CN106802379A - A kind of load switch detection of adaptive threshold and recognition methods and system - Google Patents
A kind of load switch detection of adaptive threshold and recognition methods and system Download PDFInfo
- Publication number
- CN106802379A CN106802379A CN201710127964.6A CN201710127964A CN106802379A CN 106802379 A CN106802379 A CN 106802379A CN 201710127964 A CN201710127964 A CN 201710127964A CN 106802379 A CN106802379 A CN 106802379A
- Authority
- CN
- China
- Prior art keywords
- power
- threshold value
- current
- predetermined threshold
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Current Or Voltage (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention discloses the load switch detection and recognition methods of a kind of adaptive threshold, including：Current analog signal and voltage analog signal to subscriber household electricity consumption are acquired；The current analog signal and voltage analog signal are processed, current digital signal and voltage digital signal is obtained；The power level difference in two neighboring cycle is calculated according to the current digital signal and voltage digital signal, and load is classified according to the power difference；The current strength difference in two neighboring cycle is calculated, and the current strength difference is compared with predetermined threshold value, judge whether the switch events for loading；Signal characteristic is extracted according to the voltage digital signal after the switch events for loading and current digital signal；And Classification and Identification and output loading type are carried out to load according to the signal characteristic.The present invention is classified according to different loads power, and sets different comparing threshold values, improves the accuracy rate that event detection is opened and closed to load.
Description
Technical field
The present invention relates to electric quantity metering field, and more particularly, to a kind of load switch inspection of phase adaptive threshold
Survey and recognition methods and system.
Background technology
Intelligent grid is the developing goal of following power network, and the realization of intelligent grid need to rely on non-intrusion type load and know
The realization of other technology.Non-intrusion type remained capacity technology, it is necessary to complete the detection to load switch state, load electrical feature
Extract and remained capacity.In the realization of non-intrusion type remained capacity technology, due to the diversity and complexity that load, it is difficult to
Finding a kind of method can accurately detect the on off state of all kinds of loads.
At present, the method for the switch detection of load has a lot, mainly there is Waveform Matching, current method, power method etc., should
With it is most common be current method and power method, but due to the diversity and complexity that load, take which kind of method all cannot
Ensure accuracy.The defect of presence is mainly:For low power load, the situation for missing inspection usually occur even can't detect
The situation of switch events；When usually there is repeated detection switch in powerful load；None of these methods can be simultaneous
Turn round and look at the detection of various power terminations；Load switch event detection is inaccurate, based on the load electrical carried out after switch events detection
Feature extraction and remained capacity work also cannot just ensure its accuracy.
Therefore, the degree of accuracy for how improving load switch event detection and remained capacity is urgent problem.
The content of the invention
The invention provides the load switch detection and recognition methods of a kind of adaptive threshold, by non-intrusion type remained capacity
Method is applied in intelligent appliance electricity consumption amount control system, solves load switch detection and the degree of accuracy of remained capacity is not high
Problem.
In order to solve the above problems, according to an aspect of the invention, there is provided a kind of load switch of adaptive threshold
Detection and recognition methods, methods described include：
Current analog signal and voltage analog signal to subscriber household electricity consumption are acquired；
The current analog signal and voltage analog signal are processed, current digital signal and voltage digital letter is obtained
Number；
The power level difference in two neighboring cycle, and root are calculated according to the current digital signal and voltage digital signal
Load is classified according to the power difference；
The current strength difference in two neighboring cycle is calculated, and the current strength difference is compared with predetermined threshold value
Compared with judging whether the switch events for loading, wherein the predetermined threshold value includes：It is high-power to open event predetermined threshold value, big work(
Rate is closed event predetermined threshold value, middle power and opens event predetermined threshold value, middle power and close event predetermined threshold value, small-power and opens the default threshold of event
Value and small-power close event predetermined threshold value；
Signal characteristic is extracted according to the voltage digital signal after the switch events for loading and current digital signal；And
Classification and Identification and output loading type are carried out to load according to the signal characteristic.
Preferably, wherein described processed the current analog signal and voltage analog signal, current digital is obtained
Signal and voltage digital signal include：
The voltage-mode that treatment is processed after filtering is filtered to the current analog signal and voltage analog signal
Intend signal and current analog signal；And
The voltage analog signal and current analog signal that will be processed after filtering are converted to voltage digital signal and electric current number
Word signal.
Preferably, wherein the work(that the two neighboring cycle is calculated according to the current digital signal and voltage digital signal
Rate strength difference, and load is carried out by classification included according to the power difference：
The performance number of each sample point is calculated according to the current digital signal and voltage digital signal；
Performance number according to the sample point calculates the mean power in each cycle；
The mean power in each cycle makes the difference with the mean power in previous cycle, and the power for obtaining the two neighboring cycle is strong
Degree difference；
The power level difference and predetermined power threshold range are compared, load are classified, wherein power
Strength difference [10w, 80w) corresponding low-power load, power level difference [80w, 400w) correspondence in power termination,
Corresponding high power load of the power level difference more than 400w.
