CN112994002A - Electric movable load identification method - Google Patents
Electric movable load identification method Download PDFInfo
- Publication number
- CN112994002A CN112994002A CN202110548655.2A CN202110548655A CN112994002A CN 112994002 A CN112994002 A CN 112994002A CN 202110548655 A CN202110548655 A CN 202110548655A CN 112994002 A CN112994002 A CN 112994002A
- Authority
- CN
- China
- Prior art keywords
- load
- data
- curve
- electric
- identification
- 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.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2310/00—The network for supplying or distributing electric power characterised by its spatial reach or by the load
- H02J2310/70—Load identification
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for identifying a movable electric load, which comprises the following steps: cleaning and supplementing the acquired data; carrying out normalization processing on the cleaned data, and constructing a three-dimensional vector matrix R (M, T, P) of date, 24-hour time and amplitude; wherein M is the number of days of the load curve; t is the time of 24 hours corresponding to the data, and P represents the amplitude; carrying out daily clustering analysis calculation on the three-dimensional vectors of the daily load curves of the electric equipment, and calculating to obtain N classifications; taking the clustering center of each classification as a typical characteristic curve of the electric equipment; comparing the actual load curve of each day in M days with the corresponding typical characteristic curve; if the actual load curve of each day in M days has mutation compared with the corresponding typical characteristic curve on the curve, judging that the identification object is a movable load; identification of other loads or consumers. The invention can realize real-time and accurate recognition of the movable load in a non-invasive scene.
Description
Technical Field
The invention relates to the technical field of movable load identification of a power demand side, in particular to a movable load identification method of power.
Background
In order to ensure dynamic balance between power generation (power supply) and power consumption (power demand) during operation of the power system, the supply side regulates the output of power generation, and the demand side regulates the electrical load. In order to realize the adjustment of the power load, market policies or products such as power demand response, real-time electricity price market trading, virtual power plants, power-assisted market, energy region comprehensive scheduling and the like are generated.
Taking Demand Response (Demand Response) as an example, it encourages users to reduce or divert power usage during peak periods of power consumption by regulating power rates or other financial rewards. For the power supply side, the corresponding implementation of the demand can reduce the load pressure during the peak period of power utilization and effectively balance the supply-demand relation in the power system. For the user, participation in demand response can generally reduce the electricity charge overhead.
One of the main issues in the demand response model is how much power usage can be reduced or shifted. We assume that the presence of a discontinuity in the power usage curve reflects that the user has used a large electrical consumer. Although the service time of some large-scale electric devices is difficult to transfer, such as electric furnaces for cooking, large lamps, industrial production line machine tools, welding machines, commercial building lights and the like, a large part of the service time can be transferred, such as washing machines, water boilers, dish washing machines, industrial air compressors, non-industrial air conditioners and the like. It is desirable to find these large appliances through the power usage profile to estimate the transferable power usage during peak periods of power usage by the customer. After the transferable electric quantity in the power utilization peak period is estimated, a demand response policy can be formulated more reasonably.
Thus, it is readily apparent that in a similar demand response policy, the estimation of consumer usage of large electrical consumers becomes a key to calculating transferable power usage in a power peak system. The method can estimate the consumption of the large-scale electric appliances of the user without the need of identifying the movable load of the electric power. The power load data can enable power consumers to know the power consumption of various electric equipment in different time periods in more detail, the identification of movable loads can make reasonable energy-saving plans according to specific requirements, the use of the electric equipment is adjusted, the power consumption can be reduced, the electricity fee expenditure is reduced, meanwhile, the electric power company can be helped to know the load composition of the power system more truly, the load power consumption is standardized, the service time of various loads is reasonably arranged, the utilization efficiency of a power grid is improved, the investment of the power system is reduced, and the running grid loss of the system is reduced.
Disclosure of Invention
1. The technical problem to be solved is as follows:
in view of the above technical problems, the present invention provides a power movable load identification method, which identifies whether the equipment use time can be transferred or not according to the historical power load curve of the power equipment, and evaluates the transferable power consumption. Support is provided for developing power demand response, power market transaction, virtual power plants, power-assisted markets and comprehensive scheduling of energy areas.
