CN108197817A - A kind of method of the non-intrusion type load transient state monitoring based on big data - Google Patents
A kind of method of the non-intrusion type load transient state monitoring based on big data Download PDFInfo
- Publication number
- CN108197817A CN108197817A CN201810050843.0A CN201810050843A CN108197817A CN 108197817 A CN108197817 A CN 108197817A CN 201810050843 A CN201810050843 A CN 201810050843A CN 108197817 A CN108197817 A CN 108197817A
- Authority
- CN
- China
- Prior art keywords
- intrusion type
- type load
- data
- big data
- transient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 102
- 230000001052 transient effect Effects 0.000 title claims abstract description 92
- 238000012544 monitoring process Methods 0.000 title claims abstract description 63
- 230000008569 process Effects 0.000 claims abstract description 61
- 238000005259 measurement Methods 0.000 claims abstract description 24
- 230000009467 reduction Effects 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000005516 engineering process Methods 0.000 claims description 19
- 238000000605 extraction Methods 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 7
- 238000000926 separation method Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 2
- 238000009825 accumulation Methods 0.000 abstract description 5
- 230000005611 electricity Effects 0.000 description 6
- 240000002853 Nelumbo nucifera Species 0.000 description 4
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 4
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000000205 computational method Methods 0.000 description 3
- 238000013439 planning Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000000491 multivariate analysis Methods 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
Big data of the present invention monitors applied technical field, and in particular to a kind of method of the non-intrusion type load monitoring based on big data includes the following steps：（1）The remote collection of electric power data by carrying out data monitoring and monitoring and controlling to electric system, acquires the actual measurement sample of all kinds of transient processes comprehensively；（2）The training pattern of non-intrusion type load monitoring is established, forms discriminant space；（3）The actual measurement of non-intrusion type load transient state monitoring is carried out, the transient process of actual measurement is differentiated according to discriminant space.The present invention utilizes big data technological means, from the Power system load data of the magnanimity of accumulation, obtain the data of a large amount of transient process, method according to rational load classification, suitable characteristic parameter is chosen to build the feature decision space of higher-dimension, the dimensionality reduction and sorting technique in big data technological means are recycled, can precisely and quickly realize the function of non-intrusion type load monitoring.
Description
Technical field
The invention belongs to big datas to monitor applied technical field, and in particular to a kind of non-intrusion type load based on big data
The method of monitoring.
Background technology
Today's society, computer technology have obtained rapid development, and information technology is stepped into intelligent society for the mankind and opened
Gate has driven the Developing Track for Modern Service Industry such as internet, Internet of Things, e-commerce, modern logistics, network finance, has expedited the emergence of vehicle
The new industries such as networking, intelligent grid, new energy, intelligent transportation, intelligent city, high-end equipment manufacturing develop since the mankind will
The data of processing are more and more, and desired requirement has been not achieved in simple manual analysis, so data base administration system
The extensive use of system, saves man power and material.The more doing the more big with the capacity of database, the historical data that people save bit by bit
It will be more and more.How people should face the data of magnanimity, that how the behind of mass data can just explore is significant
Information, and it was found that the value of information therefrom extracts the knowledge useful to the mankind, oneself is through becoming current big data field needs
The major issue of thinking.
In electric system, in order to realize target that people are monitored the load in electric system, non-intrusion type is born
Lotus monitoring is come into being, and from decision angle, non-intrusion type load monitoring can enable policymaker understand entire power train
The content of each type load and state and the electricity consumption per type load and electricity consumption time in system, and load switching is monitored, it obtains in real time
The operation conditions of whole system is taken, so as to reasonably guide load electricity consumption in decision-making level, load switching is scientifically arranged, realizes
The efficient and stable operation of electric system；From power planning angle, non-intrusion type load monitoring can be from whole system
Global level provides detailed electric power data, and planning personnel is enable to understand the electricity consumption rule of load on big time granularity and become
Gesture, this is to being better achieved load prediction and Electric Power Network Planning important role；It is non-to invade from the angle of each power consumer
Enter the important technology support that formula load monitoring is the forward positions such as smart home, intelligent plant concept, power consumer can be made any
When and where can understand the state of each type load in oneself local electric system, and monitor the transient state thing such as failure, switching
The generation of part ensures the normal operation of entire local electric system.
