CN110414839A - Load recognition methods and system based on quantum genetic algorithm and SVM model - Google Patents
Load recognition methods and system based on quantum genetic algorithm and SVM model Download PDFInfo
- Publication number
- CN110414839A CN110414839A CN201910686849.1A CN201910686849A CN110414839A CN 110414839 A CN110414839 A CN 110414839A CN 201910686849 A CN201910686849 A CN 201910686849A CN 110414839 A CN110414839 A CN 110414839A
- Authority
- CN
- China
- Prior art keywords
- svm
- load
- genetic algorithm
- quantum
- electric appliance
- 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
- 230000002068 genetic effect Effects 0.000 title claims abstract description 60
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000005457 optimization Methods 0.000 claims abstract description 12
- 210000000349 chromosome Anatomy 0.000 claims description 15
- 238000012544 monitoring process Methods 0.000 claims description 13
- 239000002096 quantum dot Substances 0.000 claims description 13
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 abstract description 2
- 240000002853 Nelumbo nucifera Species 0.000 description 4
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 4
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 4
- 238000005259 measurement Methods 0.000 description 3
- 239000000178 monomer Substances 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 244000131316 Panax pseudoginseng Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- Strategic Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Genetics & Genomics (AREA)
- Molecular Biology (AREA)
- Public Health (AREA)
- Physiology (AREA)
- Water Supply & Treatment (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
Abstract
The present invention relates to non-intrusion type electric appliance load identification technology fields, present invention seek to address that the problem that the existing electric appliance load recognition accuracy based on SVM model is not high, it is proposed a kind of load recognition methods based on quantum genetic algorithm and SVM model, the following steps are included: acquiring the load current and voltage data in predetermined period in real time, the electric current and voltage data are handled to obtain electric current valid data;SVM load identification model is created according to SVM algorithm, is optimized according to penalty factor and nuclear parameter of the quantum genetic algorithm to the SVM load identification model;The electric current valid data are input in the SVM load identification model after optimization, electric appliance load recognition result is obtained.The present invention it is existing be based on SVM load identification model on the basis of, optimized by penalty factor and nuclear parameter of the quantum genetic algorithm to SVM load identification model, improve load identification accuracy.
Description
Technical field
The present invention relates to non-intrusion type electric appliance load identification technology field, relate in particular to a kind of load recognition methods and
System.
Background technique
Electric energy is one of most widely used, most important energy in modern production life.It is traditional in terms of electrical energy measurement
" Every household has an ammeter " mode is to be copied to take electric energy meter and provide of that month consumption number by power department, and drawback is that user can not be known
Specific power consumption condition of certain electrical appliance within certain period.It can be said that the palm of the user to the dynamic realtime operation information of load collection
Hold also quite deficient, to solve this problem, electric load monitoring at present can be divided into two kinds:
Traditional intrusive monitoring mode installs power measurement hardware additional on each load to be measured, " one-to-one " monitor it is negative
Lotus operation information will expend a large amount of manpower object in installation, maintenance the disadvantage is that needing the original power supply circuit of failing load
Power.
Non-intrusion type load monitor system (Non-intrusive Load Monitoring System, NILMS) be
Power supply inlet install current-voltage measurement hardware, be not necessarily to failing load hardware configuration, can " one-to-many " monitor it is negative
Lotus operating condition, the Noninvasive testing power consumption that early stage proposes are based on electric appliance classification cell current, can only divide classification
Solution, cannot refine to specific electric appliance.And the transient characteristic data of electric appliance are depended on mostly, to hardware requirement height, cost is also
It is correspondingly improved, the popularization to product is unfavorable for;And some of which algorithm excessively it is complicated be inconvenient to be integrated into hardware set
In standby, training data needs to spend a large amount of human cost excessive early period.