2. The technical scheme is as follows:
a movable electric load identification method is characterized in that: the method comprises the following steps:
the method comprises the following steps: acquiring and obtaining data; acquiring historical operating data of electric equipment recorded by an electric power automatic monitoring system; the historical operation data comprises active power, reactive power, current, voltage, power factor and electricity consumption data; the recording interval of the data is in the order of minutes, including between 1 and 60 minutes; the time span of the historical operation data is K times of the period in which the repeated change of the load can be observed, and K is an integer greater than or equal to 1; the electric power automatic monitoring system comprises an SCADA system, an electricity utilization information acquisition system, an energy consumption monitoring system and a power distribution monitoring system; performing vector superposition on the loads of a plurality of preset electric equipment according to the load identification requirement, and synthesizing into a combined object; the movable load identification method of the combined object is an electric device.
Step two: data cleaning and complementing: taking an electric device as an identification object, firstly checking each historical data record of the identification object, and identifying the record with data missing; secondly, the data relationships of power, current, voltage, power factor and electric quantity at the same time are checked by applying ohm law, kirchhoff circuit law and basic power equation, and the validity of the data is verified; if the data are abnormal, the data are removed, and meanwhile, the data are identified to be missing; and finally, filling up the missing data records.
Step three: forming a normalized daily load curve three-dimensional vector: carrying out normalization processing on the data cleaned in the second step, and constructing a three-dimensional vector matrix R (M, T, P) of date, 24-hour time and amplitude; wherein M is the number of days of the load curve; t is the time of 24 hours corresponding to the data, and P represents the amplitude.
Step four: load characteristic identification; carrying out daily clustering analysis calculation on the three-dimensional vectors of the daily load curves of the electric equipment, and calculating to obtain N classifications; wherein N is an integer much less than M; taking the clustering center of each classification as a typical characteristic curve of the electric equipment, namely a load characteristic; the load of each day in M days corresponds to one of N classifications; wherein the clustering analysis comprises K-means, Gaussian mixture or spectral clustering methods.
Step five: and (3) detecting sudden change of the daily load curve: comparing the actual load curve of each day in M days with the corresponding typical characteristic curve; and if the actual load curve of each day in M days has mutation compared with the corresponding typical characteristic curve on the curve, judging the identification object as the movable load.
Step six: the size of the movable load of the recognition object and the corresponding time are acquired.
Step seven: and repeating the steps from two to six for the identification of other loads or electric equipment.
Furthermore, the supplementing method in the step two adopts the integration of a historical same-proportion load comparison reference method and a least square interpolation method, and the integration mode is the integration according to the weight; the historical comparable reference load is automatically identified according to the load cycle.
Further, the judgment criteria of mutation in the fifth step are specifically: taking a typical electricity utilization curve as an expected value mu, the actual load on the day is X, and the number of data sampling points per day is N, then utilizing a mean square error formula:
calculating the mean square error of the clustering category and the time of each identification object as a function; and if the difference value obtained by subtracting the corresponding clustering center from the electricity consumption value of the identification object at a certain time is larger than a preset threshold value, judging that the identification object is a movable load.
Further, the mutation judgment in the fifth step adopts a difference percentage detection method; the difference percentage detection method takes a typical power utilization curve as a reference value, and detects the occurrence time of the large-scale electric appliance by comparing the difference percentage with a preset threshold value; the difference percentage calculation formula is as follows:
percent difference = (true value-cluster center)/cluster center.
3. Has the advantages that:
(1) the movable load identification data of the invention is derived from the existing electric power automatic monitoring system, and can realize real-time and accurate identification of the movable load in a non-invasive scene.
(2) According to the method, the daily power utilization behaviors of all users are characterized according to historical operating data of a single power utilization equipment load or a plurality of power utilization equipment combined loads, and the clustering center of each cluster is taken as a typical power utilization curve, so that the method has good generalization capability, and the accuracy of load identification is ensured.
(3) The load identification method used by the invention is simple and practical, occupies less storage space and has short identification time when carrying out load identification, and the real-time performance of the load identification can be further improved.
Drawings
FIG. 1 is a flow chart of the method.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings.