With the continuous growth of national power consumption, the load for accessing electric system is also more and more, realizes that non-intrusion type is born
The difficulty of lotus monitoring also increases therewith, and especially under big data background, the load type for not only accessing electric system is continuously increased,
And represent that a large amount of Temporal Datas of the switching state of load also will be collected and cumulative, it is bound to cause non-intrusion type load
The problem of ability of the transient state identification of decomposition declines.
Invention content
It is an object of the invention to solve non-intrusion type caused by the load type of proposition is more, transient process data volume is big to bear
The problem of transient state discrimination of lotus monitoring is low proposes a kind of side of the non-intrusion type load transient state monitoring based on big data technology
Method using big data technological means, from the Power system load data of the magnanimity of accumulation, obtains the data of a large amount of transient process,
By building training pattern and test model, the transient state monitoring function of non-intrusion type load decomposition is realized.
The present invention is achieved by the following technical solutions：
A kind of method of the non-intrusion type load monitoring based on big data, it is characterised in that：Include the following steps：
（1）The remote collection of electric power data by carrying out data monitoring and monitoring and controlling to electric system, acquires all kinds of temporary comprehensively
The actual measurement sample of state process；
（2）The training pattern of non-intrusion type load monitoring is established, forms discriminant space；
（3）The actual measurement of non-intrusion type load transient state monitoring is carried out, the transient process of actual measurement is differentiated according to discriminant space.
Further, the step（1）In electric system host computer is mainly connected by central station, host computer is logical through transmitting
Road carries out data acquisition by RTU control devices.
Further, the step（2）Include the following steps：
（21）Collected transient state actual measurement sample is detected with detaching；
（22）Feature extraction is carried out to the transient state sample after separation；
（23）With dimension reduction method, low-dimensional discriminant space is formed；
Further, step（21）Detection process in, current waveform is sampled, and in buffer records most according to certain frequency
The waveform of nearly fixed cycle numbers if the variation that current strength occurs in real time is more than certain threshold value, judges that transient process is opened
Begin, until apparent variation is not occurring for real-time current intensity, then judge that transient process terminates, and record the wave of transient process
Shape.
Further,（22）In, the characteristic parameter before and after the transient state of extraction is formed to the feature space of higher-dimension, and remove
After unusual sample, normalization forms sample set.
Further, step（23）Feature space dimensionality reduction is carried out using the Method of Data with Adding Windows of big data technology, so as to shape
Into the discriminant space of low-dimensional.
Compared with prior art, the beneficial effects of the invention are as follows：
The design proposes a kind of method of the non-intrusion type load monitoring based on big data, it is proposed that the inspection of load transient process
Survey and separation algorithm, can detect the generation of transient process, and exclude the interference of other loads, obtain single transient-wave,
The characteristic quantity differentiated for transient process and corresponding computational methods are given, and high dimensional feature is formed with normalized characteristic quantity
Big data technology is applied to the dimensionality reduction of transient process feature space by space, and high-dimensional feature space is reduced to low-dimensional discriminant space,
Reduce the dimension of feature space.
The design non-intrusion type load caused by for current electric system internal loading type is more, transient process data volume is big
The problem of transient state discrimination of monitoring is low proposes a kind of method of the non-intrusion type load transient state monitoring based on big data technology,
Using big data technological means, from the Power system load data of the magnanimity of accumulation, the data of a large amount of transient process, foundation are obtained
The method of rational load classification chooses suitable characteristic parameter to build the feature decision space of higher-dimension, recycles big data
Dimensionality reduction and sorting technique in technological means by building training pattern and test model, precisely and can be realized quickly
The function of non-intrusion type load monitoring.In addition, the method for the present invention principle is reliable, step is simple, before having very extensive application
Scape.
It can be seen that compared with prior art, the present invention improve with prominent substantive distinguishing features and significantly, implement
Advantageous effect be also obvious.
Description of the drawings
Fig. 1 is the system hardware structure figure of electric power data remote gathering system in the present invention.
Fig. 2 is the flow chart that non-intrusion type load monitoring training pattern is built in the present invention.
Fig. 3 is structure non-intrusion type load monitoring Model Measured flow chart in the present invention.
Fig. 4 is that electric power data remote gathering system forms structure chart substantially in the present invention.
Fig. 5 is the refrigerator transient-wave figure obtained in the embodiment of the present invention 3.
Fig. 6 is the air compressor transient-wave figure obtained in the embodiment of the present invention 3.
Fig. 7 is the television set transient-wave figure obtained in the embodiment of the present invention 3.
