CN110336289A - A kind of monitoring recognition methods, device, equipment and the storage medium of power load - Google Patents
A kind of monitoring recognition methods, device, equipment and the storage medium of power load Download PDFInfo
- Publication number
- CN110336289A CN110336289A CN201910625107.8A CN201910625107A CN110336289A CN 110336289 A CN110336289 A CN 110336289A CN 201910625107 A CN201910625107 A CN 201910625107A CN 110336289 A CN110336289 A CN 110336289A
- Authority
- CN
- China
- Prior art keywords
- load
- switching event
- power
- power load
- switching
- 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
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
- H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
- H02J3/14—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load by switching loads on to, or off from, network, e.g. progressively balanced loading
-
- 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
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B70/00—Technologies for an efficient end-user side electric power management and consumption
- Y02B70/30—Systems integrating technologies related to power network operation and communication or information technologies for improving the carbon footprint of the management of residential or tertiary loads, i.e. smart grids as climate change mitigation technology in the buildings sector, including also the last stages of power distribution and the control, monitoring or operating management systems at local level
- Y02B70/3225—Demand response systems, e.g. load shedding, peak shaving
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
- Y04S20/20—End-user application control systems
- Y04S20/222—Demand response systems, e.g. load shedding, peak shaving
Abstract
The invention discloses a kind of monitoring recognition methods of power load, device, equipment and storage medium, including obtaining power load characteristic quantity sequence, power load characteristic quantity sequence is analyzed using sliding window algorithm, obtain load switching event, load switching event is analyzed based on the power load feature database pre-established, obtain switching load corresponding with load switching event, finally export switching load, without preparatory training load decomposition model, only power load feature database need to be added in the feature of each type load, switching load is identified at the time point that load switching event occurs, the situation of load opening and closing is obtained, if analyzed since zero load state, the then operating status of available all loads.Therefore the monitoring recognition methods of power load provided by the invention is not necessarily to be largely used to the data sample of load decomposition model training, enormously simplifies the calculating of non-intrusion type load monitoring identification, reduces the cost of load monitoring identification.
Description
Technical field
The present invention relates to intelligent power grid technology field, more particularly to a kind of monitoring recognition methods of power load, device,
Equipment and storage medium.
Background technique
Load monitoring is as one of most important component part of smart grid, and realizes the first step of smart grid, must
Current ammeter must be broken through can only read electricity consumption total amount automatically, and cannot analyse in depth user's internal load ingredient, the load of acquisition
The bottlenecks such as Limited information, to improve power information acquisition system and intelligent electric power utilization system, support two-way interaction service and intelligence are used
It can service, and realize energy-saving, the peak load shifting of smart grid, alleviate the situation of regional shortage of electric power.
For power load type identification problem, what is applied on the market at present is intrusive load method, specific side
Formula is to install sensor in power load or install ammeter on socket to utilize sensor or electricity when load accesses power grid
Information when table obtains work obtains its real-time electricity consumption.Although this method can more be accurately determined power load
Type and this type power load electricity consumption, but need to install additional ammeter or sensor, put into larger, also needed after installation
A large amount of maintenance work is carried out, is not suitable for popularizing in an all-round way.
To overcome the above disadvantages, those skilled in the art propose the recognition methods of non-intrusion type load monitoring, its essence is
Total power information of subscriber household bus is decomposed into the power information of each electrical equipment, and then obtains electricity consumption by load decomposition
The power informations such as equipment energy consumption situation and user power utilization rule.In specific implementation, non-intrusion type load prison in the prior art
Detection identifying method needs to be that data sample is trained study based on a large amount of power load sequence, obtains load decomposition model,
And application load decomposition model identifies power load.As it can be seen that non-intrusion type load monitoring in the prior art identification side
Method needs the load decomposition model different for user's training of different consumption habits, and training requires a large amount of data every time
Sample, calculation amount is larger, higher cost.
Data sample needed for how reducing non-intrusion type load monitoring recognition methods simplifies non-intrusion type load monitoring and knows
Other calculating reduces the cost of load monitoring identification, is those skilled in the art's technical issues that need to address.
Summary of the invention
The object of the present invention is to provide monitoring recognition methods, device, equipment and the storage mediums of a kind of power load, are used for
Data sample needed for reducing the recognition methods of non-intrusion type load monitoring simplifies the calculating of non-intrusion type load monitoring identification, drop
The cost of underload monitoring identification.
In order to solve the above technical problems, the present invention provides a kind of monitoring recognition methods of power load, comprising:
Obtain power load characteristic quantity sequence；
The power load characteristic quantity sequence is analyzed using sliding window algorithm, obtains load switching event；
The load switching event is analyzed based on the power load feature database pre-established, is obtained and the load
The corresponding switching load of switching event；
Export the switching load.
Optionally, described to analyze the power load characteristic quantity sequence using sliding window algorithm, determine load switching event
It specifically includes:
It is slided in the power load characteristic quantity sequence using the sliding window of the first length with the first step-length, when adjacent two
When power load characteristic quantity difference between the sliding window of a first length meets preset condition, determines and described bear occurs
Lotus switching event；
Using the sliding window of the second length with the second step-length the first length that the load switching event occurs sliding
It is slided in power load characteristic quantity sequence in the range of window, position the load switching event and extracts the load switching thing
The transient process of part；
Wherein, first length is greater than the second length, and the first step is long to be greater than second step-length.