The method for recognizing the movable electric load as shown in the attached figure 1 is characterized in that: the method comprises the following steps:
the method comprises the following steps: acquiring and obtaining data; acquiring historical operating data of electric equipment recorded by an electric power automatic monitoring system; the historical operation data comprises active power, reactive power, current, voltage, power factor and electricity consumption data; the recording interval of the data is in the order of minutes, including between 1 and 60 minutes; the time span of the historical operation data is K times of the period in which the repeated change of the load can be observed, and K is an integer greater than or equal to 1; the electric power automatic monitoring system comprises an SCADA system, an electricity utilization information acquisition system, an energy consumption monitoring system and a power distribution monitoring system; performing vector superposition on the loads of a plurality of preset electric equipment according to the load identification requirement, and synthesizing into a combined object; the movable load identification method of the combined object is an electric device.
Step two: data cleaning and complementing: taking an electric device as an identification object, firstly checking each historical data record of the identification object, and identifying the record with data missing; secondly, the data relationships of power, current, voltage, power factor and electric quantity at the same time are checked by applying ohm law, kirchhoff circuit law and basic power equation, and the validity of the data is verified; if the data are abnormal, the data are removed, and meanwhile, the data are identified to be missing; and finally, filling up the missing data records.
Step three: forming a normalized daily load curve three-dimensional vector: carrying out normalization processing on the data cleaned in the second step, and constructing a three-dimensional vector matrix R (M, T, P) of date, 24-hour time and amplitude; wherein M is the number of days of the load curve; t is the time of 24 hours corresponding to the data, and P represents the amplitude.
Step four: load characteristic identification; carrying out daily clustering analysis calculation on the three-dimensional vectors of the daily load curves of the electric equipment, and calculating to obtain N classifications; wherein N is an integer much less than M; taking the clustering center of each classification as a typical characteristic curve of the electric equipment, namely a load characteristic; the load of each day in M days corresponds to one of N classifications; wherein the clustering analysis comprises K-means, Gaussian mixture or spectral clustering methods.
Step five: and (3) detecting sudden change of the daily load curve: comparing the actual load curve of each day in M days with the corresponding typical characteristic curve; and if the actual load curve of each day in M days has mutation compared with the corresponding typical characteristic curve on the curve, judging the identification object as the movable load.
Step six: the size of the movable load of the recognition object and the corresponding time are acquired.
Step seven: and repeating the steps from two to six for the identification of other loads or electric equipment.
Furthermore, the supplementing method in the step two adopts the integration of a historical same-proportion load comparison reference method and a least square interpolation method, and the integration mode is the integration according to the weight; the historical comparable reference load is automatically identified according to the load cycle.
Further, the judgment criteria of mutation in the fifth step are specifically: taking a typical electricity utilization curve as an expected value mu, the actual load on the day is X, and the number of data sampling points per day is N, then utilizing a mean square error formula:
calculating the mean square error of the clustering category and the time of each identification object as a function; and if the difference value obtained by subtracting the corresponding clustering center from the electricity consumption value of the identification object at a certain time is larger than a preset threshold value, judging that the identification object is a movable load.
Further, the mutation judgment in the fifth step adopts a difference percentage detection method; the difference percentage detection method takes a typical power utilization curve as a reference value, and detects the occurrence time of the large-scale electric appliance by comparing the difference percentage with a preset threshold value; the difference percentage calculation formula is as follows:
percent difference = (true value-cluster center)/cluster center.
The specific embodiment is as follows:
in this embodiment, a process of identifying the movable load is given by taking a 96-point daily load curve of the n-day electrical load of an identified object, i.e. one load electricity reading every 15 minutes as an example.
Inputting the processed data, wherein the power utilization curve data of the identified object is
And each row in the data matrix records the load reading of every 15 minutes between 0 and 24 points on a certain day corresponding to the identification object, and m is the number of days contained in the data.
Intermediate results: (1) typical curve clustering data of the identified electric equipment, namely mx 3, wherein each row in the data matrix corresponds to an identification object id, a date and a clustering id; (2) the data is typically clustered using an electrical curve,
each row in the data matrix corresponds to the load number of a certain cluster in each 15 minutes in 24 hours, n is the cluster number, n is the load number of the cluster<<m。
Based on the method, the following steps can be output: and outputting the original power consumption, the electricity consumption of the class center, the difference value and the mean square error corresponding to each moment of each user by taking 15 minutes as a minimum time unit.