Fig. 8 is the electric light transient-wave figure obtained in the embodiment of the present invention 3.
Specific embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings：
Embodiment 1
As shown in Figures 1 to 3, a kind of method of non-intrusion type load monitoring based on big data provided by the invention, feature exist
In：Include the following steps：
（1）The remote collection of electric power data by carrying out data monitoring and monitoring and controlling to electric system, acquires all kinds of temporary comprehensively
The actual measurement sample of state process；
（2）The training pattern of non-intrusion type load monitoring is established, forms discriminant space；
（3）The actual measurement of non-intrusion type load transient state monitoring is carried out, the transient process of actual measurement is differentiated according to discriminant space.
Further, as shown in Figure 1, the step（1）In electric system host computer is mainly connected by central station, it is upper
Machine carries out data acquisition through transmission channel by RTU control devices.
Further, the step（2）Include the following steps：
（21）Collected transient state actual measurement sample is detected with detaching；
（22）Feature extraction is carried out to the transient state sample after separation；
（23）With dimension reduction method, low-dimensional discriminant space is formed；
As shown in Fig. 2, in the present embodiment, step（21）Detection process in, current waveform is sampled according to certain frequency, and
In the waveform of the nearest fixed cycle numbers of buffer records, if the variation that current strength occurs in real time is more than certain threshold value,
Judgement transient process starts, and until apparent variation is not occurring for real-time current intensity, then judges that transient process terminates, and records
The waveform of lower transient process.Step（22）In, by the feature space of the characteristic parameter formation higher-dimension before and after the transient state of extraction, and
After removing unusual sample, normalization forms sample set.Step（23）Feature is carried out using the Method of Data with Adding Windows of big data technology
Space dimensionality reduction, so as to form the discriminant space of low-dimensional.
The design proposes a kind of method of the non-intrusion type load monitoring based on big data, it is proposed that load transient process
Detection and separation algorithm, can detect the generation of transient process, and exclude the interference of other loads, obtain single transient state wave
Shape gives the characteristic quantity differentiated for transient process and corresponding computational methods, and forms higher-dimension with normalized characteristic quantity
Big data technology is applied to the dimensionality reduction of transient process feature space by feature space, and high-dimensional feature space is reduced to low-dimensional differentiates
Space reduces the dimension of feature space.
The design non-intrusion type load caused by for current electric system internal loading type is more, transient process data volume is big
The problem of transient state discrimination of monitoring is low proposes a kind of method of the non-intrusion type load transient state monitoring based on big data technology,
Using big data technological means, from the Power system load data of the magnanimity of accumulation, the data of a large amount of transient process, foundation are obtained
The method of rational load classification chooses suitable characteristic parameter to build the feature decision space of higher-dimension, recycles big data
Dimensionality reduction and sorting technique in technological means by building training pattern and test model, precisely and can be realized quickly
The function of non-intrusion type load monitoring.In addition, the method for the present invention principle is reliable, step is simple, before having very extensive application
Scape.
Embodiment 2
In electric system, SCADA system is most widely used, and technology development is also the most ripe.It is as Energy Management System
（EMS system）A most important subsystem, have information completely, improve efficiency, correctly grasp system running state, accelerate
Decision can help quick diagnosis to go out the advantages such as system fault condition, now have become the indispensable tool of power scheduling.It is right
Reliability, safety and the economic benefit of operation of power networks are improved, mitigates the burden of dispatcher, realizes power dispatching automation with showing
Dai Hua, the efficiency and horizontal aspect for improving scheduling have irreplaceable role.
Electric power data remote gathering system be build intelligent grid physical basis, system by emerging computer technology,
Communication and control technology and advanced sensing technology are applied to wherein, and remote collection so as to fulfill data completes data
Management, it is and for statistical analysis to data, timely find the exception in electric power data information, the electricity consumption to power consumer
Load is monitored and controls, and improves the electrical management efficiency and quality of electric company.
Electric power data remote system in the present embodiment as shown in figs. 1 and 4, by main website, communication channel, collecting device three
It is grouped as.
Step is carried out below（2）Establish the training pattern of non-intrusion type load monitoring.
Non-intrusion type load monitoring is divided into training and actual measurement two parts, wherein training is to form discriminant space, differentiates
Space is substantially general transient process database, and database need not be just trained again after being formed when surveying and judging；It is real
It is that the transient process to be decomposed that actual measurement obtains is projected to the discriminant space trained and obtained to survey part, to judge that it is temporary which class it belongs to
State process.