Patent No. CN105974219A discloses a kind of classifying identification method of energy-saving electric appliance load type, and this method is logical
The feature class center vector for crossing SVM algorithm, AdaBoost algorithm and monomer energy-saving electric appliance obtains monomer energy-saving electric appliance training pattern,
Variable working condition load torque identification model is obtained according to the power factor change value of each monomer energy-saving electric appliance, passes through the combination die of two kinds of models
Type carries out the Classification and Identification of energy-saving electric appliance load type, but this method due to according to the feature class center vector of each electric appliance and
Penalty factor and core ginseng after the power factor change value of each electric appliance obtains SVM load identification model, in SVM load identification model
Number has just determined, if subsequent carry out load category identification all in accordance with the SVM load identification model, when electric appliance used by a user is negative
After large change or fluctuation occur for lotus type, it is larger to will cause load category identification error, the not high problem of recognition accuracy.
Summary of the invention
Present invention seek to address that the problem that the existing electric appliance load recognition accuracy based on SVM model is not high, proposes one
Load recognition methods and system of the kind based on quantum genetic algorithm and SVM model.
The technical proposal adopted by the invention to solve the above technical problems is that: based on quantum genetic algorithm and SVM model
Load recognition methods, which comprises the following steps:
Step 1. acquires load current and voltage data in predetermined period in real time, carries out to the electric current and voltage data
Processing obtains electric current valid data;
Step 2. creates SVM load identification model according to SVM algorithm, is known according to quantum genetic algorithm to the SVM load
The penalty factor and nuclear parameter of other model optimize;
The electric current valid data are input in the SVM load identification model after optimization by step 3., obtain electric appliance load
Recognition result.
Further, it is the accuracy for improving load type identification, the electric current valid data include: current effective value,
Current maxima, current minimum and harmonic data.
Further, described according to quantum genetic algorithm pair in step 2 to realize optimization to SVM load identification model
The penalty factor and nuclear parameter of the SVM load identification model optimize and include:
Step 21. configures quantum genetic algorithm parameter, and the quantum genetic algorithm parameter includes at least: Population Size, amount
Sub- bits number, population chromosome quantum bit coding and quantum bit probability amplitude;
Step 22. according to the probability amplitude construction quantum superposition state R={ a1, a2 ..., an } in chromosome each in population,
In, ai (i=1,2 .., n) is binary string;Quantum genetic algorithm decoding solves SVM load by observation quantum superposition state R and knows
The penalty factor of other model and the current iteration value of nuclear parameter;
The current iteration value of the penalty factor and nuclear parameter is input in SVM load identification model by step 23., is obtained
The discrimination of SVM load identification model constructs fitness function according to the discrimination, is fitted according to the fitness function
Response optimum value.
It further, is raising effect of optimization, further includes:
Judge whether the fitness optimum value meets termination iterated conditional, if so, terminating iteration, otherwise, throughput
After sub- revolving door carries out chromosome update and variation, 22 are entered step.
It further, is more intuitive display electric appliance load recognition result, the electric appliance load recognition result includes: electric appliance kind
The power consumption of class, each electric appliance quantity and each electric appliance in predetermined period.
Further, the abnormal power consumption condition to understand electric appliance convenient for user, further includes:
Real-time monitoring is carried out to each electric appliance power consumption condition according to electric appliance load recognition result, if a certain electric appliance power consumption is abnormal,
Then send the prompt information of corresponding electric appliance power consumption exception.
The present invention also proposes a kind of load identifying system based on quantum genetic algorithm and SVM model, comprising:
Acquisition unit, for acquiring load current and voltage data in predetermined period in real time;
Processing unit obtains electric current valid data for being handled the electric current and voltage data;
Server, for the penalty factor and nuclear parameter according to quantum genetic algorithm to the SVM load identification model of creation
It optimizes, and electric appliance load recognition result is obtained according to the SVM load identification model after the electric current valid data and optimization.
Further, the electric current valid data include: current effective value, current maxima, current minimum and harmonic wave
Data.