In actual operation, the method is generally applied to identify and display the movable load, and a daily load bar graph of N points per day is drawn. And movable load amount exists at each acquisition time point, the upper part of the movable load amount is displayed by intense color mixing with strong contrast, and the upper part of the movable load amount accounts for the ratio. After the load shift to the daily load curve peak, the load peak horizontal lines before and after the shift are shown.
The movable load display can be carried out according to the requirement and the monthly scale, and meanwhile, the difference value between the annual load electric quantity curve of the user and the electricity consumption of the clustering center is calculated in an accumulated mode, so that the annual total electric quantity of the movable load of the equipment can be obtained.
Although the present invention has been described with reference to the preferred embodiments, it should be understood that various changes and modifications can be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (4)
1. A movable electric load identification method is characterized in that: the method comprises the following steps:
the method comprises the following steps: acquiring and obtaining data; acquiring historical operating data of electric equipment recorded by an electric power automatic monitoring system; the historical operation data comprises active power, reactive power, current, voltage, power factor and electricity consumption data; the recording interval of the data is in the order of minutes, including between 1 and 60 minutes; the time span of the historical operation data is K times of the period in which the repeated change of the load can be observed, and K is an integer greater than or equal to 1; the electric power automatic monitoring system comprises an SCADA system, an electricity utilization information acquisition system, an energy consumption monitoring system and a power distribution monitoring system; performing vector superposition on the loads of a plurality of preset electric equipment according to the load identification requirement, and synthesizing into a combined object; the movable load identification method of the combined object is an electric device;
step two: data cleaning and complementing: taking an electric device as an identification object, firstly checking each historical data record of the identification object, and identifying the record with data missing; secondly, the data relationships of power, current, voltage, power factor and electric quantity at the same time are checked by applying ohm law, kirchhoff circuit law and basic power equation, and the validity of the data is verified; if the data are abnormal, the data are removed, and meanwhile, the data are identified to be missing; finally, the missing data records are filled up;
step three: forming a normalized daily load curve three-dimensional vector: carrying out normalization processing on the data cleaned in the second step, and constructing a three-dimensional vector matrix R (M, T, P) of date, 24-hour time and amplitude; wherein M is the number of days of the load curve; t is the time of 24 hours corresponding to the data, and P represents the amplitude;
step four: load characteristic identification; carrying out daily clustering analysis calculation on the three-dimensional vectors of the daily load curves of the electric equipment, and calculating to obtain N classifications; wherein N is an integer much less than M; taking the clustering center of each classification as a typical characteristic curve of the electric equipment, namely a load characteristic; the load of each day in M days corresponds to one of N classifications; wherein the clustering analysis comprises a K-means, Gaussian mixture or spectral clustering method;
step five: and (3) detecting sudden change of the daily load curve: comparing the actual load curve of each day in M days with the corresponding typical characteristic curve; if the actual load curve of each day in M days has mutation compared with the corresponding typical characteristic curve on the curve, judging that the identification object is a movable load;
step six: acquiring the size and the corresponding time of the movable load of the identification object;
step seven: and repeating the steps from two to six for the identification of other loads or electric equipment.
2. An electric movable load identification method according to claim 1, characterized in that: the supplementing method in the step two adopts the integration of two methods, namely a historical same-proportion load comparison reference method and a least square method interpolation method, and the integration mode is integration according to weight; the historical comparable reference load is automatically identified according to the load cycle.
3. An electric movable load identification method according to claim 1, characterized in that: the judgment standard of the mutation in the fifth step is specifically as follows: taking a typical electricity utilization curve as an expected value mu, the actual load on the day is X, and the number of data sampling points per day is N, then utilizing a mean square error formula:
calculating the mean square error of the clustering category and the time of each identification object as a function; and if the difference value obtained by subtracting the corresponding clustering center from the electricity consumption value of the identification object at a certain time is larger than a preset threshold value, judging that the identification object is a movable load.