Non-intrusion type load monitoring is completed, first has to that the transient process of electric system is detected and detached, and then
Feature extraction is carried out to the transient process isolated, the feature of extraction is more, more can accurately characterize the spy of different transient processes
Point, that is, form high-dimensional feature space, the problem of will bringing in calculating and parsing along with high dimensional feature space, so needing
Eigenvectors matrix is subjected to dimension-reduction treatment, forms low-dimensional discriminant space, that is, forms transient process database.Non-intrusion type load
The training pattern flow of monitoring is as shown in Figure 2.
Its algorithm steps is expressed as：
<1>Comprehensively acquire the actual measurement sample of all kinds of transient processes.The step is mainly completed by SCADA system.
<2>The detection of transient state is carried out to above-mentioned sample with detaching.Current waveform is sampled according to certain frequency, and is being buffered
Area records the waveform of nearest fixed cycle numbers, if the variation that current strength occurs in real time is more than certain threshold value, judgement is temporary
State process starts, and until apparent variation is not occurring for real-time current intensity, then judges that transient process terminates, and records transient state
The waveform of process.
<3>Feature extraction forms feature space.Feature extraction is carried out to the transient state sample after separation, since transient state occurs
It is front and rear to bring the surge of the moment of current value and performance number and cause the bright of current effective value, active power and reactive power
Aobvious variation and the minor change of voltage effective value.Current effective value difference △ I, voltage in the present embodiment before and after selection transient state have
Dash current maximum value Imax, voltage in valid value difference △ V, active power difference △ P, reactive power difference △ Q, transient process
Maximum value Vmax, active power maximum value Pmax, reactive power maximum value Qmax, dash current maximum value and transient state presteady state electricity
Flowing the ratio I crest of virtual value, the ratio Vcrest of transient voltage maximum value and transient state presteady state voltage effective value, transient state has
Work(power maximum value and ratio Pcrest, transient reactive power maximum value and the transient state presteady state of transient state presteady state reactive power value
Class label belonging to the ratio Qcrest of reactive power value, transient state lasting periodicity T and load is as characteristic parameter, shape
Into the feature space of higher-dimension, and after removing unusual sample, normalization forms sample set
<4>With dimension reduction method, low-dimensional discriminant space is formed.For this sample set, with the Data Dimensionality Reduction side in big data technology
Method carries out feature space dimensionality reduction, so as to form the discriminant space of low-dimensional.
The classification problem of a multivariate statistics amount on non-intrusion type load monitoring question essence, solve such issues that
When, the dimension in space is the key factor for determining classification quality, on the one hand, feature space dimension is too low, can lead to feature extraction
Not comprehensively, some clusters can not effectively be distinguished；On the other hand, feature space dimension is excessively high and can bring asking in calculating and parsing
Topic, it is many in the effective calculating of lower dimensional space and analytic method, in higher dimensional space with regard to unworkable.Method of Data with Adding Windows at present
It is divided into two kinds：Linear dimensionality reduction and Method of Nonlinear Dimensionality Reduction.The application research be a multivariate statistics amount linear classification problem,
Have represent meaning to have principal component analysis in linear dimension reduction method（PCA）With linear discriminant analysis method（LDA）Algorithm.And PCA
Advantage with the minimum information of loss initial data, therefore PCA algorithms is taken to carry out linear dimensionality reduction in the present embodiment.
Then, it enters step（3）The actual measurement of non-intrusion type load monitoring.
After the completion of discriminant space structure, it is possible to be differentiated for the transient process of actual measurement.By the actual measurement sample of higher-dimension
Eigen vector according to best projection direction projection to low-dimensional discriminant space, after the completion of projection, it is also necessary to judge subpoint on earth
Belong to which cluster, choose suitable clustering algorithm carry out cluster can tell different transient processes.
The type of the electrical equipment in electric system is innumerable at present, in non-intrusion type load monitoring field, not yet
Form the standard of unified load classification.In addition to load is divided into first order load, two stage loads, three-level by the grade according to load
Outside load, resistive load, capacitive load and inductive load can also be divided into according to part throttle characteristics, can be drawn according to the purposes of load
It is divided into agriculture load, industrial load, resident living load and commercial drivers load.In the classification of household electrical appliance, it is generally divided into
Six major class：Resistance-type, pump type, motor driving, electronics feed, electronic power supply control and fluorescent lamp.The application is using big data
The advantage of technology is that it according to different demands, can establish the mould of different supervised or the study of non-supervisory formula
Type according to actual needs, carries out Active Learning, the clustering algorithm for the later stage establishes the base of decision according to different criteria for classifications
Plinth.