Preferably, wherein the current strength difference for calculating two neighboring cycle, and by the current strength difference with
Predetermined threshold value is compared, and judges whether that the switch events for loading include：
The corresponding current strength difference of high power load is opened into event predetermined threshold value and high-power pass thing with high-power respectively
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than high-power, loads
Event is opened, is the pass event for loading if the current strength difference is less than high-power pass event predetermined threshold value；
The corresponding current strength difference of middle power termination is opened into event predetermined threshold value with middle power respectively and middle power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than middle power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than middle power；And
The corresponding current strength difference of low-power load is opened into event predetermined threshold value with small-power respectively and small-power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than small-power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than small-power.
Preferably, wherein the signal characteristic includes：Frequency-region signal feature and time domain signal characteristics, wherein the frequency domain is believed
Number feature includes：Frequency domain energy, harmonic wave vector sum harmonic distortion characteristic parameter, the time domain signal characteristics include：Electric current is maximum
Value, current average, electric current root mean square, form factor, crest factor, peak factor, peak-to-peak value and instantaneous power.
Preferably, wherein described carry out Classification and Identification and output loading type and include according to the signal characteristic to load：
According to the signal characteristic, load is identified using support vector machines algorithm and output loading type.
According to another aspect of the present invention, there is provided a kind of load switch detection of adaptive threshold and identifying system,
The system includes：Signal gathering unit, signal conversion unit, load taxon, switch events judging unit, signal characteristic
Extraction unit and remained capacity unit,
The signal gathering unit, current analog signal and voltage analog signal to subscriber household electricity consumption are acquired；
The signal conversion unit, is processed the current analog signal and voltage analog signal, obtains electric current number
Word signal and voltage digital signal；
The load taxon, the two neighboring cycle is calculated according to the current digital signal and voltage digital signal
Power level difference, and load is classified according to the power difference；
The switch events judging unit, calculates the current strength difference in two neighboring cycle, and by the current strength
Difference is compared with predetermined threshold value, judges whether the switch events for loading, wherein the predetermined threshold value includes：It is high-power
Open event predetermined threshold value, high-power pass event predetermined threshold value, middle power and open event predetermined threshold value, the middle power pass default threshold of event
Value, small-power open event predetermined threshold value and small-power closes event predetermined threshold value；
The signal characteristic abstraction unit, according to voltage digital signal and current digital after the switch events for loading
Signal extraction signal characteristic；And
The remained capacity unit, Classification and Identification and output loading type are carried out according to the signal characteristic to load.
Preferably, wherein the signal conversion unit is processed the current analog signal and voltage analog signal,
Obtaining current digital signal and voltage digital signal includes：
The voltage-mode that treatment is processed after filtering is filtered to the current analog signal and voltage analog signal
Intend signal and current analog signal；And
The voltage analog signal and current analog signal that will be processed after filtering are converted to voltage digital signal and electric current number
Word signal.
Preferably, wherein the load taxon calculates adjacent according to the current digital signal and voltage digital signal
Two power level differences in cycle, and load is carried out by classification included according to the power difference：
The performance number of each sample point is calculated according to the current digital signal and voltage digital signal；
Performance number according to the sample point calculates the mean power in each cycle；
The mean power in each cycle makes the difference with the mean power in previous cycle, and the power for obtaining the two neighboring cycle is strong
Degree difference；And
The power level difference and predetermined power threshold range are compared, load are classified, wherein power
Strength difference [10w, 80w) corresponding low-power load, power level difference [80w, 400w) correspondence in power termination,
Corresponding high power load of the power level difference more than 400w.
Preferably, wherein the switch events judging unit calculates the current strength difference in two neighboring cycle, and by institute
State current strength difference to be compared with predetermined threshold value, judge whether that the switch events for loading include：
The corresponding current strength difference of high power load is opened into event predetermined threshold value and high-power pass thing with high-power respectively
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than high-power, loads
Event is opened, is the pass event for loading if the current strength difference is less than high-power pass event predetermined threshold value；
The corresponding current strength difference of middle power termination is opened into event predetermined threshold value with middle power respectively and middle power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than middle power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than middle power；And
The corresponding current strength difference of low-power load is opened into event predetermined threshold value with small-power respectively and small-power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than small-power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than small-power.
Preferably, wherein the signal characteristic includes：Frequency-region signal feature and time domain signal characteristics, wherein the frequency domain is believed
Number feature includes：Frequency domain energy, harmonic wave vector sum harmonic distortion characteristic parameter, the time domain signal characteristics include：Electric current is maximum
Value, current average, electric current root mean square, form factor, crest factor, peak factor, peak-to-peak value and instantaneous power.