Optionally, described that the load switching event is analyzed based on the power load feature database pre-established, it obtains
To switching load corresponding with the load switching event, specifically include:
Power load transient state in the one of transient characteristic quantity of the load switching event and the power load feature database is special
Sign is matched, and matched load is judged whether there is；
If it is present with the matched load for the switching load；
If it does not exist, then carrying out load combination based on the steady state characteristic amount of the load switching event using genetic algorithm
Analysis, obtains load corresponding with the load switching event and combines；
Load combinatory analysis is carried out based on the one of transient characteristic quantity of the load switching event using multi-variable decision tree algorithm
To verify the load combination that the genetic algorithm obtains, judgement verifies whether to pass through；
If the verification passes, then the load group obtained with the genetic algorithm is combined into the switching load；
If verifying does not pass through, it is determined that the switching load is not record load, and do not record the temporary of load for described
State feature and the steady state characteristic for not recording load are loaded into the power load feature database.
Optionally, the electricity consumption in the one of transient characteristic quantity by the load switching event and the power load feature database
Load transient characteristic is matched, and is judged whether there is matched load, is specifically included:
Steady state characteristic in conjunction with the power load feature database, using the genetic algorithm based on the load switching event
Amount carries out analysis matching, obtains the first matching result；
In conjunction with the power load feature database, using the multi-variable decision tree algorithm based on the load switching event
One of transient characteristic quantity carries out analysis matching, obtains the second matching result；
Judge whether first matching result and second matching result are consistent；
If it is, determining that there are the matched loads, with first matching result and second matching result
Corresponding load is the matched load；
The matched load is not present if it is not, then determining.
Optionally, described that load combination point is carried out based on the steady state characteristic amount of the load switching event using genetic algorithm
Analysis, obtains load corresponding with the load switching event and combines, specifically include:
Using open state as first state, with closed state for the second state, with the first state and second shape
State random coded obtains individual, forms initial population with the individual of the first preset number, is minimised as objective function with deviation,
The sum of deviation with each individual is population entirety deviation；
It is reduced to Evolutionary direction with the whole deviation, operation is iterated based on the initial population, until obtaining
Meet the population of the default condition of convergence；
The load is determined in the smallest individual according to the deviation in the population for meeting the default condition of convergence
Combination；
Wherein, the code length of each individual is the quantity for the load recorded in the power load feature database, institute
State the total amount true value steady state characteristic corresponding with the individual for the steady state characteristic amount that deviation is the load switching event
The absolute value of the difference of the total amount match value of amount.
Optionally, further includes:
The load operating region at log history time point；
Correspondingly, described obtain individual with the first state and the second state random coded, specifically:
It is combined according to the load operating region of the previous time point at time point where the load switching event to each described
Individual carries out initial code, and carries out random coded to the rest segment on each individual.
Optionally, further includes:
The time point that the load switching event does not occur is analyzed to obtain using genetic algorithm and described institute does not occur
State the operating load combination at the time point of load switching event；
Record the operating load combination at each time point.
In order to solve the above technical problems, the present invention also provides a kind of monitoring identification devices of power load, comprising:
Acquiring unit, for obtaining power load characteristic quantity sequence；
Extraction unit obtains load switching thing for analyzing the power load characteristic quantity sequence using sliding window algorithm
Part；
Analytical unit, for being analyzed based on the power load feature database pre-established the load switching event,
Obtain switching load corresponding with the load switching event；
Output unit, for exporting the switching load.
In order to solve the above technical problems, the present invention also provides a kind of monitorings of power load to identify equipment, comprising:
Memory, for storing instruction, described instruction include the monitoring identification side of power load described in above-mentioned any one
The step of method；
Processor, for executing described instruction.
In order to solve the above technical problems, it is stored thereon with computer program the present invention also provides a kind of storage medium, it is described
It is realized when computer program is executed by processor as described in above-mentioned any one the step of the monitoring recognition methods of power load.
The monitoring recognition methods of power load provided by the present invention, including power load characteristic quantity sequence is obtained, it utilizes
It slides window algorithm and analyzes power load characteristic quantity sequence, load switching event is obtained, based on the power load feature pre-established
Load switching event is analyzed in library, obtains switching load corresponding with load switching event, finally exports switching load, nothing
Preparatory training load decomposition model is needed, only power load feature database need to be added in the feature of each type load, load switching is occurring
The time point of event identifies switching load, the situation of load opening and closing has been obtained, if opened from zero load state
Begin to analyze, then the operating status of available all loads.Therefore power load provided by the invention monitoring recognition methods without
It need to be largely used to the data sample of load decomposition model training, enormously simplify the calculating of non-intrusion type load monitoring identification, drop
The cost of low load monitoring identification.The present invention also provides monitoring identification device, equipment and the storage medium of a kind of power load,
With above-mentioned beneficial effect, details are not described herein.