Further, it is described according to quantum genetic algorithm to the penalty factor and nuclear parameter of the SVM load identification model
It optimizes and includes:
Quantum genetic algorithm parameter is configured, the quantum genetic algorithm parameter includes at least: Population Size, quantum digit
Mesh, population chromosome quantum bit coding and quantum bit probability amplitude;
Quantum superposition state R={ a1, a2 ..., an } is constructed according to the probability amplitude in chromosome each in population, wherein ai (i
=1,2 .., n) it is binary string;Quantum genetic algorithm decoding solves SVM load identification model by observation quantum superposition state R
The current iteration value of penalty factor and nuclear parameter;
The current iteration value of the penalty factor and nuclear parameter is input in SVM load identification model, SVM load is obtained
The discrimination of identification model, constructs fitness function according to the discrimination, obtains fitness most according to the fitness function
Good value.
Further, further includes:
Power consumption monitoring modular, for carrying out real-time monitoring to each electric appliance power consumption condition according to electric appliance load recognition result, if
A certain electric appliance power consumption is abnormal, then sends the prompt information of corresponding electric appliance power consumption exception.
The beneficial effects of the present invention are: the load identification side of the present invention based on quantum genetic algorithm and SVM model
Method and system, it is existing be based on SVM load identification model on the basis of, by quantum genetic algorithm to SVM load identify mould
The penalty factor and nuclear parameter of type optimize, and obtain penalty factor and the optimal iterative value of nuclear parameter, and according to the SVM after optimization
Load identification model realizes the fast and accurately identification to electric appliance load type, improves the accuracy of load identification, ensure that
The stability and reliability of electric operation.
Detailed description of the invention
Fig. 1 is the process of the load recognition methods based on quantum genetic algorithm and SVM model described in the embodiment of the present invention
Schematic diagram;
Fig. 2 is the structure of the load identifying system based on quantum genetic algorithm and SVM model described in the embodiment of the present invention
Schematic diagram.
Specific embodiment
Embodiments of the present invention are described in detail below in conjunction with attached drawing.
Load recognition methods of the present invention based on quantum genetic algorithm and SVM model, comprising the following steps: step
S1. the load current and voltage data in predetermined period are acquired in real time, and the electric current and voltage data are handled to obtain electricity
Flow valid data;Step S2. creates SVM load identification model according to SVM algorithm, negative to the SVM according to quantum genetic algorithm
The penalty factor and nuclear parameter of lotus identification model optimize;After the electric current valid data are input to optimization by step S3.
In SVM load identification model, electric appliance load recognition result is obtained.
Firstly, power monitoring system is installed in the resident for needing to carry out load identification, it is real-time according to the pre- period
The electric current and voltage data of all electrical equipments of resident are acquired, and collected electric current and voltage data are converted to and can be counted
The digital signal of calculation, pre-processes electric current and voltage data, obtains the electric current valid data of algorithm needs, then, according to
The power factor change value of SVM algorithm, the feature class center vector of each electric appliance and each electric appliance creates SVM load identification model, and
According to quantum genetic algorithm in the SVM load identification model penalty factor and nuclear parameter optimize, finally, according to optimization
SVM load identification model afterwards carries out the load identification of electric appliance type.
Wherein, load current and voltage data can be acquired by current sensor and voltage sensor, by adopting
Collection chip acquires load current and voltage data in predetermined period in real time, can be by converter to the electric current and electricity of acquisition
Pressure data are AD converted, and are obtained electric current and the corresponding digital signal of voltage data, are then pre-processed to the digital signal,
The valid data of algorithm needs, such as current effective value, current maxima, current minimum and harmonic data are obtained, to electric current
The acquisition and processing of data and voltage data can locally be carried out in resident, and electric current valid data can be uploaded by communication module
It is performed corresponding processing to cloud server, e.g., GPRS communication module can reduce local computing pressure.