The Model Measured flow of non-intrusion type load monitoring is as shown in figure 3, its algorithm steps is expressed as：
<1>Always end detects transient process, and it is detached to system.
<2>Transient process after being detached to detection carries out feature extraction, forms feature space.
<3>The optimal projection direction obtained according to training projects to the discriminant space of low-dimensional.
<4>Judge that it belongs to the transient process which type load occurs with clustering algorithm.
Embodiment 3
A kind of method of non-intrusion type load monitoring based on big data proposed by the present invention, analyzes theory basis, and
Correlation model flow is listed, in the present embodiment, to obtain the transient-wave figure of electrical equipment needs to pass through following steps：
（1）The remote collection of electric power data by carrying out data monitoring and monitoring and controlling to electric system, acquires all kinds of temporary comprehensively
The actual measurement sample of state process；
（2）The training pattern of non-intrusion type load monitoring is established, forms discriminant space；
（3）The actual measurement of non-intrusion type load transient state monitoring is carried out, the transient process of actual measurement is differentiated according to discriminant space.
In the present embodiment, with electrical equipment refrigerator common in electric system, air compressor, television set and point
It for lamp, is modeled according to above step, obtains transient-wave figure as shown in Fig. 5 to 8, it can be seen that the beneficial effect of the application
Fruit is：
1）Detection and the separation algorithm of load transient process are proposed, can detect the generation of transient process, and it is negative to exclude other
The interference of lotus obtains the single transient-wave as shown in Fig. 5 to 8.
2）Give the characteristic quantity differentiated for transient process and corresponding computational methods, and with normalized characteristic quantity structure
Into high-dimensional feature space.
3）The application value of PCA dimension-reduction algorithms is analyzed, applies it to the dimensionality reduction of transient process feature space, by higher-dimension
Feature space is reduced to low-dimensional discriminant space, reduces the dimension of feature space.
The application non-intrusion type load caused by for current electric system internal loading type is more, transient process data volume is big
The problem of transient state discrimination of monitoring is low proposes a kind of method of the non-intrusion type load transient state monitoring based on big data technology,
Using big data technological means, from the Power system load data of the magnanimity of accumulation, the data of a large amount of transient process, foundation are obtained
The method of rational load classification chooses suitable characteristic parameter to build the feature decision space of higher-dimension, recycles big data
Dimensionality reduction and sorting technique in technological means by building training pattern and test model, precisely and can be realized quickly
The function of non-intrusion type load monitoring.
Above-mentioned technical proposal is one embodiment of the present invention, for those skilled in the art, at this
On the basis of disclosure of the invention application process and principle, it is easy to make various types of improvement or deformation, be not limited solely to this
Invent the described method of above-mentioned specific embodiment, therefore previously described mode is only preferred, and and without limitation
The meaning of property.
Claims (6)
- A kind of 1. method of the non-intrusion type load monitoring based on big data, it is characterised in that：Include the following steps：The remote collection of electric power data by carrying out data monitoring and monitoring and controlling to electric system, acquires all kinds of transient state comprehensively The actual measurement sample of process；The training pattern of non-intrusion type load monitoring is established, forms discriminant space；（3）The actual measurement of non-intrusion type load transient state monitoring is carried out, the transient process of actual measurement is differentiated according to discriminant space.
- 2. a kind of method of the non-intrusion type load monitoring based on big data as described in claim 1, it is characterised in that：It is described Step（1）In electric system host computer is mainly connected by central station, host computer is carried out through transmission channel by RTU control devices Data acquire.
- 3. a kind of method of the non-intrusion type load monitoring based on big data as described in claim 1, it is characterised in that：It is described Step（2）Include the following steps：（21）Collected transient state actual measurement sample is detected with detaching；（22）Feature extraction is carried out to the transient state sample after separation；（23）With dimension reduction method, low-dimensional discriminant space is formed.
- 4. a kind of method of the non-intrusion type load monitoring based on big data as claimed in claim 3, it is characterised in that：Step （21）Detection process in, current waveform is sampled according to certain frequency, and in the wave of the nearest fixed cycle numbers of buffer records Shape if the variation that current strength occurs in real time is more than certain threshold value, judges that transient process starts, until real-time current is strong Apparent variation is not occurring for degree, then judges that transient process terminates, and record the waveform of transient process.