Detailed description of the invention
It, below will be to embodiment or existing for the clearer technical solution for illustrating the embodiment of the present invention or the prior art
Attached drawing needed in technical description is briefly described, it should be apparent that, the accompanying drawings in the following description is only this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of flow chart of the monitoring recognition methods of power load provided in an embodiment of the present invention；
Fig. 2 is the flow chart of the specific embodiment of step S102 in a kind of Fig. 1 provided in an embodiment of the present invention；
Fig. 3 (a) is a kind of event monitoring schematic diagram of the sliding window of first length provided in an embodiment of the present invention；
Fig. 3 (b) is a kind of event monitoring schematic diagram of the sliding window of second length provided in an embodiment of the present invention；
Fig. 4 is the flow chart of the specific embodiment of step S103 in a kind of Fig. 1 provided in an embodiment of the present invention；
Fig. 5 is the flow chart of the specific embodiment of step S401 in a kind of Fig. 4 provided in an embodiment of the present invention；
Fig. 6 is a kind of schematic diagram of individual UVR exposure provided in an embodiment of the present invention；
Fig. 7 is the flow chart of the monitoring recognition methods of another power load provided in an embodiment of the present invention；
Fig. 8 is a kind of structural schematic diagram of the monitoring identification device of power load provided in an embodiment of the present invention；
Fig. 9 is a kind of structural schematic diagram of the monitoring identification equipment of power load provided in an embodiment of the present invention.
Specific embodiment
Core of the invention is to provide monitoring recognition methods, device, equipment and the storage medium of a kind of power load, is used for
Data sample needed for reducing the recognition methods of non-intrusion type load monitoring simplifies the calculating of non-intrusion type load monitoring identification, drop
The cost of underload monitoring identification.
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Fig. 1 is a kind of flow chart of the monitoring recognition methods of power load provided in an embodiment of the present invention.
As shown in Figure 1, the monitoring recognition methods of power load provided in an embodiment of the present invention includes:
S101: power load characteristic quantity sequence is obtained.
The type of power load characteristic quantity includes power, harmonic wave, current amplitude etc., can be first passed through in advance to power load work
The Current Voltage of work is analyzed, and determines the type of characteristic quantity for identification.According to the type of characteristic quantity, by each power load
All types of characteristic quantities be added in power load feature databases, the match cognization to subsequent step works.
In specific implementation, common household power load is acquired from unlatching by Data Data capture card and Hall sensor
To work normally again arrive close data, data are carried out with matlab include signal denoising etc. pretreatment, with reduce measure
Error restores the real data of open state.
S102: power load characteristic quantity sequence is analyzed using sliding window algorithm, obtains load switching event.
Power load characteristic quantity sequence is the time series of power load characteristic quantity.Remember that power load characteristic quantity sequence is(k=1,2,3 ...), according to the transient characterisitics of power load be arranged one section of appropriate length continuous array
(including m numerical value), the continuous array of this section are a sliding window.
It is calculated by the following formula the average value of array in sliding window:
Wherein, k is the initial time of sliding window.
Adjacent sliding is handled by the way that reasonable window sliding step-length d (time of i.e. each moving window counts) is arranged
Power load characteristic quantity sequence in window subtracts current point in time with the average value of array in the sliding window at previous time point
The average value of array, obtains difference DELTA in sliding windowk, it is arranged threshold value H (H > 0), works as Δk> H or Δk<-H is considered as when current
Between point have load switching event.
S103: analyzing load switching event based on the power load feature database pre-established, obtains throwing with load
Cut the corresponding switching load of event.
In in practical situations, switching load is divided into single load and load combines two kinds of situations.Load switching event is sent out
The variable quantity of characteristic quantity and the characteristic quantity of this type of each load in power load feature database when raw compare, if there is
Error then thinks to be matched to the corresponding load of this feature amount no more than the characteristic quantity of error threshold, it is believed that has been matched to single negative
Lotus, i.e. the load switching event are as caused by the opening or closing of this single load.If in power load feature database
It fails to match for each load, then in addition to the load switching event be multiple loads open or close the case where causing other than,
It is also possible that increasing new load type in system, then the modes analysis load switching thing such as genetic algorithm can be first passed through
Part checks whether the load combination in available load switching event, if be unable to get, illustrates it is to increase new bear
Lotus type.
S104: output switching load.
Export switching load, i.e., the load opened or closed in output load switching event.When acquired in step s101
Power load characteristic quantity sequence is the operating status of all loads after then available since system zero load condition.
The monitoring recognition methods of power load provided in an embodiment of the present invention, including power load characteristic quantity sequence is obtained,
Power load characteristic quantity sequence is analyzed using sliding window algorithm, load switching event is obtained, based on the power load pre-established
Feature database analyzes load switching event, obtains switching load corresponding with load switching event, and it is negative finally to export switching
Lotus is not necessarily to preparatory training load decomposition model, only power load feature database need to be added in the feature of each type load, load is occurring
The time point of switching event identifies switching load, the situation of load opening and closing has been obtained, if from zero load shape
State starts to analyze, then the operating status of available all loads.Therefore the monitoring of power load provided in an embodiment of the present invention
Recognition methods is not necessarily to be largely used to the data sample of load decomposition model training, enormously simplifies the identification of non-intrusion type load monitoring
Calculating, reduce load monitoring identification cost.
Fig. 2 is the flow chart of the specific embodiment of step S102 in a kind of Fig. 1 provided in an embodiment of the present invention；Fig. 3 (a)
For a kind of event monitoring schematic diagram of the sliding window of first length provided in an embodiment of the present invention；Fig. 3 (b) is the embodiment of the present invention
The event monitoring schematic diagram of the sliding window of the second length of one kind of offer.