Beyond the clouds in server, become according to the power factor of SVM algorithm, the feature class center vector of each electric appliance and each electric appliance
Change value creates SVM load identification model, and specific creation method belongs to the prior art, and details are not described herein again, according to quantum genetic
Algorithm carries out load category identification to the SVM load identification model specifically:
Step S21. configures quantum genetic algorithm parameter, and the quantum genetic algorithm parameter includes at least: Population Size n,
Quantum bits number m, population P={ P1, P2 ..., Pn } chromosome Pi (i=1,2 ..., n) quantum bit coding and quantum bit are general
Rate width;
Step S22. according to the probability amplitude construction quantum superposition state R={ a1, a2 ..., an } in chromosome each in population,
In, ai (i=1,2 .., n) is binary string;Quantum genetic algorithm decoding solves SVM load by observation quantum superposition state R and knows
The penalty factor of other model and the current iteration value of nuclear parameter;
The current iteration value of the penalty factor and nuclear parameter is input in SVM load identification model by step S23., is obtained
The discrimination for obtaining SVM load identification model constructs fitness function according to the discrimination, is obtained according to the fitness function
Fitness optimum value;
Step S24. judges whether the fitness optimum value meets termination iterated conditional, if so, iteration is terminated, it is no
Then, after carrying out chromosome update and variation by Quantum rotating gate, 22 are entered step.
Generation is carried out according to quantum genetic algorithm, meets the fitness optimum value for terminating iterated conditional until obtaining, wherein eventually
Only iterated conditional can be the preset the number of iterations of completion or fitness optimum value meets preset range, obtain fitness most
After good value, i.e. the corresponding penalty factor of expression fitness optimum value and the discrimination of nuclear parameter is best, terminates iteration at this time, according to
Step S1 electric current valid data obtained are input to the SVM with best identified rate corresponding penalty factor and nuclear parameter
In load identification model, final electric appliance load recognition result is obtained, wherein electric appliance load recognition result may include: electric appliance
The power consumption of type, each electric appliance quantity and each electric appliance in predetermined period, cloud server can send out electric appliance load recognition result
It send to user terminal so that user checks, such as smart phone.
It optionally, can be according to the electric appliance load recognition result to each electric appliance after obtaining electric appliance load recognition result
Power consumption condition carries out real-time monitoring and sends the prompt information of corresponding electric appliance power consumption exception, electric appliance if a certain electric appliance power consumption is abnormal
Power consumption exception information also can be transmitted to user terminal, understand the electric appliance of abnormal power consumption in time convenient for user.
Based on the above-mentioned technical proposal, the present invention also proposes a kind of based on the knowledge of the load of quantum genetic algorithm and SVM model
Other system characterized by comprising
Acquisition unit, for acquiring load current and voltage data in predetermined period in real time;
Processing unit obtains electric current valid data for being handled the electric current and voltage data;
Server, for the penalty factor and nuclear parameter according to quantum genetic algorithm to the SVM load identification model of creation
It optimizes, and electric appliance load recognition result is obtained according to the SVM load identification model after the electric current valid data and optimization.
Optionally, the electric current valid data include: current effective value, current maxima, current minimum and harmonic number
According to.
Optionally, it is described according to quantum genetic algorithm to the penalty factor of the SVM load identification model and nuclear parameter into
Row optimizes
Quantum genetic algorithm parameter is configured, the quantum genetic algorithm parameter includes at least: Population Size, quantum digit
Mesh, population chromosome quantum bit coding and quantum bit probability amplitude;
Quantum superposition state R={ a1, a2 ..., an } is constructed according to the probability amplitude in chromosome each in population, wherein ai (i
=1,2 .., n) it is binary string;Quantum genetic algorithm decoding solves SVM load identification model by observation quantum superposition state R
The current iteration value of penalty factor and nuclear parameter;
The current iteration value of the penalty factor and nuclear parameter is input in SVM load identification model, SVM load is obtained
The discrimination of identification model, constructs fitness function according to the discrimination, obtains fitness most according to the fitness function
Good value.
Optionally, further includes: power consumption monitoring modular, for according to electric appliance load recognition result to each electric appliance power consumption condition into
Row real-time monitoring sends the prompt information of corresponding electric appliance power consumption exception if a certain electric appliance power consumption is abnormal.
