope of the electricity utilization characteristic line of the nth day in the k-day calendar history data at the time j, qnThe weighting factor corresponding to the day.
Claims (1)
1. A method for repairing abnormal energy consumption data of an illumination socket based on a KNN improved algorithm is characterized by comprising the following steps:
s1, calculating the slope of the electricity utilization characteristic line of the historical data: aiming at a certain type of energy consumption per unit time, calculating the slope of the power utilization characteristic line of historical data per unit time, and assuming a common power consumptionmA historical data electricity utilization characteristic line in unit time, for any historical data electricity utilization characteristic line in unit timeiOf 1 atjTime and firstj+The electricity consumption at 1 moment is respectivelyE i, j AndE i, j+1 then it is firstjFrom time to timej The slope of the power utilization characteristic line of the historical data in the +1 moment isThe calculation formula of (2) is as follows:
Wherein,t jis the unit time length;
s2, calculating the slope of the electricity utilization characteristic line of the abnormal data in unit time: calculating the slope of the abnormal data electricity utilization characteristic line by the same method as the step S1l j,;
S3, searching sample data: slope of electricity utilization characteristic line based on abnormal datal j Slope of power utilization characteristic line from historical dataL i,jEuclidean distance (similarity) of (m days), finding a ranking from the historical data setkHour energy consumption data for the day from the minimum day; slope of electrical characteristic line for abnormal datal j Slope of power utilization characteristic line from historical dataL i,jEuclidean distance ofp iThe calculation formula is as follows:
S4, determining a weighting coefficient: using the inverse Euclidean distance andkthe ratio of the sum of the inverse Tian-Euclidean distances is used as a weighting coefficient for weighting and summing abnormal data; the weighting coefficient formula is:
Wherein,q t calculated for KNN algorithm as nearest Euclidean distancekIn the middle of the daytThe weight of the day is calculated,p tis as followstThe Euclidean distance of day;
s5, repairing abnormal data: according tokObtaining the repairing data of the abnormal data by the sample data of the day and the weighting coefficient; suppose thatjThe time is abnormal data, and the data repair formula is as follows:
Whereine jIn order to calculate the repair energy consumption data,t jin terms of the length of the unit of time,L n,jis composed ofkThe first in the calendar history datanThe electricity utilization characteristic line of the sky isjThe slope of the time of day is,q nthe weighting factor corresponding to the day.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811342483.8A CN109508743B (en) | 2018-11-13 | 2018-11-13 | KNN improved algorithm-based method for repairing abnormal energy consumption data of lighting socket |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811342483.8A CN109508743B (en) | 2018-11-13 | 2018-11-13 | KNN improved algorithm-based method for repairing abnormal energy consumption data of lighting socket |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109508743A CN109508743A (en) | 2019-03-22 |
CN109508743B true CN109508743B (en) | 2022-03-04 |
Family
ID=65748204
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811342483.8A Active CN109508743B (en) | 2018-11-13 | 2018-11-13 | KNN improved algorithm-based method for repairing abnormal energy consumption data of lighting socket |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109508743B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111625525B (en) * | 2020-05-26 | 2023-05-26 | 哈尔滨工业大学 | Environment data repairing/filling method and system |
CN111723085B (en) * | 2020-06-28 | 2023-06-23 | 山东建筑大学 | Method and system for cleaning electric energy consumption data of public building air conditioner |
CN112765549B (en) * | 2021-01-20 | 2022-10-21 | 大连理工大学 | KNN-Matrix method for repairing abnormal data of building energy consumption monitoring |
CN114253953A (en) * | 2021-11-18 | 2022-03-29 | 国电南京自动化股份有限公司 | Power distribution network multidimensional data processing method and system based on cluster analysis |
CN117969946B (en) * | 2024-04-01 | 2024-06-14 | 山东普源电力工程有限公司 | Lighting system energy consumption detection method and system |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104199961A (en) * | 2014-09-19 | 2014-12-10 | 北京建筑技术发展有限责任公司 | Data mining based public building energy consumption monitoring platform data processing method |
GB201509507D0 (en) * | 2015-06-02 | 2015-07-15 | Passivsystems Ltd | Fault detection in energy generation arrangements |
CN108228862A (en) * | 2018-01-12 | 2018-06-29 | 上海市建筑科学研究院 | A kind of method for processing abnormal data and system of energy consumption monitoring platform |
CN108694473A (en) * | 2018-06-15 | 2018-10-23 | 常州瑞信电子科技有限公司 | Building energy consumption prediction technique based on RBF neural |
-
2018
- 2018-11-13 CN CN201811342483.8A patent/CN109508743B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN109508743A (en) | 2019-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109508743B (en) | KNN improved algorithm-based method for repairing abnormal energy consumption data of lighting socket | |
Wang et al. | A review of approaches to low-carbon transition of high-rise residential buildings in China | |
CN103912966B (en) | A kind of earth source heat pump refrigeration system optimal control method | |
Ma et al. | Building energy research in Hong Kong: a review | |
CN110796307B (en) | Distributed load prediction method and system for comprehensive energy system | |
CN107220472B (en) | Energy consumption instrument based on building area and energy efficiency ratio and energy consumption evaluation method | |
CN110851892B (en) | Park level comprehensive energy system aided design method and system | |
CN105760963A (en) | Tourist area resident household electricity prediction method | |
CN116992548B (en) | Design optimization method and system for building and regional comprehensive energy system | |
CN112907074A (en) | Energy efficiency sensitive index detection method and system for comprehensive energy system user | |
CN111738498A (en) | Comprehensive energy system robust planning method and system | |
CN116205425A (en) | Low-carbon park cold-hot electric load prediction method based on typical database | |
CN109146284B (en) | Accounting system and method for comprehensive utilization amount and scale of renewable energy sources of urban building | |
Cheng et al. | A study of urban form and the integration of energy supply technologies | |
CN112257964B (en) | Load-intensive urban intelligent park demand aggregation modeling method | |
CN109359134B (en) | Method for identifying energy consumption implicit abnormal data of lighting socket based on data mining | |
Medrano et al. | Economics and climate change emissions analysis of a bioclimatic institutional building with trigeneration and solar support | |
CN111126863A (en) | System and method for evaluating passive energy and active energy supply of public institution | |
CN112712213B (en) | Method and system for predicting deep migration learning energy consumption of concentrated air conditioning house | |
CN116224161A (en) | Judgment method based on polarity reverse connection error wiring abnormality of voltage transformer | |
CN113488990B (en) | Micro-grid optimal scheduling method based on improved bat algorithm | |
CN114548533A (en) | Design method and device for dynamic load of building cooling, heating and power, acquisition method for dynamic load in area, computer and storage medium | |
Wu et al. | Field measurement research on indoor physical environment in natural ventilation season of rural housings in northeast and southeast China | |
CN117217094B (en) | Photovoltaic optimal configuration method, system and equipment for urban ground and roof | |
Asmawi et al. | Review on Building Energy Management Systems by Retrofitting Existing Buildings with Green Technology |
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 | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20231218 Address after: Room 2205, 2 / F, 56a22, Torch Road, Dalian hi tech Industrial Park, 116000, Liaoning Province Patentee after: Dalian Qunzhi Technology Co.,Ltd. Address before: 116024 No. 2 Ling Road, Ganjingzi District, Liaoning, Dalian Patentee before: DALIAN University OF TECHNOLOGY |
|
TR01 | Transfer of patent right |