On the basis of the above embodiments, in another embodiment, as shown in Fig. 2, the monitoring of power load shown in FIG. 1
The step S102 of recognition methods can specifically include:
S201: it is slided in power load characteristic quantity sequence using the sliding window of the first length with the first step-length, when adjacent
When power load characteristic quantity difference between the sliding window of two the first length meets preset condition, determines and load switching occurs
Event.
S202: using the sliding window of the second length with the second step-length occur load switching event the first length sliding
It is slided in power load characteristic quantity sequence in the range of window, position load switching event and extracts the transient state of load switching event
Process.
Wherein, the first length is greater than the second length, and the first step is long long greater than second step.
To guarantee to accurately identify and position load switching event, needs to be arranged the sliding window of appropriate length, however grow
Sliding window and short sliding window respectively have superiority and inferiority.As shown in Fig. 3 (a), since to have of short duration power steady for the opening process of some loads
Determine process, when amount analyzes power load characteristic quantity sequence characterized by power, if the sliding window length of setting is shorter, has very much
May be imperfect to the extraction of the transient process of load switching time, the time point that load switching event occurs judges by accident.
However, transient process is longer than true transient process much if sliding window length is too long, transient process extraction will affect
Characteristic quantity accuracy.For the load of a type, a kind of sliding of appropriate length can also be set by repetition test
Window, but for a plurality of types of loads, a kind of sliding window of length will will appear above-mentioned problems.
Therefore the embodiment of the present invention provides a kind of improved sliding window algorithm, and the sliding window of two kinds of length is combined, can be with
Referred to as length slides window algorithm.As shown in Fig. 3 (a), it is by sliding window (i.e. long sliding window) determination of the first length first
It is no to have load switching event, when the sliding window of the first length is in such as Fig. 3 (b) sliding window that is shown, then passing through the second length
The power load characteristic quantity sequence of (i.e. short sliding window) in the range of the sliding window of the first length of load switching event occurs
Upper sliding, the definite time point for positioning load switching event and occurring, and it is precisely separating, extracts the transient state mistake of load switching event
Journey.
Fig. 4 is the flow chart of the specific embodiment of step S103 in a kind of Fig. 1 provided in an embodiment of the present invention；Fig. 5 is
The flow chart of the specific embodiment of step S401 in a kind of Fig. 4 provided in an embodiment of the present invention；Fig. 6 mentions for the embodiment of the present invention
A kind of schematic diagram of the individual UVR exposure supplied.
On the basis of the above embodiments, in another embodiment, as shown in figure 4, the monitoring of power load shown in FIG. 1
The step S103 of recognition methods can specifically include:
S401: by the power load transient characteristic in the one of transient characteristic quantity of load switching event and power load feature database into
Row matching, judges whether there is matched load；If it is, entering step S402；If it is not, then entering step S403.
In specific implementation, first assume that load switching event is caused by single load switching, first by load switching event
One of transient characteristic quantity and the transient characteristic of each load in power load feature database compare, select consistent load.
As shown in figure 5, step S401 can specifically include:
S501: it in conjunction with power load feature database, is carried out using genetic algorithm based on the steady state characteristic amount of load switching event
Analysis matching, obtains the first matching result.
S502: in conjunction with power load feature database, the transient state using multi-variable decision tree algorithm based on load switching event is special
Sign amount carries out analysis matching, obtains the second matching result.
S503: judge whether the first matching result and the second matching result are consistent；If it is, entering step S504；Such as
Fruit is no, then enters step S505.
S504: determine that, there are matched load, load corresponding with the first matching result and the second matching result is matching
Load.
S505: it determines and matched load is not present.
Accuracy is matched to improve, transient process is analyzed with Analysis of Genetic Algorithms steady-state process and multi-variable decision tree algorithm
In conjunction with mode carry out the match cognization process of single load.If the matching result of two kinds of algorithms is inconsistent, explanation is not present
Matched load, load switching event may be that switching has occurred in multiple loads in similar time, it is also possible to have and not record
Switching has occurred in load.
S402: using matched load as switching load.
S403: using genetic algorithm based on load switching event steady state characteristic amount carry out load combinatory analysis, obtain with
The corresponding load combination of load switching event.
In specific implementation, step S403 may include:
Using open state as first state, with closed state for the second state, compiled at random with first state and the second state
Code obtains individual, forms initial population with the individual of the first preset number, objective function is minimised as with deviation, with each individual
The sum of deviation be population entirety deviation；
It is reduced to Evolutionary direction with whole deviation, operation is iterated based on initial population, is preset until obtaining and meeting
The population of the condition of convergence；
Determine load combination in the smallest individual according to deviation in the population for meeting the default condition of convergence；
Wherein, the code length of each individual is the quantity for the load recorded in power load feature database, and deviation is negative
The difference of the total amount true value of the steady state characteristic amount of lotus switching event and the total amount match value of the corresponding steady state characteristic amount of individual it is absolute
Value.