It is appreciated that since the load identifying system of the present invention based on quantum genetic algorithm and SVM model is to use
In the system for realizing the load recognition methods based on quantum genetic algorithm and SVM model, for disclosed system, by
Method disclosed in Yu Qiyu is corresponding, so description is relatively simple, related place illustrates referring to the part of method.Due to
Therefore the accuracy that load identification is capable of in the above-mentioned load recognition methods based on quantum genetic algorithm and SVM model is realized above-mentioned
The system of load recognition methods based on quantum genetic algorithm and SVM model equally can be improved the accuracy of load identification.
Claims (10)
1. the load recognition methods based on quantum genetic algorithm and SVM model, which comprises the following steps:
Step 1. acquires load current and voltage data in predetermined period in real time, handles the electric current and voltage data
Obtain electric current valid data;
Step 2. creates SVM load identification model according to SVM algorithm, identifies mould to the SVM load according to quantum genetic algorithm
The penalty factor and nuclear parameter of type optimize;
The electric current valid data are input in the SVM load identification model after optimization by step 3., obtain the identification of electric appliance load
As a result.
2. the load recognition methods based on quantum genetic algorithm and SVM model as described in claim 1, which is characterized in that institute
Stating electric current valid data includes: current effective value, current maxima, current minimum and harmonic data.
3. the load recognition methods based on quantum genetic algorithm and SVM model as described in claim 1, which is characterized in that step
It is described that packet is optimized according to penalty factor and nuclear parameter of the quantum genetic algorithm to the SVM load identification model in rapid 2
It includes:
Step 21. configures quantum genetic algorithm parameter, and the quantum genetic algorithm parameter includes at least: Population Size, quantum bit
Number, population chromosome quantum bit coding and quantum bit probability amplitude;
Step 22. constructs quantum superposition state R={ a1, a2 ..., an } according to the probability amplitude in chromosome each in population, wherein ai
(i=1,2 .., n) is binary string;Quantum genetic algorithm decoding solves SVM load identification model by observation quantum superposition state R
Penalty factor and nuclear parameter current iteration value;
The current iteration value of the penalty factor and nuclear parameter is input in SVM load identification model by step 23., obtains SVM
The discrimination of load identification model constructs fitness function according to the discrimination, is adapted to according to the fitness function
Spend optimum value.
4. the load recognition methods based on quantum genetic algorithm and SVM model as claimed in claim 3, which is characterized in that also
Include:
Judge whether the fitness optimum value meets termination iterated conditional, is otherwise revolved by quantum if so, terminating iteration
After revolving door carries out chromosome update and variation, 22 are entered step.
5. the load recognition methods based on quantum genetic algorithm and SVM model as described in claim 1, which is characterized in that institute
State the power consumption that electric appliance load recognition result includes: electric appliance type, each electric appliance quantity and each electric appliance in predetermined period.
6. the load recognition methods based on quantum genetic algorithm and SVM model as described in claim 1, which is characterized in that also
Include:
Real-time monitoring is carried out to each electric appliance power consumption condition according to electric appliance load recognition result to send out if a certain electric appliance power consumption is abnormal
Send the prompt information of corresponding electric appliance power consumption exception.
7. the load identifying system based on quantum genetic algorithm and SVM model characterized by comprising
Acquisition unit, for acquiring load current and voltage data in predetermined period in real time;
Processing unit obtains electric current valid data for being handled the electric current and voltage data;
Server, for being carried out according to penalty factor and nuclear parameter of the quantum genetic algorithm to the SVM load identification model of creation
Optimization, and electric appliance load recognition result is obtained according to the SVM load identification model after the electric current valid data and optimization.
8. the load identifying system based on quantum genetic algorithm and SVM model as claimed in claim 7, which is characterized in that institute
Stating electric current valid data includes: current effective value, current maxima, current minimum and harmonic data.