- 5. a kind of method of the non-intrusion type load monitoring based on big data as claimed in claim 3, it is characterised in that：Step （22）In, the characteristic parameter before and after the transient state of extraction is formed to the feature space of higher-dimension, and after removing unusual sample, normalization Form sample set.
- 6. a kind of method of the non-intrusion type load monitoring based on big data as claimed in claim 3, it is characterised in that：Step （23）Feature space dimensionality reduction is carried out using the Method of Data with Adding Windows of big data technology, so as to form the discriminant space of low-dimensional.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710847363 | 2017-09-19 | ||
| CN2017108473632 | 2017-09-19 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108197817A true CN108197817A (en) | 2018-06-22 |
Family
ID=62589822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810050843.0A Pending CN108197817A (en) | 2017-09-19 | 2018-01-18 | A kind of method of the non-intrusion type load transient state monitoring based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108197817A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108921424A (en) * | 2018-06-28 | 2018-11-30 | 广东电网有限责任公司 | A kind of electric power data method for detecting abnormality, device, equipment and readable storage medium storing program for executing |
| CN109270368A (en) * | 2018-09-03 | 2019-01-25 | 四川长虹电器股份有限公司 | A kind of non-intrusion type electrical load recognition methods based on steady-state current |
| CN110426554A (en) * | 2019-08-09 | 2019-11-08 | 威胜集团有限公司 | Household loads electric characteristic extracting method, device and computer readable storage medium |
| CN110879537A (en) * | 2018-09-06 | 2020-03-13 | 珠海格力电器股份有限公司 | Method and device for processing transient stability of intelligent household energy system |
| CN111222768A (en) * | 2019-12-29 | 2020-06-02 | 杭州拓深科技有限公司 | Non-invasive load identification-electricity utilization behavior analysis electricity utilization judgment method and system |
| 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 |
| CN111722028A (en) * | 2019-03-19 | 2020-09-29 | 华北电力大学 | Load identification method based on high-frequency data |
| CN112446416A (en) * | 2020-10-09 | 2021-03-05 | 国网山东省电力公司日照供电公司 | Non-invasive load monitoring and identifying method and system |
| CN112564090A (en) * | 2020-11-20 | 2021-03-26 | 东北电力大学 | MBLDA-based AC/DC system transient voltage stability monitoring method |
| CN112701683A (en) * | 2020-12-23 | 2021-04-23 | 国网黑龙江省电力有限公司双鸭山供电公司 | Non-invasive load detection and decomposition method, system and storage medium |
| CN113034305A (en) * | 2021-02-10 | 2021-06-25 | 上海千居智科技有限公司 | Non-invasive load monitoring event classification method and storage medium |
| CN113297742A (en) * | 2021-05-27 | 2021-08-24 | 东北大学 | Self-energy non-invasive system based on independent component analysis and modeling method thereof |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106526280A (en) * | 2016-08-25 | 2017-03-22 | 北京电力经济技术研究院 | Testing method and separating method for railway power load impact transient process |
| CN106786534A (en) * | 2016-12-28 | 2017-05-31 | 天津求实智源科技有限公司 | A kind of non-intrusive electrical load transient process discrimination method and system |
-
2018
- 2018-01-18 CN CN201810050843.0A patent/CN108197817A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106526280A (en) * | 2016-08-25 | 2017-03-22 | 北京电力经济技术研究院 | Testing method and separating method for railway power load impact transient process |
| CN106786534A (en) * | 2016-12-28 | 2017-05-31 | 天津求实智源科技有限公司 | A kind of non-intrusive electrical load transient process discrimination method and system |
Non-Patent Citations (1)
| Title |
|---|
| 李雨轩: "非侵入式负荷分解算法的综合研究", 《中国优秀硕士学位论文全文数据库•工程科技II辑》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108921424A (en) * | 2018-06-28 | 2018-11-30 | 广东电网有限责任公司 | A kind of electric power data method for detecting abnormality, device, equipment and readable storage medium storing program for executing |
| CN109270368A (en) * | 2018-09-03 | 2019-01-25 | 四川长虹电器股份有限公司 | A kind of non-intrusion type electrical load recognition methods based on steady-state current |
| CN109270368B (en) * | 2018-09-03 | 2020-09-25 | 四川长虹电器股份有限公司 | Non-invasive power load identification method based on steady-state current |
| CN110879537A (en) * | 2018-09-06 | 2020-03-13 | 珠海格力电器股份有限公司 | Method and device for processing transient stability of intelligent household energy system |