As shown in fig. 6, setting the quantity of the load in system as N, first state is indicated with " 1 ", the second state " 0 " table
Show, generates M individual S at randomi(i=1 ... M) constitutes initial population { S1,S2,S3,…,SM, then it is 0-1 that each individual, which is N,
Vector, i.e.,The power amount of being characterized such as is selected, for the general power P (t) of any time point sampling,
Power features amount when each load operation is expressed as P=[P1,P2,P3,…,PN], then i-th of corresponding fitting general power of individual
For SiPT。
Objective function is shown below:
Min F (i, t, P)=P (t)-SiPT (2)
The condition of convergence can be less than preset value for deviation, or iterative calculation reaches preset times, or full simultaneously
Both foots.
S404: load combinatory analysis is carried out based on the one of transient characteristic quantity of load switching event using multi-variable decision tree algorithm
To verify the load combination that genetic algorithm obtains, judgement verifies whether to pass through；If it is, entering step S405；If it is not, then
Enter step S406.
Similarly, load combination is carried out based on the one of transient characteristic quantity of load switching event using multi-variable decision tree algorithm again
Analysis improves the accuracy of load combination identification to verify the load combination that genetic algorithm obtains.
S405: the load group obtained with genetic algorithm is combined into switching load.
S406: determine that switching load is not record load, and do not record the transient characteristic for not recording load and load
Steady state characteristic is loaded into power load feature database.
When load switching event neither caused by single load switching, nor then being said caused by multiple load switchings
It is bright to have new load, i.e., load not being recorded, switching occurring, the transient characteristic for not recording load and steady state characteristic are loaded into electricity consumption and born
Lotus feature database is to carry out the update of power load feature database.
On the basis of the embodiment described in Fig. 4, in another embodiment, for the optimization for further speeding up algorithm, reduce excellent
The number of iterations needed for changing, the monitoring recognition methods of power load provided in an embodiment of the present invention further include:
The load operating region at log history time point；
Correspondingly, individual is obtained with first state and the second state random coded, specifically:
Each individual is carried out according to the load operating region of the previous time point at time point where load switching event initial
Coding, and random coded is carried out to the rest segment on each individual.
In specific implementation, negative for the operation of each time point if being analyzed since system is in zero load state
Lotus combination is all recorded, and contents may include time point, load type, load operating region, load opening time point
(if the load is in the open state), load opening time etc..
When load switching event occurs, the load operating region of the previous time point of current point in time is searched, according to this
Load operating region first carries out initial code to individual.For example, determine that previous time point load 1,4,9 is in the open state,
He is in close state load, then for each individual, the coding of 1,4,8,9 position of load is first set to 1, then to other positions
Carry out random coded.In the operating status of load 1,4,8,9, there is no being intersected, become in the case where variation as a result,
Different coding number reduces, so that the optimization of algorithm is accelerated, the number of iterations needed for reducing optimization.And even if load 1,4,
8, there is the load closed in 9, also can largely effectively reduce required calculating.Similarly, it can also will be in and close shape
The load of state is first encoded to 0, carries out random coded to other positions.
Fig. 7 is the flow chart of the monitoring recognition methods of another power load provided in an embodiment of the present invention.
On the basis of the above embodiments, in another embodiment, for time point of load switching event does not occur, this
The monitoring recognition methods for another power load that inventive embodiments provide further include:
S701: the time point that load switching event does not occur is analyzed using genetic algorithm, load throwing is not occurred
Cut the operating load combination at the time point of event.
S702: the operating load combination at each time point is recorded.
Step S701 and step S702 since system suitable for there is the prison for carrying out power load the state of load operation
The case where surveying identification, specific implementation can refer to above-described embodiment.
The corresponding each embodiment of monitoring recognition methods of power load as detailed above, on this basis, the present invention is also
Disclose the monitoring identification device of power load corresponding with the above method.
Fig. 8 is a kind of structural schematic diagram of the monitoring identification device of power load provided in an embodiment of the present invention.
As shown in figure 8, the monitoring identification device of power load provided in an embodiment of the present invention includes:
Acquiring unit 801, for obtaining power load characteristic quantity sequence；
Extraction unit 802 obtains load switching thing for analyzing power load characteristic quantity sequence using sliding window algorithm
Part；
Analytical unit 803 is obtained for being analyzed based on the power load feature database pre-established load switching event
To switching load corresponding with load switching event；
Output unit 804, for exporting switching load.
Since the embodiment of device part is corresponded to each other with the embodiment of method part, the embodiment of device part is asked
Referring to the description of the embodiment of method part, wouldn't repeat here.
Fig. 9 is a kind of structural schematic diagram of the monitoring identification equipment of power load provided in an embodiment of the present invention.
As shown in figure 9, the monitoring identification equipment of power load provided in an embodiment of the present invention includes:
Memory 901, for storing instruction, described instruction include the prison of power load described in above-mentioned any one embodiment
The step of detection identifying method；
Processor 902, for executing described instruction.
The monitoring of power load provided in this embodiment identifies equipment, due to that can call memory storage by processor
Computer program, realize as above-mentioned any embodiment provide power load monitoring recognition methods the step of, so this use
The monitoring identification equipment of electric load has the same actual effect of monitoring recognition methods with above-mentioned power load.
This programme in order to better understand, the embodiment of the present invention also provide a kind of storage medium, store on the storage medium
There is computer program, realizes that the monitoring for the power load that any embodiment as above is mentioned is known when computer program is executed by processor
The step of other method.
Storage medium provided in this embodiment, since the computer journey of storage medium storage can be called by processor
The step of sequence, the monitoring recognition methods for the power load that realization such as above-mentioned any embodiment provides, so this storage medium has
With the same actual effect of monitoring recognition methods of above-mentioned power load.