9. the load identifying system based on quantum genetic algorithm and SVM model as claimed in claim 7, which is characterized in that institute
It states and optimizes according to penalty factor and nuclear parameter of the quantum genetic algorithm to the SVM load identification model and include:
Quantum genetic algorithm parameter is configured, the quantum genetic algorithm parameter includes at least: Population Size, quantum bits number, kind
Group's chromosome quantum bit coding and quantum bit probability amplitude;
Quantum superposition state R={ a1, a2 ..., an } is constructed according to the probability amplitude in chromosome each in population, wherein ai (i=1,
2 .., n) it is binary string;Quantum genetic algorithm decoding, the punishment of SVM load identification model is solved by observation quantum superposition state R
The current iteration value of the factor and nuclear parameter;
The current iteration value of the penalty factor and nuclear parameter is input in SVM load identification model, the identification of SVM load is obtained
The discrimination of model, constructs fitness function according to the discrimination, obtains fitness optimum value according to the fitness function.
10. the load identifying system based on quantum genetic algorithm and SVM model as claimed in claim 7, which is characterized in that also
Include:
Power consumption monitoring modular, for carrying out real-time monitoring to each electric appliance power consumption condition according to electric appliance load recognition result, if a certain
Electric appliance power consumption is abnormal, then sends the prompt information of corresponding electric appliance power consumption exception.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910686849.1A CN110414839A (en) | 2019-07-29 | 2019-07-29 | Load recognition methods and system based on quantum genetic algorithm and SVM model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910686849.1A CN110414839A (en) | 2019-07-29 | 2019-07-29 | Load recognition methods and system based on quantum genetic algorithm and SVM model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110414839A true CN110414839A (en) | 2019-11-05 |
Family
ID=68363619
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910686849.1A Pending CN110414839A (en) | 2019-07-29 | 2019-07-29 | Load recognition methods and system based on quantum genetic algorithm and SVM model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110414839A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111289817A (en) * | 2020-02-14 | 2020-06-16 | 珠海格力电器股份有限公司 | Method, device and system for monitoring faults of electric appliance and storage medium |
CN111612052A (en) * | 2020-05-12 | 2020-09-01 | 国网河北省电力有限公司电力科学研究院 | Non-invasive load decomposition method based on improved genetic algorithm |
CN113129163A (en) * | 2021-03-29 | 2021-07-16 | 上海思创电器设备有限公司 | Load monitoring system applied to algorithm core unit |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104537382A (en) * | 2015-01-12 | 2015-04-22 | 杭州电子科技大学 | Electromyographic signal gait recognition method for optimizing support vector machine based on genetic algorithm |
CN106053067A (en) * | 2016-05-24 | 2016-10-26 | 广东石油化工学院 | Bearing fault diagnosis method based on quantum genetic algorithm optimized support vector machine |
CN108416251A (en) * | 2018-01-08 | 2018-08-17 | 中国矿业大学 | Efficient human motion recognition method based on quantum genetic algorithm optimization |
CN109359665A (en) * | 2018-08-28 | 2019-02-19 | 中国农业大学 | A kind of family's electric load recognition methods and device based on support vector machines |
CN109633301A (en) * | 2018-12-03 | 2019-04-16 | 四川长虹电器股份有限公司 | Non-intrusion type electric appliance load recognition methods based on quantum genetic optimization |
CN109813978A (en) * | 2018-12-25 | 2019-05-28 | 武汉中原电子信息有限公司 | A kind of non-intruding load-type recognition methods of variation characteristic between comprehensive transient characteristic and stable state |
CN109902339A (en) * | 2019-01-18 | 2019-06-18 | 昆明理工大学 | A kind of Fault Diagnosis of Roller Bearings based on IAGA-SVM |
-
2019
- 2019-07-29 CN CN201910686849.1A patent/CN110414839A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104537382A (en) * | 2015-01-12 | 2015-04-22 | 杭州电子科技大学 | Electromyographic signal gait recognition method for optimizing support vector machine based on genetic algorithm |