Preferably, wherein the remained capacity unit carries out Classification and Identification to load and exports negative according to the signal characteristic
Carrying type includes：Load is identified using support vector machines algorithm according to the signal characteristic and output loading type.
The beneficial effects of the present invention are：
1. classified according to different loads power, then set based on current method inspection according to different load classification types
The different threshold values of open and close are surveyed, the Automatic adjusument based on current method detection is realized, is improve and load is opened and closed
The accuracy rate of event detection.
2. after the load switch condition detection method by adaptive threshold detects that all kinds of loads are opened and closed, wink is extracted
Feature on state process frequency domain and time domain, the identification to loading is realized using SVM algorithm, improves the recognition accuracy of load.
Brief description of the drawings
By reference to the following drawings, illustrative embodiments of the invention can be more fully understood by：
Fig. 1 is the load switch detection and recognition methods 100 of a kind of adaptive threshold according to embodiment of the present invention
Flow chart；
Fig. 2 is the flow chart of the load sorting technique 200 according to embodiment of the present invention；And
Fig. 3 is load switch detection and the identifying system 300 of a kind of adaptive threshold according to embodiment of the present invention
Structural representation.
Specific embodiment
With reference now to accompanying drawing, illustrative embodiments of the invention are introduced, however, the present invention can use many different shapes
Formula is implemented, and is not limited to embodiment described herein, there is provided these embodiments are to disclose at large and fully
The present invention, and fully pass on the scope of the present invention to person of ordinary skill in the field.For showing for being illustrated in the accompanying drawings
Term in example property implementation method is not limitation of the invention.In the accompanying drawings, identical cells/elements are attached using identical
Icon is remembered.
Unless otherwise indicated, term (including scientific and technical terminology) used herein has to person of ordinary skill in the field
It is common to understand implication.Further it will be understood that the term limited with usually used dictionary, is appreciated that and it
The linguistic context of association area has consistent implication, and is not construed as Utopian or excessively formal meaning.
Fig. 1 is the load switch detection and recognition methods 100 of a kind of adaptive threshold according to embodiment of the present invention
Flow chart.As shown in figure 1, be applied to non-intrusion type load identification method in intelligent appliance electricity consumption amount control system by the present invention,
Solve the problems, such as that load switch detection and the degree of accuracy of remained capacity be not high.By the voltage that load is detected in load porch
Signal and current signal simultaneously calculate power；The mean power of power cycle and the power level difference of adjacent periods are obtained, and
Size according to the power level difference difference load is classified, and is then set not for different types of load
Same threshold value, realizes the Automatic adjusument of current method load detecting, greatly improves and opens and closes event detection to load
Accuracy rate, then extract the feature opened and closed in transient process frequency domain and time domain, the knowledge loaded using SVM algorithm
Not, improve the recognition accuracy of load.Load switch detection and the recognition methods 100 from step 101 of embodiment of the present invention
Place starts, and the current analog signal and voltage analog signal of subscriber household electricity consumption are acquired in step 101.
Preferably, processed in the step 102 pair current analog signal and voltage analog signal, obtain current digital
Signal and voltage digital signal.Preferably, wherein described processed the current analog signal and voltage analog signal, obtain
Obtaining current data signal and voltage digital signal include：Place is filtered to the current analog signal and voltage analog signal
Reason, to remove the noise in signal, the voltage analog signal and current analog signal for being processed after filtering；And will pass through
The voltage analog signal and current analog signal of filtering process are converted to voltage digital signal and current digital signal.
Preferably, the two neighboring cycle is calculated according to the current digital signal and voltage digital signal in step 103
Power level difference, and load is classified according to the power difference.Fig. 2 is according to the load of embodiment of the present invention point
The flow chart of class method 200.As shown in Fig. 2 the load sorting technique is since step 201 place, in step 201 according to described
Current digital signal and voltage digital signal calculate the performance number of each sample point.
Preferably, the mean power in each cycle is calculated according to the performance number of the sample point in step 202.
Preferably, the mean power in each cycle of step 203 makes the difference with the mean power in previous cycle, obtains adjacent
Two power level differences in cycle.
Preferably, the power level difference and predetermined power threshold range are compared in step 204, to load into
Row classification, wherein power level difference [10w, 80w) corresponding low-power load, power level difference [80w, 400w)
Power termination in correspondence, corresponding high power load of the power level difference more than 400w.In embodiments of the present invention, utilize
Voltage digital signal is multiplied with current digital signal, calculates the performance number of each sampled point, then to the power in each cycle
Mean power is obtained, allows each cycle to subtract the mean power in its previous cycle as average power difference value, will be average
Load of the power difference between 10w-80w is classified as low-power load, and load of the average power difference value between 80w-400w is returned
For mid power is loaded, average power difference value is classified as high power load in the load more than 400w.