In several embodiments provided herein, it should be understood that disclosed method, apparatus, equipment and storage
Medium may be implemented in other ways.For example, the apparatus embodiments described above are merely exemplary, for example, mould
The division of block, only a kind of logical function partition, there may be another division manner in actual implementation, for example, multiple modules or
Component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point is shown
The mutual coupling, direct-coupling or communication connection shown or discussed can be through some interfaces, between device or module
Coupling or communication connection are connect, can be electrical property, mechanical or other forms.As illustrated by the separation member module can be or
It may not be and be physically separated, the component shown as module may or may not be physical module, it can
It is in one place, or may be distributed on multiple network modules.Part therein can be selected according to the actual needs
Or whole modules achieve the purpose of the solution of this embodiment.
It, can also be in addition, can integrate in a processing module in each functional module in each embodiment of the application
It is that modules physically exist alone, can also be integrated in two or more modules in a module.Above-mentioned integrated mould
Block both can take the form of hardware realization, can also be realized in the form of software function module.
If integrated module is realized and when sold or used as an independent product in the form of software function module, can
To be stored in a computer readable storage medium.Based on this understanding, the technical solution of the application substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
The whole of (can be personal computer, funcall device or the network equipment etc.) execution each embodiment method of the application
Or part steps.And storage medium above-mentioned may include: USB flash disk, mobile hard disk, read-only memory (Read-Only Memory,
ROM), random access memory (Random Access Memory, RAM), magnetic or disk etc. are various can store program
The medium of code.
Above to a kind of monitoring recognition methods, device, equipment and the storage medium of power load provided by the present invention into
It has gone and has been discussed in detail.Each embodiment is described in a progressive manner in specification, the highlights of each of the examples are with
The difference of other embodiments, the same or similar parts in each embodiment may refer to each other.For disclosed in embodiment
For device, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method portion
It defends oneself bright.It should be pointed out that for those skilled in the art, in the premise for not departing from the principle of the invention
Under, it can be with several improvements and modifications are made to the present invention, these improvement and modification also fall into the protection of the claims in the present invention
In range.
It should also be noted that, in the present specification, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Claims (10)
1. a kind of monitoring recognition methods of power load characterized by comprising
Obtain power load characteristic quantity sequence；
The power load characteristic quantity sequence is analyzed using sliding window algorithm, obtains load switching event；
The load switching event is analyzed based on the power load feature database pre-established, is obtained and the load switching
The corresponding switching load of event；
Export the switching load.
2. monitoring recognition methods according to claim 1, which is characterized in that described to analyze the use using sliding window algorithm
Electric load characteristic quantity sequence determines that load switching event specifically includes:
It is slided in the power load characteristic quantity sequence using the sliding window of the first length with the first step-length, when two neighboring
When power load characteristic quantity difference between the sliding window of first length meets preset condition, determine that the load, which occurs, to be thrown
Cut event；
Using the sliding window of the second length with the second step-length the first length that the load switching event occurs sliding window
It is slided in power load characteristic quantity sequence in range, position the load switching event and extracts the load switching event
Transient process；
Wherein, first length is greater than the second length, and the first step is long to be greater than second step-length.
3. monitoring recognition methods according to claim 1, which is characterized in that described special based on the power load pre-established
The load switching event is analyzed in sign library, is obtained switching load corresponding with the load switching event, is specifically included:
By the power load transient characteristic in the one of transient characteristic quantity of the load switching event and the power load feature database into
Row matching, judges whether there is matched load；
If it is present with the matched load for the switching load；
If it does not exist, then carrying out load combination point based on the steady state characteristic amount of the load switching event using genetic algorithm
Analysis, obtains load corresponding with the load switching event and combines；
Load combinatory analysis is carried out to test based on the one of transient characteristic quantity of the load switching event using multi-variable decision tree algorithm
The load combination that the genetic algorithm obtains is demonstrate,proved, judgement verifies whether to pass through；
If the verification passes, then the load group obtained with the genetic algorithm is combined into the switching load；
If verifying does not pass through, it is determined that the switching load is not record load, and the transient state for not recording load is special
The steady state characteristic for not recording load of seeking peace is loaded into the power load feature database.
4. monitoring recognition methods according to claim 3, which is characterized in that the transient state by the load switching event
Characteristic quantity is matched with the power load transient characteristic in the power load feature database, is judged whether there is matched negative
Lotus specifically includes:
In conjunction with the power load feature database, steady state characteristic amount using the genetic algorithm based on the load switching event into
Row analysis matching, obtains the first matching result；
Transient state in conjunction with the power load feature database, using the multi-variable decision tree algorithm based on the load switching event
Characteristic quantity carries out analysis matching, obtains the second matching result；
Judge whether first matching result and second matching result are consistent；
If it is, determine that there are the matched loads, it is corresponding with first matching result and second matching result
Load be the matched load；
The matched load is not present if it is not, then determining.