4. An electric movable load identification method according to claim 1, characterized in that: judging mutation in the fifth step by adopting a difference percentage detection method; the difference percentage detection method takes a typical power utilization curve as a reference value, and detects the occurrence time of the large-scale electric appliance by comparing the difference percentage with a preset threshold value; the difference percentage calculation formula is as follows:
percent difference = (true value-cluster center)/cluster center.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110548655.2A CN112994002B (en) | 2021-05-20 | 2021-05-20 | Electric movable load identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110548655.2A CN112994002B (en) | 2021-05-20 | 2021-05-20 | Electric movable load identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112994002A true CN112994002A (en) | 2021-06-18 |
| CN112994002B CN112994002B (en) | 2021-08-24 |
Family
ID=76337043
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110548655.2A Active CN112994002B (en) | 2021-05-20 | 2021-05-20 | Electric movable load identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112994002B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115459270A (en) * | 2022-11-03 | 2022-12-09 | 西安国智电子科技有限公司 | Method and device for configuring urban peak electricity consumption, computer equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110628A (en) * | 2019-04-24 | 2019-08-09 | 华为技术有限公司 | A kind of detection method and detection device of frequency synthesizer deterioration |
| US10816537B2 (en) * | 2016-06-23 | 2020-10-27 | Roche Sequencing Solutions, Inc. | Formation and calibration of nanopore sequencing cells |
-
2021
- 2021-05-20 CN CN202110548655.2A patent/CN112994002B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10816537B2 (en) * | 2016-06-23 | 2020-10-27 | Roche Sequencing Solutions, Inc. | Formation and calibration of nanopore sequencing cells |
| CN110110628A (en) * | 2019-04-24 | 2019-08-09 | 华为技术有限公司 | A kind of detection method and detection device of frequency synthesizer deterioration |
Non-Patent Citations (1)
| Title |
|---|
| 闫安: "城市智能交通动态预测模型的研究及应用", 《中国优秀硕士学位论文全文数据库工程科技II辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115459270A (en) * | 2022-11-03 | 2022-12-09 | 西安国智电子科技有限公司 | Method and device for configuring urban peak electricity consumption, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112994002B (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116646933B (en) | Big data-based power load scheduling method and system | |
| Dallinger et al. | Integration of intermittent renewable power supply using grid-connected vehicles–A 2030 case study for California and Germany | |
| CN107340492A (en) | Electric power meter failure analysis methods with scene anticipation are excavated based on big data | |
| JP2015529926A (en) | Method and system for estimating recoverable utility revenues | |
| CN114004296A (en) | Method and system for reversely extracting monitoring points based on power load characteristics | |
| CN111817299B (en) | Fuzzy reasoning-based intelligent identification method for line loss rate abnormal cause of power distribution station | |
| CN106780125B (en) | method for calculating abnormal collection emergency degree based on average monthly power consumption | |
| CN107798475A (en) | A kind of formulating method and device of Demand-side load adjustment scheme | |
| Yao et al. | A novel category-specific pricing strategy for demand response in microgrids | |
| CN114865627A (en) | Power distribution method, device and equipment based on supply and demand relationship | |
| CN112288303A (en) | Method and device for determining line loss rate | |
| CN112994002B (en) | Electric movable load identification method | |
| CN111310854A (en) | Low false alarm rate electricity stealing detection secondary screening method based on electricity load fingerprint | |
| CN105914735B (en) | A kind of power distribution network economic load tidal current computing method | |
| CN115684809A (en) | User side power consumption test system based on virtual power plant | |
| CN110675018B (en) | Line loss anomaly analysis and closed-loop management method based on expert rules | |
| CN114140176B (en) | Adjustable capacity prediction method and device for load aggregation platform | |
| CN106682817B (en) | Judgment method for collecting abnormal emergency degree | |
| Grigoras et al. | Customer classification and load profiling using data from Smart Meters | |
| CN114841832B (en) | Power consumer portrait label establishing method based on secondary clustering of power loads | |
| CN112113316B (en) | Method for extracting air conditioner load | |
| CN114580757A (en) | Load response prediction method and system based on industry type of large power consumption user | |
| Gerbec et al. | A methodology to classify distribution load profiles | |
| CN114463059A (en) | thermal load-green energy transaction electricity price prediction method of t-distribution GARCH model | |
| Wu et al. | A unified model for diagnosing energy usage abnormalities in regional integrated energy service systems |
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 |