Preferably, calculate the current strength difference in two neighboring cycle in step 104, and by the current strength difference with
Predetermined threshold value is compared, and judges whether the switch events for loading, wherein the predetermined threshold value includes：It is high-power to open event
Predetermined threshold value, high-power pass event predetermined threshold value, middle power open event predetermined threshold value, middle power and close event predetermined threshold value, small work(
Rate opens event predetermined threshold value and small-power closes event predetermined threshold value.Preferably, wherein the electric current for calculating the two neighboring cycle
Strength difference, and the current strength difference is compared with predetermined threshold value, judge whether the switch events bag for loading
Include：
The corresponding current strength difference of high power load is opened into event predetermined threshold value and high-power pass thing with high-power respectively
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than high-power, loads
Event is opened, is the pass event for loading if the current strength difference is less than high-power pass event predetermined threshold value；
The corresponding current strength difference of middle power termination is opened into event predetermined threshold value with middle power respectively and middle power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than middle power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than middle power；And
The corresponding current strength difference of low-power load is opened into event predetermined threshold value with small-power respectively and small-power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than small-power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than small-power.In this hair
In bright implementation method, judge that the high-power event threshold of opening of switch events is 17 for high power load setting, if the electricity
Intensity of flow difference is more than 17, then it is assumed that have high power load to open；The big work(for judging switch events is set for high power load
It is -17 that rate closes event threshold, if the current strength difference is less than -17, then it is assumed that there is high power load to close.
It is 1 to judge that the middle power of switch events opens event threshold for mid power load setting, if the current strength
Difference is more than 1, then it is assumed that have middle power termination to open；Set for mid power load and judge that the middle power of switch events closes thing
Part threshold value is -1, if the current strength difference is less than -1, then it is assumed that have middle power termination to close.
It is 0.07 to judge that the small-power of switch events opens event threshold for low-power load setting, if the current strength
Difference is more than 0.07, then it is assumed that have low-power load to open；Set for low-power load and judge that the small-power of switch events is closed
Event threshold is -0.07, if the current strength difference is less than -0.07, then it is assumed that there is low-power load to close.
Preferably, in step 105 according to the voltage digital signal and current digital signal after the switch events for loading
Extract signal characteristic.Preferably, wherein the signal characteristic includes：Frequency-region signal feature and time domain signal characteristics, wherein described
Frequency-region signal feature includes：Frequency domain energy, harmonic wave vector sum harmonic distortion characteristic parameter, the time domain signal characteristics include：Electricity
Stream maximum, current average, electric current root mean square, form factor, crest factor, peak factor, peak-to-peak value and instantaneous power.
Preferably, Classification and Identification and output loading type are carried out to load according to the signal characteristic in step 106.It is preferred that
Ground, wherein described carry out Classification and Identification and output loading type and include according to the signal characteristic to load：According to the signal
Feature, is identified and output loading type using support vector machines algorithm to load.In embodiments of the present invention,
In advance extract all kinds of load frequency domains and time domain on signal characteristic set up database as sample set, wherein the frequency-region signal
Feature includes：Frequency domain energy, harmonic wave vector sum harmonic distortion characteristic parameter, the time domain signal characteristics include：Current maxima,
Current average, electric current root mean square, form factor, crest factor, peak factor, peak-to-peak value and instantaneous power, then will be current
The signal characteristic that the transient process that is intercepted when the event of detecting is turned on and off is extracted as test set, and by the test set number
Remained capacity is carried out according to being input in LibSvm, when load matched degree is higher than 90%, is considered as load and is classified successfully and export negative
Carry type.
Fig. 3 is load switch detection and the identifying system 300 of a kind of adaptive threshold according to embodiment of the present invention
Structural representation.As shown in figure 3, the load switch detection and identifying system 300 include：Signal gathering unit 301, signal turns
Change unit 302, load taxon 303, switch events judging unit 304, signal characteristic abstraction unit 305 and remained capacity list
Unit 306.
Preferably, the current analog signal and voltage analog of subscriber household electricity consumption are believed in the signal gathering unit 301
Number it is acquired.
Preferably, processed in 302 pairs of current analog signals of the signal conversion unit and voltage analog signal,
Obtain current digital signal and voltage digital signal.Preferably, wherein in 302 pairs of current analogs of the signal conversion unit
Signal and voltage analog signal are processed, and obtaining current digital signal and voltage digital signal includes：To the current analog
Signal and voltage analog signal are filtered the voltage analog signal and current analog signal that treatment is processed after filtering；With
And the voltage analog signal that will process after filtering and current analog signal are converted to voltage digital signal and current digital signal.