5. according to monitoring recognition methods described in claim 3 or 4 any one, which is characterized in that described to apply genetic algorithm
Steady state characteristic amount based on the load switching event carries out load combinatory analysis, obtains corresponding with the load switching event
Load combination, specifically includes:
Using open state as first state, with closed state for the second state, with the first state and second state with
Machine encodes to obtain individual, forms initial population with the individual of the first preset number, objective function is minimised as with deviation, with each
The sum of deviation of the individual is population entirety deviation；
It is reduced to Evolutionary direction with the whole deviation, operation is iterated based on the initial population, until being met
The population of the default condition of convergence；
Determine the load combination in the smallest individual according to the deviation in the population for meeting the default condition of convergence；
Wherein, the code length of each individual is the quantity for the load recorded in the power load feature database, described inclined
Difference is the total amount true value steady state characteristic amount corresponding with the individual of the steady state characteristic amount of the load switching event
The absolute value of the difference of total amount match value.
6. monitoring recognition methods according to claim 5, which is characterized in that further include:
The load operating region at log history time point；
Correspondingly, described obtain individual with the first state and the second state random coded, specifically:
It is combined according to the load operating region of the previous time point at time point where the load switching event to each individual
Initial code is carried out, and random coded is carried out to the rest segment on each individual.
7. monitoring recognition methods according to claim 1, which is characterized in that further include:
Using genetic algorithm to the time point that the load switching event does not occur analyzed to obtain it is described do not occur it is described negative
The operating load at the time point of lotus switching event combines；
Record the operating load combination at each time point.
8. a kind of monitoring identification device of power load characterized by comprising
Acquiring unit, for obtaining power load characteristic quantity sequence；
Extraction unit obtains load switching event for analyzing the power load characteristic quantity sequence using sliding window algorithm；
Analytical unit is obtained for being analyzed based on the power load feature database pre-established the load switching event
Switching load corresponding with the load switching event；
Output unit, for exporting the switching load.
9. a kind of monitoring of power load identifies equipment characterized by comprising
Memory, for storing instruction, described instruction include that the monitoring of power load described in claim 1 to 7 any one is known
The step of other method；
Processor, for executing described instruction.
10. a kind of storage medium, is stored thereon with computer program, which is characterized in that the computer program is held by processor
It is realized when row as described in claim 1 to 7 any one the step of the monitoring recognition methods of power load.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910625107.8A CN110336289A (en) | 2019-07-11 | 2019-07-11 | A kind of monitoring recognition methods, device, equipment and the storage medium of power load |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910625107.8A CN110336289A (en) | 2019-07-11 | 2019-07-11 | A kind of monitoring recognition methods, device, equipment and the storage medium of power load |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110336289A true CN110336289A (en) | 2019-10-15 |
Family
ID=68146401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910625107.8A Pending CN110336289A (en) | 2019-07-11 | 2019-07-11 | A kind of monitoring recognition methods, device, equipment and the storage medium of power load |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110336289A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110954744A (en) * | 2019-11-18 | 2020-04-03 | 浙江工业大学 | Non-invasive load monitoring method based on event detection |
| CN111351974A (en) * | 2020-03-12 | 2020-06-30 | 华北电力大学 | Environment-friendly monitoring method and system based on electricity consumption perception |
| CN111366800A (en) * | 2020-03-11 | 2020-07-03 | 北京慧飒科技有限责任公司 | Non-invasive intelligent identification method for electrical load |
| CN111505559A (en) * | 2020-04-29 | 2020-08-07 | 国网重庆市电力公司营销服务中心 | Load identification electric energy meter detection method and device |
| CN112234996A (en) * | 2020-10-29 | 2021-01-15 | 广东电网有限责任公司计量中心 | Compression method, device, equipment and storage medium of power load data |
| CN113030564A (en) * | 2021-03-10 | 2021-06-25 | 南方电网数字电网研究院有限公司 | Load identification method based on double-core intelligent electric meter system |
| CN113222401A (en) * | 2021-05-10 | 2021-08-06 | 湖南省计量检测研究院 | Non-invasive load type intelligent identification method, device, server and system |
| CN113255236A (en) * | 2021-07-07 | 2021-08-13 | 浙江大学 | Non-invasive load self-adaptive identification method based on twin network |
| CN113447740A (en) * | 2021-06-21 | 2021-09-28 | 天津大学 | Non-invasive load event global optimization matching method and system |
| CN113487242A (en) * | 2021-08-19 | 2021-10-08 | 浙江华云清洁能源有限公司 | Energy efficiency management method, system, equipment and storage medium |
| CN113765096A (en) * | 2021-07-22 | 2021-12-07 | 深圳供电局有限公司 | Power load analysis method and device, computer equipment and storage medium |
| CN114019281A (en) * | 2021-11-04 | 2022-02-08 | 国网四川省电力公司营销服务中心 | Non-invasive load monitoring method and system based on die body excavation and semi-supervision method |