CN106053067A (en) * | 2016-05-24 | 2016-10-26 | 广东石油化工学院 | Bearing fault diagnosis method based on quantum genetic algorithm optimized support vector machine |
CN108416251A (en) * | 2018-01-08 | 2018-08-17 | 中国矿业大学 | Efficient human motion recognition method based on quantum genetic algorithm optimization |
CN109359665A (en) * | 2018-08-28 | 2019-02-19 | 中国农业大学 | A kind of family's electric load recognition methods and device based on support vector machines |
CN109633301A (en) * | 2018-12-03 | 2019-04-16 | 四川长虹电器股份有限公司 | Non-intrusion type electric appliance load recognition methods based on quantum genetic optimization |
CN109813978A (en) * | 2018-12-25 | 2019-05-28 | 武汉中原电子信息有限公司 | A kind of non-intruding load-type recognition methods of variation characteristic between comprehensive transient characteristic and stable state |
CN109902339A (en) * | 2019-01-18 | 2019-06-18 | 昆明理工大学 | A kind of Fault Diagnosis of Roller Bearings based on IAGA-SVM |
Non-Patent Citations (2)
Title |
---|
涂京 等: "基于监督学习的非侵入式负荷监测算法比较", 《电力自动化设备》 * |
高辉: "基于量子遗传算法的支持向量机人脸识别技术", 《电脑知识与技术》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111289817A (en) * | 2020-02-14 | 2020-06-16 | 珠海格力电器股份有限公司 | Method, device and system for monitoring faults of electric appliance and storage medium |
CN111612052A (en) * | 2020-05-12 | 2020-09-01 | 国网河北省电力有限公司电力科学研究院 | Non-invasive load decomposition method based on improved genetic algorithm |
CN113129163A (en) * | 2021-03-29 | 2021-07-16 | 上海思创电器设备有限公司 | Load monitoring system applied to algorithm core unit |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110414839A (en) | Load recognition methods and system based on quantum genetic algorithm and SVM model | |
CN106936129B (en) | Power load identification method and system based on multi-feature fusion | |
WO2023168950A1 (en) | Data collection method and system for smart meter-reading terminal | |
CN110188826A (en) | Household electrical appliance operating status non-invasive inspection methods based on intelligent electric meter data | |
CN109359665B (en) | Household appliance load identification method and device based on support vector machine | |
CN109633301B (en) | Non-invasive electrical appliance load identification method based on quantum genetic optimization | |
Yu et al. | Non-intrusive adaptive load identification based on siamese network | |
CN105447082A (en) | Distributed clustering method for mass load curves | |
CN110011618A (en) | The diagnostic device of photovoltaic array failure based on fuzzy C-means clustering neural network | |
CN113036759A (en) | Power consumer load fine-grained identification method and identification system | |
CN116707331B (en) | Inverter output voltage high-precision adjusting method and system based on model prediction | |
Azaza et al. | Finite state machine household's appliances models for non-intrusive energy estimation | |
CN111563827A (en) | Load decomposition method based on electrical appliance physical characteristics and residential electricity consumption behaviors | |
CN114970633A (en) | LSTM-based non-invasive electrical appliance identification method, system and equipment | |
CN113884734B (en) | Non-invasive electricity consumption abnormality diagnosis method and device | |
CN110443481B (en) | Power distribution automation terminal state evaluation system and method based on hybrid K-nearest neighbor algorithm | |
CN116467648A (en) | Early monitoring method for nonlinear platform power failure based on Internet of things table | |
CN113361454B (en) | Non-supervision optimization-based deep learning non-invasive load monitoring method | |
de Souza et al. | An effective CPT-based nonintrusive load monitoring for cognitive meters | |
Han et al. | Non‐intrusive load monitoring by voltage–current trajectory enabled asymmetric deep supervised hashing | |
Zhou et al. | Non-intrusive extraction and forecasting of residential electric vehicle charging load | |
CN103427491A (en) | Automatic collecting and monitoring system for intelligent power grid data | |
Ma et al. | Deep flexible transmitter networks for non-intrusive load monitoring of power distribution networks | |
CN115828091A (en) | Non-invasive load identification method and system based on end-cloud cooperation | |
CN113269478A (en) | Concentrator abnormal data reminding method and system based on multiple models |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191105 |