Preferably, phase is calculated according to the current digital signal and voltage digital signal in the load taxon 303
Adjacent two power level differences in cycle, and load is classified according to the power difference.Preferably, wherein described negative
The power level that taxon 303 is carried according to the current digital signal and voltage digital signal calculating two neighboring cycle is poor
Being worth, and carry out classification to load according to the power difference includes：
The performance number of each sample point is calculated according to the current digital signal and voltage digital signal；
Performance number according to the sample point calculates the mean power in each cycle；
The mean power in each cycle makes the difference with the mean power in previous cycle, and the power for obtaining the two neighboring cycle is strong
Degree difference；And
The power level difference and predetermined power threshold range are compared, load are classified, wherein power
Strength difference [10w, 80w) corresponding low-power load, power level difference [80w, 400w) correspondence in power termination,
Corresponding high power load of the power level difference more than 400w.
Preferably, the current strength difference in two neighboring cycle is calculated in the switch events judging unit 304, and by institute
State current strength difference to be compared with predetermined threshold value, judge whether the switch events for loading, wherein the predetermined threshold value
Including：High-power event predetermined threshold value, high-power pass event predetermined threshold value, the middle power opened opens event predetermined threshold value, middle power pass
Event predetermined threshold value, small-power open event predetermined threshold value and small-power closes event predetermined threshold value.Preferably, wherein in the switch
Event judging unit 304 calculates the current strength difference in two neighboring cycle, and by the current strength difference and predetermined threshold value
It is compared, judges whether that the switch events for loading include：
The corresponding current strength difference of high power load is opened into event predetermined threshold value and high-power pass thing with high-power respectively
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than high-power, loads
Event is opened, is the pass event for loading if the current strength difference is less than high-power pass event predetermined threshold value；
The corresponding current strength difference of middle power termination is opened into event predetermined threshold value with middle power respectively and middle power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than middle power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than middle power；And
The corresponding current strength difference of low-power load is opened into event predetermined threshold value with small-power respectively and small-power closes thing
Part predetermined threshold value is compared, if the current strength difference opens event predetermined threshold value more than small-power, loads
Event is opened, is the pass event for loading if the current strength difference closes event predetermined threshold value less than small-power.
Preferably, in the signal characteristic abstraction unit 305 according to the voltage digital letter after the switch events for loading
Number and current digital signal extract signal characteristic.Preferably, wherein the signal characteristic includes：Frequency-region signal feature and time domain are believed
Number feature, wherein the frequency-region signal feature includes：Frequency domain energy, harmonic wave vector sum harmonic distortion characteristic parameter, the time domain
Signal characteristic includes：Current maxima, current average, electric current root mean square, form factor, crest factor, peak factor, peak-
Peak value and instantaneous power.
Preferably, load is carried out by Classification and Identification and is exported according to the signal characteristic in the remained capacity unit 306
Loadtype.Preferably, wherein the remained capacity unit carries out Classification and Identification and export according to the signal characteristic to load
Loadtype includes：Load is identified using support vector machines algorithm according to the signal characteristic and output loading class
Type.
A kind of load switch detection of adaptive threshold of embodiments of the invention and identifying system 300 are another with of the invention
The load switch detection and recognition methods 100 of a kind of adaptive threshold of one embodiment are corresponding, will not be repeated here.
The present invention is described by reference to a small amount of implementation method.However, it is known in those skilled in the art, as
What subsidiary Patent right requirement was limited, except the present invention other embodiments disclosed above equally fall of the invention
In the range of.
Normally, all terms for using in the claims are all solved according to them in the usual implication of technical field
Release, unless clearly defined in addition wherein.It is all of to be all opened ground with reference to " one/described/be somebody's turn to do [device, component etc.] "
At least one of described device, component etc. example is construed to, unless otherwise expressly specified.Any method disclosed herein
Step need not all be run with disclosed accurate order, unless explicitly stated otherwise.
Claims (12)
1. the load switch detection and recognition methods of a kind of adaptive threshold, it is characterised in that methods described includes：
Current analog signal and voltage analog signal to subscriber household electricity consumption are acquired；
The current analog signal and voltage analog signal are processed, current digital signal and voltage digital signal is obtained；
The power level difference in two neighboring cycle is calculated according to the current digital signal and voltage digital signal, and according to institute
Power difference is stated to classify load；
The current strength difference in two neighboring cycle is calculated, and the current strength difference is compared with predetermined threshold value, sentenced
The disconnected switch events for whether loading, wherein the predetermined threshold value includes：It is high-power to open event predetermined threshold value, high-power pass thing
Part predetermined threshold value, middle power open event predetermined threshold value, middle power and close event predetermined threshold value, small-power and opens event predetermined threshold value and small
Power closes event predetermined threshold value；
Signal characteristic is extracted according to the voltage digital signal after the switch events for loading and current digital signal；And
Classification and Identification and output loading type are carried out to load according to the signal characteristic.
2. method according to claim 1, it is characterised in that described to the current analog signal and voltage analog signal
Processed, obtaining current digital signal and voltage digital signal includes：
The voltage analog letter that treatment is processed after filtering is filtered to the current analog signal and voltage analog signal
Number and current analog signal；And
The voltage analog signal and current analog signal that will be processed after filtering are converted to voltage digital signal and current digital letter
Number.