| CN114184870A (en) * | 2021-12-14 | 2022-03-15 | 河北科技大学 | Non-invasive load identification method and equipment |
| CN114996182A (en) * | 2022-05-23 | 2022-09-02 | 中国计量大学 | Steady state data flow output device for load recognition model training |
| EP4166950A1 (en) * | 2021-10-16 | 2023-04-19 | Elspec Engineering Ltd. | Electrical parameter monitoring |
-
2019
- 2019-07-11 CN CN201910625107.8A patent/CN110336289A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| 曹敏 等: ""基于暂态过程的非侵入式负荷监测研究"", 《水电能源科学》, vol. 36, no. 8, 30 August 2018 (2018-08-30), pages 177 - 180 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110954744A (en) * | 2019-11-18 | 2020-04-03 | 浙江工业大学 | Non-invasive load monitoring method based on event detection |
| CN111366800A (en) * | 2020-03-11 | 2020-07-03 | 北京慧飒科技有限责任公司 | Non-invasive intelligent identification method for electrical load |
| CN111351974A (en) * | 2020-03-12 | 2020-06-30 | 华北电力大学 | Environment-friendly monitoring method and system based on electricity consumption perception |
| CN111505559A (en) * | 2020-04-29 | 2020-08-07 | 国网重庆市电力公司营销服务中心 | Load identification electric energy meter detection method and device |
| CN112234996A (en) * | 2020-10-29 | 2021-01-15 | 广东电网有限责任公司计量中心 | Compression method, device, equipment and storage medium of power load data |
| CN113030564A (en) * | 2021-03-10 | 2021-06-25 | 南方电网数字电网研究院有限公司 | Load identification method based on double-core intelligent electric meter system |
| CN113222401A (en) * | 2021-05-10 | 2021-08-06 | 湖南省计量检测研究院 | Non-invasive load type intelligent identification method, device, server and system |
| CN113447740A (en) * | 2021-06-21 | 2021-09-28 | 天津大学 | Non-invasive load event global optimization matching method and system |
| CN113255236A (en) * | 2021-07-07 | 2021-08-13 | 浙江大学 | Non-invasive load self-adaptive identification method based on twin network |
| CN113255236B (en) * | 2021-07-07 | 2021-10-08 | 浙江大学 | Non-invasive load self-adaptive identification method based on twin network |
| CN113765096A (en) * | 2021-07-22 | 2021-12-07 | 深圳供电局有限公司 | Power load analysis method and device, computer equipment and storage medium |
| CN113765096B (en) * | 2021-07-22 | 2024-01-19 | 深圳供电局有限公司 | Electrical load analysis method, apparatus, computer device, and storage medium |
| CN113487242A (en) * | 2021-08-19 | 2021-10-08 | 浙江华云清洁能源有限公司 | Energy efficiency management method, system, equipment and storage medium |
| EP4166950A1 (en) * | 2021-10-16 | 2023-04-19 | Elspec Engineering Ltd. | Electrical parameter monitoring |
| CN114019281A (en) * | 2021-11-04 | 2022-02-08 | 国网四川省电力公司营销服务中心 | Non-invasive load monitoring method and system based on die body excavation and semi-supervision method |
| CN114184870A (en) * | 2021-12-14 | 2022-03-15 | 河北科技大学 | Non-invasive load identification method and equipment |
| CN114996182A (en) * | 2022-05-23 | 2022-09-02 | 中国计量大学 | Steady state data flow output device for load recognition model training |
| CN114996182B (en) * | 2022-05-23 | 2024-04-26 | 中国计量大学 | Steady-state data stream output device for training load identification model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110336289A (en) | A kind of monitoring recognition methods, device, equipment and the storage medium of power load | |
| CN111612651B (en) | Abnormal electric quantity data detection method based on long-term and short-term memory network | |
| CN111368890A (en) | Fault detection method and device and information physical fusion system | |
| CN112699913A (en) | Transformer area household variable relation abnormity diagnosis method and device | |
| CN109543901A (en) | Short-Term Load Forecasting Method based on information fusion convolutional neural networks model | |
| CN111967512B (en) | Abnormal electricity utilization detection method, system and storage medium | |
| CN112149873B (en) | Low-voltage station line loss reasonable interval prediction method based on deep learning | |
| CN108765189B (en) | Open type partial discharge big data management system based on intelligent diagnosis algorithm | |
| CN110045209A (en) | Detection method, device, equipment and the readable storage medium storing program for executing of electricity consumption data exception | |
| CN110619479B (en) | Fault judgment model modeling method and fault judgment method for photovoltaic module | |
| CN104408667A (en) | Comprehensive assessment method and system of power quality | |
| CN111884347B (en) | Power data centralized control system for multi-source power information fusion | |
| CN109767054A (en) | Efficiency cloud appraisal procedure and edge efficiency gateway based on deep neural network algorithm | |
| CN103269279B (en) | A kind of boss stands associating topology Identification method | |
| CN113358993B (en) | Online fault diagnosis method and system for multi-level converter IGBT | |
| CN109633448A (en) | Identify the method, apparatus and terminal device of cell health state | |
| CN113780684A (en) | Intelligent building user energy consumption behavior prediction method based on LSTM neural network | |
| CN112633658A (en) | Low-voltage distribution area topological relation identification method based on CNN-LSTM | |
| CN114547816A (en) | Intelligent analysis method and device for reliability of power distribution network | |
| CN114021425A (en) | Power system operation data modeling and feature selection method and device, electronic equipment and storage medium | |
| CN110968703B (en) | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm | |
| CN112904148A (en) | Intelligent cable operation monitoring system, method and device | |
| CN113010985A (en) | Non-invasive load identification method based on parallel AANN | |
| CN112595918A (en) | Low-voltage meter reading fault detection method and device | |
| CN111310920A (en) | Method for applying deep learning neural network technology to signal acquisition device |
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: 20191015 |
|
| RJ01 | Rejection of invention patent application after publication |