| CN111722028A (en) * | 2019-03-19 | 2020-09-29 | 华北电力大学 | Load identification method based on high-frequency data |
| CN110426554A (en) * | 2019-08-09 | 2019-11-08 | 威胜集团有限公司 | Household loads electric characteristic extracting method, device and computer readable storage medium |
| CN111222768A (en) * | 2019-12-29 | 2020-06-02 | 杭州拓深科技有限公司 | Non-invasive load identification-electricity utilization behavior analysis electricity utilization judgment method and system |
| 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 |
| CN112446416A (en) * | 2020-10-09 | 2021-03-05 | 国网山东省电力公司日照供电公司 | Non-invasive load monitoring and identifying method and system |
| CN112564090A (en) * | 2020-11-20 | 2021-03-26 | 东北电力大学 | MBLDA-based AC/DC system transient voltage stability monitoring method |
| CN112564090B (en) * | 2020-11-20 | 2023-04-07 | 东北电力大学 | MBLDA-based AC/DC system transient voltage stability monitoring method |
| CN112701683A (en) * | 2020-12-23 | 2021-04-23 | 国网黑龙江省电力有限公司双鸭山供电公司 | Non-invasive load detection and decomposition method, system and storage medium |
| CN113034305A (en) * | 2021-02-10 | 2021-06-25 | 上海千居智科技有限公司 | Non-invasive load monitoring event classification method and storage medium |
| CN113297742A (en) * | 2021-05-27 | 2021-08-24 | 东北大学 | Self-energy non-invasive system based on independent component analysis and modeling method thereof |
| CN113297742B (en) * | 2021-05-27 | 2023-10-31 | 东北大学 | Self-energy non-invasive system based on independent component analysis and modeling method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108197817A (en) | A kind of method of the non-intrusion type load transient state monitoring based on big data | |
| CN103872782B (en) | A kind of power quality data integrated service system | |
| CN111624931B (en) | Industrial park electricity utilization internet intelligent operation and maintenance management and control system and method | |
| CN113570200B (en) | Power grid running state monitoring method and system based on multidimensional information | |
| CN104091070A (en) | Rail transit fault diagnosis method and system based on time series analysis | |
| CN112926749A (en) | Intelligent power grid information equipment monitoring system and method | |
| CN106682835B (en) | Data-driven complex electromechanical system service quality state evaluation method | |
| CN113036759B (en) | Fine granularity identification method and identification system for power consumer load | |
| CN113011089B (en) | Method and device for optimizing efficiency of mechanical production well system based on deep learning | |
| CN113466597A (en) | Intelligent detection method for equipment state of rail transit power supply system | |
| CN103729804A (en) | On-line decision support method for responding to power quality early warning | |
| CN113902104A (en) | Non-invasive load monitoring method combining unsupervised domain self-adaptive strategy and attention mechanism | |
| CN109711664B (en) | Power transmission and transformation equipment health assessment system based on big data | |
| CN109491339B (en) | Big data-based substation equipment running state early warning system | |
| CN111327118B (en) | Non-invasive power load identification method | |
| CN112615428A (en) | Line loss analysis and treatment system and method | |
| CN117559443A (en) | Ordered power utilization control method for large industrial user cluster under peak load | |
| CN117610934A (en) | Power grid security risk assessment early warning system | |
| CN103529337B (en) | The recognition methods of nonlinear correlation relation between equipment failure and electric quantity information | |
| CN114924518A (en) | Intelligent information processing system and method for water affairs | |
| Gurbuz et al. | Comprehensive non-intrusive load monitoring process: Device event detection, device feature extraction and device identification using KNN, random forest and decision tree | |
| CN106649034B (en) | Visual intelligent operation and maintenance method and platform | |
| CN111391694B (en) | Multi-level rapid data monitoring method and system for operation and maintenance of charging station | |
| Yao et al. | AdaBoost-CNN: a hybrid method for electricity theft detection | |
| CN114662576A (en) | Non-invasive storage battery car charging detection method and system based on supervised classification |
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: 20180622 |
|
| RJ01 | Rejection of invention patent application after publication |