3. method according to claim 1, it is characterised in that described to be believed according to the current digital signal and voltage digital
Number calculating the power level difference in two neighboring cycle, and carry out classification to load according to the power difference includes：
The performance number of each sample point is calculated according to the current digital signal and voltage digital signal；
Performance number according to the sample point calculates the mean power in each cycle；
The mean power in each cycle makes the difference with the mean power in previous cycle, and the power level for obtaining the two neighboring cycle is poor
Value；
The power level difference and predetermined power threshold range are compared, load are classified, wherein power level
Difference [10w, 80w) corresponding low-power load, power level difference [80w, 400w) correspondence in power termination, power
Corresponding high power load of the strength difference more than 400w.
4. method according to claim 1, it is characterised in that the current strength difference in the calculating two neighboring cycle,
And the current strength difference is compared with predetermined threshold value, judge whether that the switch events for loading include：
The corresponding current strength difference of high power load is opened into event predetermined threshold value and high-power pass event is pre- with high-power respectively
If threshold value is compared, if the current strength difference opens event predetermined threshold value more than high-power, thing is opened for what is loaded
Part, is the pass event for loading if the current strength difference is less than high-power pass event predetermined threshold value；
The corresponding current strength difference of middle power termination is opened into event predetermined threshold value with middle power respectively and middle power pass event is pre-
If threshold value is compared, if the current strength difference opens event predetermined threshold value more than middle power, thing is opened for what is loaded
Part, is the pass event for loading if the current strength difference closes event predetermined threshold value less than middle power；And
The corresponding current strength difference of low-power load is opened into event predetermined threshold value with small-power respectively and small-power pass event is pre-
If threshold value is compared, if the current strength difference opens event predetermined threshold value more than small-power, thing is opened for what is loaded
Part, is the pass event for loading if the current strength difference closes event predetermined threshold value less than small-power.
5. method according to claim 1, it is characterised in that the signal characteristic includes：Frequency-region signal feature and time domain
Signal characteristic, wherein the frequency-region signal feature includes：Frequency domain energy, harmonic wave vector sum harmonic distortion characteristic parameter, when described
Domain signal characteristic includes：Current maxima, current average, electric current root mean square, form factor, crest factor, peak factor,
Peak-to-peak value and instantaneous power.
6. method according to claim 1, it is characterised in that described that classification knowledge is carried out to load according to the signal characteristic
Not simultaneously output loading type does not include：According to the signal characteristic, load is identified and defeated using support vector machines algorithm
Go out loadtype.
7. a kind of load switch of adaptive threshold is detected and identifying system, it is characterised in that the system includes：Signal acquisition
Unit, signal conversion unit, load taxon, switch events judging unit, signal characteristic abstraction unit and remained capacity list
Unit,
The signal gathering unit, current analog signal and voltage analog signal to subscriber household electricity consumption are acquired；
The signal conversion unit, is processed the current analog signal and voltage analog signal, obtains current digital letter
Number and voltage digital signal；
The load taxon, the power in two neighboring cycle is calculated according to the current digital signal and voltage digital signal
Strength difference, and load is classified according to the power difference；
The switch events judging unit, calculates the current strength difference in two neighboring cycle, and by the current strength difference
It is compared with predetermined threshold value, judges whether the switch events for loading, wherein the predetermined threshold value includes：It is high-power to open thing
Part predetermined threshold value, high-power pass event predetermined threshold value, middle power open event predetermined threshold value, middle power and close event predetermined threshold value, small
Power opens event predetermined threshold value and small-power closes event predetermined threshold value；
The signal characteristic abstraction unit, according to voltage digital signal and current digital signal after the switch events for loading
Extract signal characteristic；And
The remained capacity unit, Classification and Identification and output loading type are carried out according to the signal characteristic to load.
8. system according to claim 7, it is characterised in that the signal conversion unit to the current analog signal and
Voltage analog signal is processed, and obtaining current digital signal and voltage digital signal includes：
The voltage analog letter that treatment is processed after filtering is filtered to the current analog signal and voltage analog signal
Number and current analog signal；And
The voltage analog signal and current analog signal that will be processed after filtering are converted to voltage digital signal and current digital letter
Number.
9. system according to claim 7, it is characterised in that the load taxon is according to the current digital signal
With the power level difference that voltage digital signal calculates two neighboring cycle, and load is classified according to the power difference
Including：
The performance number of each sample point is calculated according to the current digital signal and voltage digital signal；
Performance number according to the sample point calculates the mean power in each cycle；
The mean power in each cycle makes the difference with the mean power in previous cycle, and the power level for obtaining the two neighboring cycle is poor
Value；And
The power level difference and predetermined power threshold range are compared, load are classified, wherein power level
Difference [10w, 80w) corresponding low-power load, power level difference [80w, 400w) correspondence in power termination, power
Corresponding high power load of the strength difference more than 400w.
10. system according to claim 7, it is characterised in that the switch events judging unit calculates two neighboring week
The current strength difference of phase, and the current strength difference is compared with predetermined threshold value, judge whether that what is loaded opens
Pass event includes：
The corresponding current strength difference of high power load is opened into event predetermined threshold value and high-power pass event is pre- with high-power respectively
If threshold value is compared, if the current strength difference opens event predetermined threshold value more than high-power, thing is opened for what is loaded
Part, is the pass event for loading if the current strength difference is less than high-power pass event predetermined threshold value；
The corresponding current strength difference of middle power termination is opened into event predetermined threshold value with middle power respectively and middle power pass event is pre-
If threshold value is compared, if the current strength difference opens event predetermined threshold value more than middle power, thing is opened for what is loaded
Part, is the pass event for loading if the current strength difference closes event predetermined threshold value less than middle power；And
The corresponding current strength difference of low-power load is opened into event predetermined threshold value with small-power respectively and small-power pass event is pre-
If threshold value is compared, if the current strength difference opens event predetermined threshold value more than small-power, thing is opened for what is loaded
Part, is the pass event for loading if the current strength difference closes event predetermined threshold value less than small-power.
11. systems according to claim 7, it is characterised in that the signal characteristic includes：Frequency-region signal feature and time domain
Signal characteristic, wherein the frequency-region signal feature includes：Frequency domain energy, harmonic wave vector sum harmonic distortion characteristic parameter, when described
Domain signal characteristic includes：Current maxima, current average, electric current root mean square, form factor, crest factor, peak factor,
Peak-to-peak value and instantaneous power.
12. systems according to claim 7, it is characterised in that the remained capacity unit is according to the signal characteristic pair
Load carries out Classification and Identification and output loading type to be included：According to the signal characteristic using support vector machines algorithm to negative
Load is identified and output loading type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710127964.6A CN106802379B (en) | 2017-03-06 | 2017-03-06 | The load switch detection and recognition methods of a kind of adaptive threshold and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710127964.6A CN106802379B (en) | 2017-03-06 | 2017-03-06 | The load switch detection and recognition methods of a kind of adaptive threshold and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106802379A true CN106802379A (en) | 2017-06-06 |
| CN106802379B CN106802379B (en) | 2019-07-09 |
Family
ID=58987830
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710127964.6A Active CN106802379B (en) | 2017-03-06 | 2017-03-06 | The load switch detection and recognition methods of a kind of adaptive threshold and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106802379B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107478926A (en) * | 2017-07-14 | 2017-12-15 | 宁波三星医疗电气股份有限公司 | A kind of electricity consumption monitoring arrangement and method |
| CN108181533A (en) * | 2018-01-18 | 2018-06-19 | 中国海洋大学 | A kind of electric load intelligent measurement of non-intrusion type and hierarchical classification method |
| CN108459222A (en) * | 2018-03-22 | 2018-08-28 | 中国海洋大学 | A kind of electric appliance fault detection method and system |
| CN108872666A (en) * | 2018-09-20 | 2018-11-23 | 广东石油化工学院 | A kind of load switch event detecting method and system |
| CN110196354A (en) * | 2019-04-23 | 2019-09-03 | 广东石油化工学院 | A kind of detection method and device of the switch events of load |
| CN110751385A (en) * | 2019-10-08 | 2020-02-04 | 威胜集团有限公司 | Non-invasive load identification method, terminal device and storage medium |
| CN112034245A (en) * | 2020-11-03 | 2020-12-04 | 天津滨电电力工程有限公司 | Household washing machine identification method for embedded terminal |
| CN112611931A (en) * | 2020-12-23 | 2021-04-06 | 南方电网电力科技股份有限公司 | Method, system and storage medium for identifying and processing malignant load based on discrete waveform |
| CN113131608A (en) * | 2019-12-30 | 2021-07-16 | 华为数字技术（苏州）有限公司 | Method, device and power distribution system for managing power distribution system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120136593A1 (en) * | 2009-07-30 | 2012-05-31 | Intelligent Sustainable Energy Limited | Identifying the operation of a specified type of appliance |
| CN204101613U (en) * | 2014-10-20 | 2015-01-14 | 中国长江电力股份有限公司 | A kind of non-contact switch amount condition checkout gear |
| CN105823948A (en) * | 2016-05-18 | 2016-08-03 | 国网上海市电力公司 | Non-invasive resident load identification method |
| CN106124850A (en) * | 2016-07-20 | 2016-11-16 | 国网江西省电力公司南昌供电分公司 | A kind of load identification system for domestic intelligent electricity meter based on similarity algorithm and load recognition methods thereof |
-
2017
- 2017-03-06 CN CN201710127964.6A patent/CN106802379B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120136593A1 (en) * | 2009-07-30 | 2012-05-31 | Intelligent Sustainable Energy Limited | Identifying the operation of a specified type of appliance |
| CN204101613U (en) * | 2014-10-20 | 2015-01-14 | 中国长江电力股份有限公司 | A kind of non-contact switch amount condition checkout gear |
| CN105823948A (en) * | 2016-05-18 | 2016-08-03 | 国网上海市电力公司 | Non-invasive resident load identification method |
| CN106124850A (en) * | 2016-07-20 | 2016-11-16 | 国网江西省电力公司南昌供电分公司 | A kind of load identification system for domestic intelligent electricity meter based on similarity algorithm and load recognition methods thereof |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107478926A (en) * | 2017-07-14 | 2017-12-15 | 宁波三星医疗电气股份有限公司 | A kind of electricity consumption monitoring arrangement and method |
| CN107478926B (en) * | 2017-07-14 | 2020-10-09 | 宁波三星医疗电气股份有限公司 | Power utilization monitoring device and method |
| CN108181533A (en) * | 2018-01-18 | 2018-06-19 | 中国海洋大学 | A kind of electric load intelligent measurement of non-intrusion type and hierarchical classification method |
| CN108459222A (en) * | 2018-03-22 | 2018-08-28 | 中国海洋大学 | A kind of electric appliance fault detection method and system |
| CN108872666A (en) * | 2018-09-20 | 2018-11-23 | 广东石油化工学院 | A kind of load switch event detecting method and system |
| CN110196354A (en) * | 2019-04-23 | 2019-09-03 | 广东石油化工学院 | A kind of detection method and device of the switch events of load |
| CN110196354B (en) * | 2019-04-23 | 2021-09-17 | 广东石油化工学院 | Method and device for detecting switching event of load |
| CN110751385A (en) * | 2019-10-08 | 2020-02-04 | 威胜集团有限公司 | Non-invasive load identification method, terminal device and storage medium |
| CN113131608A (en) * | 2019-12-30 | 2021-07-16 | 华为数字技术（苏州）有限公司 | Method, device and power distribution system for managing power distribution system |
| CN112034245A (en) * | 2020-11-03 | 2020-12-04 | 天津滨电电力工程有限公司 | Household washing machine identification method for embedded terminal |
| CN112611931A (en) * | 2020-12-23 | 2021-04-06 | 南方电网电力科技股份有限公司 | Method, system and storage medium for identifying and processing malignant load based on discrete waveform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106802379B (en) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106802379B (en) | The load switch detection and recognition methods of a kind of adaptive threshold and system | |
| US10145885B2 (en) | Electrical leakage detection apparatus | |
| CN106569129B (en) | Motor safety detection method and system and electric machine control system | |
| CN111027408A (en) | Load identification method based on support vector machine and V-I curve characteristics | |
| CN101706527A (en) | Method for detecting arc faults based on time-frequency characteristics of high-frequency current component | |
| CN105510687A (en) | Empirical mode decomposition-based voltage anomaly characteristic identification method | |
| CN106908671A (en) | A kind of non-intrusion type household loads intelligent detecting method and system | |
| CN110516788A (en) | Bi-LSTM algorithm-based non-invasive load detection method and system | |
| CN113567794B (en) | Electric bicycle indoor charging identification method and system based on dynamic time warping | |
| CN210111945U (en) | Arc fault detection system on direct current side of photovoltaic system | |
| CN106855597B (en) | A kind of non-intrusion type power quality interference source online adaptive monitoring system and method | |
| CN101819261B (en) | Three-phase source phase failure and power failure detection circuit | |
| CN108181533A (en) | A kind of electric load intelligent measurement of non-intrusion type and hierarchical classification method | |
| CN109034277A (en) | Power Quality Disturbance Classification Method based on multi-feature fusion and system | |
| CN114584069B (en) | Photovoltaic direct-current arc detection method and system | |
| CN106908657A (en) | A kind of load switch detection means and method based on power | |
| CN116724241A (en) | Battery system arc discharge detection method, detection device and battery energy storage system | |
| CN114236234A (en) | Electrical appliance characteristic identification method based on fundamental wave and harmonic wave mixed criterion | |
| Bilski et al. | The rule-based method for the non-intrusive electrical appliances identification | |
| CN104778348A (en) | Voltage sag action characteristic modeling method of low-voltage release | |
| CN108459222A (en) | A kind of electric appliance fault detection method and system | |
| CN114002550A (en) | Direct-current power distribution network ground fault line selection method and system | |
| CN115112989B (en) | Non-invasive load monitoring method based on low-frequency data | |
| CN105891635A (en) | Digital signal processing-based electric appliance type identification device and method | |
| CN110703124B (en) | Method for detecting and protecting three-phase power supply and load phase loss |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |