CN109981011A - A kind of generator parameter identification method - Google Patents
A kind of generator parameter identification method Download PDFInfo
- Publication number
- CN109981011A CN109981011A CN201910266221.6A CN201910266221A CN109981011A CN 109981011 A CN109981011 A CN 109981011A CN 201910266221 A CN201910266221 A CN 201910266221A CN 109981011 A CN109981011 A CN 109981011A
- Authority
- CN
- China
- Prior art keywords
- streams
- river
- ocean
- identified
- cost function
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P9/00—Arrangements for controlling electric generators for the purpose of obtaining a desired output
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P2103/00—Controlling arrangements characterised by the type of generator
- H02P2103/20—Controlling arrangements characterised by the type of generator of the synchronous type
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Control Of Eletrric Generators (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The present invention relates to a kind of generator parameter identification methods, characterized by the following steps: (1) according to the synchronous motor system to be recognized, Synchronous Machine Models are established, signal to be identified is obtained and calculate the damping time constant of the DC component of the signal to be identified；(2) water round-robin algorithm is used, the Synchronous Machine Models based on foundation recognize the other parameters in the signal to be identified of acquisition, obtain the optimal result of PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE.The present invention has good convergence and higher parameter identification precision, and compared with using the parameter identification result of genetic algorithm, this method has the advantages that parameter setting is few, convergence is good and Identification Errors are small, can be widely applied to generator parameter identification field.
Description
Technical field
The present invention relates to a kind of generator parameter identification methods, especially with regard to a kind of synchronous hair based on water round-robin algorithm
Nonlinear model for minor parameter identification method.
Background technique
Synchronous generator is the important equipment of electric system, and the accuracy of its parameter directly influences Power System Analysis
Calculating, Control System Design, stable operation and control.It is influenced by multiple factors, the parameter that producer provides is difficult to accurately describe same
Walk the practical dynamic process of generator.Field engineering in practice, mainly obtains synchronous motor by three-phase sudden short circuit test
Transient parameter.
Parameter identification is the important means for obtaining Generator Parameters, and common parameter identification method has: least square
Method, Prony (Pu Luoni) method and intelligent optimization algorithm.Least square method is very sensitive to noise, is easily trapped into local optimum, ginseng
There are large errors for number identification；Prony method is very sensitive to noise, and order determines difficulty, seriously affects parameter identification precision.
Genetic algorithm is replicated, is intersected and mutation operation in intelligent optimization algorithm, and evolutionary rate is slower, precocious receive easily occurs
It holds back, and its performance has biggish dependence to parameter；Particle swarm optimization algorithm is easily trapped into locally optimal solution, generates precocious existing
As, and optimization performance has biggish dependence to parameter setting.
Water round-robin algorithm (Water Cycle Algorithm, WCA) is inspired in the Nature, is the Nature according to the observation
Water is flowed to the process of ocean by river, river, lake in water cycle process, and a kind of novel optimization algorithm proposed.The algorithm has
Good convergence rate, convergence precision and stability can pick out the parameter of synchronous generator nonlinear model well.
Summary of the invention
In view of the above-mentioned problems, being based on water round-robin algorithm the object of the present invention is to provide a kind of generator parameter identification method
The parameter of synchronous generator is recognized, there is good convergence and higher parameter identification precision.
To achieve the above object, the present invention takes following technical scheme: a kind of generator parameter identification method comprising with
Lower step:
(1) according to the synchronous motor system to be recognized, Synchronous Machine Models are established, obtain signal to be identified and are calculated
The damping time constant T of the DC component of signal to be identifieda；
(2) water round-robin algorithm, the Synchronous Machine Models based on foundation, to the other parameters of the signal to be identified of acquisition are used
It is recognized, obtains the optimal result of PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE；The other parameters include longitudinal axis ultra-transient reactance x "d, it is horizontal
Axis ultra-transient reactance x "q, d-axis transient reactance x 'd, the longitudinal axis surpass transient time constant T "d, longitudinal axis transient time constant T 'd, short circuit
Initial phase angle φ0, synchronous reactance xd。
Further, in the step (1), according to the synchronous motor system to be recognized, Synchronous Machine Models is established, are obtained
The method for taking signal to be identified and calculating the damping time constant of the DC component of signal to be identified, comprising the following steps:
(1.1) synchronous motor system to be recognized is analyzed, the Synchronous Machine Models to tally with the actual situation are established, it is described same
Walk the identification range in motor model comprising whole parameters and each parameter to be identified to be identified；
(1.2) obtain short circuit current signal when three-phase suddenly-applied short circuit occurs for the synchronous motor system to be recognized as to
Identification signal；
(1.3) according to the signal to be identified of acquisition, the damping time constant T of its DC component is calculateda。
Further, in the step (1.3), the damping time constant T of the DC component of parameter to be identified is calculatedaSide
Method are as follows:
The coenvelope line and lower envelope line for extracting short circuit current first, calculate the half of the sum of envelope, obtain direct current point
Measure i0；
Secondly to DC component i0Absolute value take logarithm obtain ln (|-i0|)；
Then it carries out curve fitting, obtains the damping time constant T of DC componenta。
Further, in the step (2), using water round-robin algorithm, the Synchronous Machine Models based on foundation, to acquisition
Other parameters in signal to be identified are recognized, the method for obtaining the optimal result of PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE, including following
Step:
(2.1) cost function calculation formula is determined；
(2.2) water round-robin algorithm control parameter, including rainfall layer number N are setpop, river and ocean sum Nsr, minimum
Initial value dmax, water round-robin algorithm maximum number of iterations T, Optimal Parameters number Nvar, and parameter or empirical value are provided according to producer
Range is recognized for each parameter setting to be identified；
(2.3) initial population is generated at random, forms initial streams, river and ocean；
(2.4) streams population dividing is multiple streams layers as model parameter and is separately input to Synchronous Machine Models, counted
It calculates Synchronous Machine Models and signal when three-phase suddenly-applied short circuit occurs, and calculate the cost function value J of each streams layeri；
(2.5) size of more each streams layer cost function, selects the smallest cost function value JiCorresponding streams layer
As ocean, N is selectedRiverA small cost function value JiCorresponding streams layer is used as river, and determine flow to specified river and
The streams number of ocean；
(2.6) position in river is flowed to streams respectively, streams flows to the position of ocean and the position of river direction ocean
It sets and is updated, and position exchange is carried out according to the cost function value of streams each after update, river, ocean；
(2.7) judge whether to meet evaporation conditions, enter step (2.8) if meeting, otherwise enter step (2.9)；
(2.8) according to the whether close enough ocean in river and streams, rainfall is carried out in different ways, is formed new
Precipitation；
(2.9) minimum when previous iteration is updated；
(2.10) judge whether to reach maximum number of iterations, if it is, terminating iteration, export PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE
Optimal result, otherwise, return step (2.6), until iteration terminates.
Further, in the step (2.1), determining cost function calculation formula are as follows:
In formula: N is the number for acquiring data；yiFor the ith sample value of real system output；For identified parameters model
The ith sample value of output.
Further, in the step (2.6), the position in river is flowed to streams respectively, streams flows to the position of ocean
And the position of river direction ocean is updated, and is carried out according to the cost function value of streams each after update, river, ocean
The method of position exchange, comprising the following steps:
The position that (2.6.1) flows to the position in river to streams respectively and streams flows to ocean is updated；
In formula: rand is equally distributed random number between 0 to 1；I-th is respectively indicated to change
Generation during, streams, river and ocean current location；Indicate the new position in streams；C is the coefficient of location updating；
The new position in streams is input to Synchronous Machine Models as model parameter by (2.6.2), calculates the synchronous motor mould
Signal when three-phase suddenly-applied short circuit occurs for type, and the cost function value in each streams is calculated according to step (2.1)；If the cost in streams
Functional value is less than the cost function value in river, then the position in river and streams is exchanged；If the small Yu Haiyang of the cost function value in streams
Cost function value, then the position in ocean and streams exchange；
(2.6.3) is updated the position of river direction ocean, calculation formula are as follows:
In formula: rand is equally distributed random number between 0 to 1；Respectively indicate i-th iteration process
In, the current location in river and ocean；Indicate the new position in river；C is the coefficient of location updating；
The new position in river is input to Synchronous Machine Models as model parameter by (2.6.4), calculates the synchronous motor mould
Signal when three-phase suddenly-applied short circuit occurs for type, and calculates the cost function value in each river；If the small Yu Haiyang of the cost function in river
Cost function value, then the position in ocean and river exchange.
Further, in the step (2.7), evaporation conditions are as follows:
In formula,WithRespectively during i-th iteration, the position of ocean and river；I=1,2 ..., Nsr-1。
Further, in the step (2.8), according to the whether close enough ocean in river and streams, using different sides
Formula carries out rainfall, the method for forming new precipitation are as follows:
IfOr rand < 0.1, i=1,2 ..., Nsr- 1, then rainfall is carried out using following formula
Journey forms new precipitation:
In formula:For the latest position for newly forming streams, UB and LB are respectively the upper and lower boundary of variable, rand 0
Equally distributed random number between to 1；
IfRainfall is carried out using following formula, forms new drop
Water:
In formula: randn is the random number of normal distribution；μ indicates the coefficient of region of search range near ocean.
Further, in the step (2.9), minimumMore new formula are as follows:
In formula: T is water round-robin algorithm maximum number of iterations,Minimum during i-th iteration,For i+1
Minimum in secondary iterative process.
The invention adopts the above technical scheme, which has the following advantages: 1, the present invention according to parameter to be identified not
Together, it is recognized using parameter of the different methods to synchronous motor system, three-phase suddenly-applied short circuit is occurred according to synchronous motor
Actual short electric current calculates the damping time constant of DC component, and calculated result is more accurate.2, the present invention is calculated using water circulation
Method, the Synchronous Machine Models based on foundation recognize the other parameters in addition to damping time constant of synchronous motor, have
Good convergence and higher parameter identification precision, compared with using the parameter identification result of genetic algorithm, this method has
The advantage that parameter setting is few, convergence is good and Identification Errors are small.Therefore, the present invention can be widely applied to generator parameter identification
Field.
Detailed description of the invention
Fig. 1 is short circuit current waveform figure of the present invention；
Fig. 2 is A phase short circuit current waveform figure in the embodiment of the present invention；
Fig. 3 is that current waveform figure is recognized in the embodiment of the present invention；
Fig. 4 is the comparison figure that current waveform and measured waveform are recognized in the embodiment of the present invention.
Specific embodiment
The present invention is described in detail below with reference to the accompanying drawings and embodiments.
As shown in Figure 1, a kind of generator parameter identification method proposed by the present invention, comprising the following steps:
(1) according to the synchronous motor system to be recognized, Synchronous Machine Models are established, signal to be identified is obtained and calculates and be somebody's turn to do
The damping time constant T of the DC component of signal to be identifieda。
Specifically, the following steps are included:
(1.1) synchronous motor system to be recognized is analyzed, establishes the Synchronous Machine Models to tally with the actual situation, the synchronization
It include the identification range for the whole parameters and each parameter to be identified to be recognized in motor model, wherein each parameter to be identified
Identification range provide parameter or empirical value according to producer and determine.
(1.2) short circuit current signal when three-phase suddenly-applied short circuit occurs for the synchronous motor system to be recognized is obtained, as
Signal to be identified.
(1.3) according to the short circuit current signal of acquisition, the die-away time for calculating the DC component of the parameter to be identified is normal
Number Ta。
The method for calculating the damping time constant of DC component are as follows: extract the coenvelope line and lower envelope of short circuit current first
Line calculates the half of the sum of envelope, obtains DC component i0；Secondly to DC component i0Absolute value take logarithm obtain ln (|-
i0|), it then carries out curve fitting, can be obtained the damping time constant T of DC componenta。
(2) water round-robin algorithm, the Synchronous Machine Models based on foundation, to other ginsengs in the signal to be identified of acquisition are used
Number is recognized, and the optimal result of PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE is obtained.
Specifically, the following steps are included:
(2.1) cost function calculation formula is determined, as PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE side of the evaluation based on water round-robin algorithm
The precision of method；Wherein, the calculation formula of cost function value are as follows:
In formula: N is the number for acquiring data；yiFor the ith sample value of real system output；For identified parameters model
The ith sample value of output.
(2.2) water round-robin algorithm control parameter, including rainfall layer number N are setpop, river and ocean sum Nsr, minimum
Initial value dmax, the maximum water round-robin algorithm generation number T, Optimal Parameters number N of changingvar=7, and parameter or experience are provided according to producer
Value is each parameter (x " to be identifiedd、x″q、x′d、T″d、T′d、φ0、xd) the suitable identification range of setting.
(2.3) according to the signal to be identified obtained in step (1), initial population is generated at random, forms initial streams (rain
Drop), river and ocean.
Nsr=NRiver+1 (3)
NSmall stream=Npop-Nsr (4)
In formula: NvarIt is a N for the dimension (numbers of Optimal Parameters) of search spacevarTie up optimization problem；NRiverFor river
Flow number；NsrFor river and ocean sum；NSmall streamFor streams number.
(2.4) raindrop population dividing is multiple raindrop layers as model parameter and is separately input to Synchronous Machine Models, counted
It calculates Synchronous Machine Models and signal when three-phase suddenly-applied short circuit occurs, and calculate the cost function value J of each raindrop layeri.Wherein, each rain
Drop layer expression formula be
(2.5) size of more each raindrop layer cost function selects the smallest cost function value JiCorresponding raindrop layer
As ocean, N is selectedRiverA small cost function value JiCorresponding raindrop layer determines that flow direction refers to as river, and by formula (5)
Determine the streams number of river and ocean.
In formula: NSnFor the streams number for flowing to specific river or ocean, round { f } take f to round up after integer value,
N=1,2 ..., Nsr。
(2.6) position in river is flowed to streams respectively, streams flows to the position of ocean and the position of river direction ocean
It sets and is updated, and position exchange is carried out according to the cost function value of streams each after update, river, ocean；
Specifically, the following steps are included:
The position that (2.6.1) flows to the position in river to streams respectively and streams flows to ocean is updated.
In formula: rand is equally distributed random number between 0 to 1；I-th is respectively indicated to change
Generation during, streams, river and ocean current location；C is the coefficient of location updating, takes empirical value 2.
The new position in streams is separately input to Synchronous Machine Models as model parameter by (2.6.2), is calculated this and is synchronized electricity
Signal when three-phase suddenly-applied short circuit occurs for machine model, and the cost function value in each streams is calculated according to step (2.1).If streams
Cost function value is less than the cost function value in river, then the position in river and streams is exchanged；If the cost function value in streams is less than
The cost function value of ocean, then the position in ocean and streams is exchanged.
(2.6.3) is updated the position of river direction ocean.
In formula: rand is equally distributed random number between 0 to 1；Respectively indicate i-th iteration process
In, the current location in river and ocean；C is the coefficient of location updating, takes empirical value 2.
The new position in river is input to Synchronous Machine Models as model parameter by (2.6.4), calculates the synchronous motor mould
Signal when three-phase suddenly-applied short circuit occurs for type, and the cost function value in each river is calculated according to step (2.1).If the cost in river
The cost function value of the small Yu Haiyang of function, then the position in ocean and river is exchanged.
(2.7) judge whether to meet evaporation conditions, enter step (2.8) if meeting, otherwise enter step (2.9).
Wherein, evaporation conditions are as follows:
In formula,WithRespectively during i-th iteration, the position of ocean and river；I=1,2 ..., Nsr-
1；dmaxFor a very small constant.
(2.8) according to the whether close enough ocean in river and streams, rainfall is carried out in different ways, is formed new
Precipitation.
IfOr rand < 0.1, i=1,2 ..., Nsr- 1, then rainfall is carried out using formula (9)
Journey forms new precipitation.
In formula:For the latest position for newly forming streams；UB and LB is respectively the upper and lower boundary of variable；Rand is fixed
Justice is same as above.
IfRainfall is carried out using formula (10), forms new drop
Water.
In formula: randn is the random number of normal distribution；μ indicates the coefficient of region of search range near ocean, and μ is smaller, then
Search range is closer far from ocean (optimal solution), and general μ takes 0.1.
(2.9) minimum when previous iteration is updated, more new formula is following formula:
In formula: T is water round-robin algorithm maximum number of iterations,Minimum during i-th iteration,For i+1
Minimum in secondary iterative process.
(2.10) judge whether to reach maximum number of iterations, if it is, terminating iteration, export PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE
Optimal result, otherwise, return step (2.6), until iteration terminates.
Embodiment one
In the present embodiment, after Sudden Three-phase Short Circuit occurs for synchronous motor under no-load condition, the stator current in A phase can be indicated
Are as follows:
In formula: TaFor the damping time constant of DC component, T 'dFor longitudinal axis transient time constant, T "dSurpass transient state for the longitudinal axis
Time constant, xdFor synchronous reactance, x 'dFor d-axis transient reactance, x "dFor longitudinal axis ultra-transient reactance, x "qSurpass transient state electricity for horizontal axis
It is anti-, φ0For short-circuit initial phase angle, E is generator no-load emf, and e (t) is noise current.
Simulation analysis: T is carried out first by taking ideal synchronous machine as an example for the validity of verification methoda=0.4348s, T 'd=
0.8250s, T "d=0.0170s, xd=2.1690 (per unit values, similarly hereinafter), x 'd=0.2290, x "d=0.1830, x "q=
0.1855, short-circuit initial phase angle φ0=π/3, E=1.Current waveform after motor generation Sudden Three-phase Short Circuit is as shown in Figure 1.
The damping time constant T of the DC component of the electric current is calculated firsta.The coenvelope line of short circuit current is extracted under
Envelope calculates the half of the sum of envelope, obtains DC component i0, to i0Absolute value take logarithm obtain ln (|-i0|), so
After carry out curve fitting, can be obtained the damping time constant T of DC componenta.T is obtained by calculationa=0.4348s.
Then optimizing, parameter setting are carried out to other parameters using water round-robin algorithm are as follows: maximum number of iterations T=500,
Rainfall layer number Npop=80, river and ocean sum Nsr=4, minimum dmax=10-5, Optimal Parameters number Nvar=7, and root
Parameter is provided according to producer or empirical value is each parameter (x " to be identifiedd、x″q、x″d、T″d、T′d、φ0、xd) set and suitably distinguish
Know range.Its parameter identification result are as follows: x "d=0.1830, x "q=0.1855, x 'd=0.2290, T "d=0.0170s, T 'd=
0.8255s, φ0=0.3333 π, xd=2.1760.As can be seen that method proposed by the present invention has very high identification precision.
In order to further verify the performance of the generator parameter identification method proposed by the present invention based on water round-robin algorithm, adopt
Parameter identification, parameter setting are carried out to it with common genetic algorithm are as follows: population scale=80, number of parameters=7, maximum are lost
Passage number=500, crossover probability=0.9.The parameter identification result of this method are as follows: x "d=0.1844, x "q=0.1868, x 'd=
0.2340, T "d=0.0233s, T 'd=0.9281s, φ0=0.3326 π, xd=3.8427.As can be seen that taking identical kind
When group's number and the number of iterations, genetic algorithm obtains identification result and true value is widely different, and identification precision is poor.
Embodiment two
By taking some hydropower station generator as an example, having for the generator parameter identification method based on water round-robin algorithm of proposition is verified
Effect property.Rated voltage 11kV, rated current 656.1A, rated power factor 0.85 (lag), rated speed
428.6r/min.Under 0.20p.u. voltage, three-phase sudden short circuit test, the motor A phase short circuit electricity of actual measurement are implemented to generator
It is as shown in Figure 2 to flow waveform.
The damping time constant T of the DC component of the electric current is calculated firsta.The coenvelope line of short circuit current is extracted under
Envelope calculates the half of the sum of envelope, obtains DC component i0, to i0Absolute value take logarithm obtain ln (|-i0|), so
After carry out curve fitting, can be obtained the damping time constant T of DC componenta.T is obtained by calculationa=0.1853s.
Then optimizing, parameter setting are carried out to other parameters using water round-robin algorithm are as follows: maximum number of iterations T=500,
Rainfall layer number Npop=80, river and ocean sum Nsr=4, minimum dmax=10-5, Optimal Parameters number Nvar=7, and root
Parameter is provided according to producer or empirical value is each parameter (x " to be identifiedd、x″q、x′d、T″d、T′d、φ0、xd) set and suitably distinguish
Know range.Its parameter identification result are as follows: x "d=0.1890, x "q=0.1537, x 'd=0.2561, T "d=0.0635s, T 'd=
0.8073s, φ0=1.8170 π, xd=0.9683.Short circuit current waveform is calculated with these parameters, as shown in Figure 3.Identification electricity
The comparison figure for flowing waveform and measured waveform is as shown in Figure 4.
From Fig. 2~4 as can be seen that identification current waveform and actual experimental waveform have it is extraordinary coincide, illustrate to use
The parameter that present invention identification obtains is more accurately, to demonstrate the validity of the discrimination method.
The various embodiments described above are merely to illustrate the present invention, wherein the structure of each component, connection type and manufacture craft etc. are all
It can be varied, all equivalents and improvement carried out based on the technical solution of the present invention should not exclude
Except protection scope of the present invention.
Claims (9)
1. a kind of generator parameter identification method, it is characterised in that the following steps are included:
(1) according to the synchronous motor system to be recognized, Synchronous Machine Models are established, obtain signal to be identified and are calculated wait distinguish
Know the damping time constant T of the DC component of signala；
(2) water round-robin algorithm is used, the Synchronous Machine Models based on foundation carry out the other parameters of the signal to be identified of acquisition
Identification, obtains the optimal result of PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE；The other parameters include longitudinal axis ultra-transient reactance x "d, horizontal axis it is super
Transient reactance x "q, d-axis transient reactance x 'd, the longitudinal axis surpass transient time constant T "d, longitudinal axis transient time constant T 'd, short-circuit first phase
Angle φ0, synchronous reactance xd。
2. a kind of generator parameter identification method as described in claim 1, it is characterised in that: in the step (1), according to institute
The synchronous motor system to be recognized, establishes Synchronous Machine Models, obtains signal to be identified and calculates the direct current point of signal to be identified
The method of the damping time constant of amount, comprising the following steps:
(1.1) synchronous motor system to be recognized is analyzed, the Synchronous Machine Models to tally with the actual situation are established, it is described to synchronize electricity
It include the identification range of whole parameters and each parameter to be identified to be identified in machine model；
(1.2) the short circuit current signal when acquisition synchronous motor system to be recognized generation three-phase suddenly-applied short circuit is as to be identified
Signal；
(1.3) according to the signal to be identified of acquisition, the damping time constant T of its DC component is calculateda。
3. a kind of generator parameter identification method as claimed in claim 2, it is characterised in that: in the step (1.3), calculate
The damping time constant T of the DC component of parameter to be identifiedaMethod are as follows:
The coenvelope line and lower envelope line for extracting short circuit current first, calculate the half of the sum of envelope, obtain DC component i0；
Secondly to DC component i0Absolute value take logarithm obtain ln (|-i0|)；
Then it carries out curve fitting, obtains the damping time constant T of DC componenta。
4. a kind of generator parameter identification method as described in claim 1, it is characterised in that: in the step (2), using water
Round-robin algorithm, the Synchronous Machine Models based on foundation recognize the other parameters in the signal to be identified of acquisition, obtain same
The method for walking the optimal result of parameter of electric machine identification, comprising the following steps:
(2.1) cost function calculation formula is determined；
(2.2) water round-robin algorithm control parameter, including rainfall layer number N are setpop, river and ocean sum Nsr, minimum is initial
Value dmax, water round-robin algorithm maximum number of iterations T, Optimal Parameters number Nvar, and be every according to producer's offer parameter or empirical value
A parameter x " to be identifiedd、x″q、x′d、T″d、T′d、φ0、xdSetting identification range；
(2.3) initial population is generated at random, forms initial streams, river and ocean；
(2.4) streams population dividing is multiple streams layers as model parameter and is separately input to Synchronous Machine Models, calculated same
It walks motor model and signal when three-phase suddenly-applied short circuit occurs, and calculate the cost function value J of each streams layeri；
(2.5) size of more each streams layer cost function, selects the smallest cost function value JiCorresponding streams layer is as sea
Ocean selects NRiverA small cost function value JiCorresponding streams layer is used as river, and determines and flow to specified river and ocean
Streams number；
(2.6) position in river is flowed to streams respectively, streams flow to ocean position and river direction ocean position into
Row updates, and carries out position exchange according to the cost function value of streams each after update, river, ocean；
(2.7) judge whether to meet evaporation conditions, enter step (2.8) if meeting, otherwise enter step (2.9)；
(2.8) according to the whether close enough ocean in river and streams, rainfall is carried out in different ways, forms new drop
Water；
(2.9) minimum when previous iteration is updated；
(2.10) judge whether to reach maximum number of iterations, if it is, terminating iteration, export PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE most
It is excellent as a result, otherwise, return step (2.6), until iteration terminates.
5. a kind of generator parameter identification method as claimed in claim 4, it is characterised in that: in the step (2.1), determine
Cost function calculation formula are as follows:
In formula: N is the number for acquiring data；yiFor the ith sample value of real system output；For the output of identified parameters model
Ith sample value.
6. a kind of generator parameter identification method as claimed in claim 4, it is characterised in that: in the step (2.6), respectively
The position in river is flowed to streams, the position for the position and river direction ocean that streams flows to ocean is updated, and according to
The method of the cost function value progress position exchange in each streams, river, ocean after update, comprising the following steps:
The position that (2.6.1) flows to the position in river to streams respectively and streams flows to ocean is updated；
In formula: rand is equally distributed random number between 0 to 1；Respectively indicate i-th iteration mistake
Cheng Zhong, streams, river and ocean current location；Indicate the new position in streams；C is the coefficient of location updating；
The new position in streams is input to Synchronous Machine Models as model parameter by (2.6.2), calculates Synchronous Machine Models hair
Signal when raw three-phase suddenly-applied short circuit, and according to the cost function value in each streams of step (2.1) calculating；If the cost function in streams
Value is less than the cost function value in river, then the position in river and streams is exchanged；If the generation of the small Yu Haiyang of the cost function value in streams
Valence functional value, then the position in ocean and streams is exchanged；
(2.6.3) is updated the position of river direction ocean, calculation formula are as follows:
In formula: rand is equally distributed random number between 0 to 1；During respectively indicating i-th iteration, river
The current location of stream and ocean；Indicate the new position in river；C is the coefficient of location updating；
The new position in river is input to Synchronous Machine Models as model parameter by (2.6.4), calculates Synchronous Machine Models hair
Signal when raw three-phase suddenly-applied short circuit, and calculate the cost function value in each river；If the generation of the small Yu Haiyang of the cost function in river
Valence functional value, then the position in ocean and river is exchanged.
7. a kind of generator parameter identification method as claimed in claim 4, it is characterised in that: in the step (2.7), evaporation
Condition are as follows:
In formula,WithRespectively during i-th iteration, the position of ocean and river；I=1,2 ..., Nsr-1。
8. a kind of generator parameter identification method as claimed in claim 4, it is characterised in that: in the step (2.8), according to
The whether close enough ocean in river and streams, carries out rainfall, the method for forming new precipitation in different ways are as follows:
IfOr rand < 0.1, i=1,2 ..., Nsr- 1, then rainfall, shape are carried out using following formula
The precipitation of Cheng Xin:
In formula:For the latest position for newly forming streams, UB and LB are respectively the upper and lower boundary of variable, rand be 0 to 1 it
Between equally distributed random number；
IfI=1,2 ..., NS1, rainfall is carried out using following formula, forms new precipitation:
In formula: randn is the random number of normal distribution；μ indicates the coefficient of region of search range near ocean.
9. a kind of generator parameter identification method as claimed in claim 4, it is characterised in that: minimum in the step (2.9)
ValueMore new formula are as follows:
In formula: T is water round-robin algorithm maximum number of iterations,Minimum during i-th iteration,Repeatedly for i+1 time
Minimum during generation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910266221.6A CN109981011B (en) | 2019-04-03 | 2019-04-03 | Generator parameter identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910266221.6A CN109981011B (en) | 2019-04-03 | 2019-04-03 | Generator parameter identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109981011A true CN109981011A (en) | 2019-07-05 |
| CN109981011B CN109981011B (en) | 2021-03-09 |
Family
ID=67082561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910266221.6A Active CN109981011B (en) | 2019-04-03 | 2019-04-03 | Generator parameter identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109981011B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04165999A (en) * | 1990-10-26 | 1992-06-11 | Hitachi Ltd | Exciter for synchronous machine |
| CN103529698A (en) * | 2013-10-17 | 2014-01-22 | 广东电网公司电力科学研究院 | Method for distinguishing parameter of power generator speed regulating system |
| CN106169747A (en) * | 2016-07-20 | 2016-11-30 | 河海大学 | A kind of double fed induction generators parameter identification method |
| CN106788092A (en) * | 2017-02-28 | 2017-05-31 | 南京工程学院 | A kind of PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE method based on atom decomposition |
-
2019
- 2019-04-03 CN CN201910266221.6A patent/CN109981011B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04165999A (en) * | 1990-10-26 | 1992-06-11 | Hitachi Ltd | Exciter for synchronous machine |
| CN103529698A (en) * | 2013-10-17 | 2014-01-22 | 广东电网公司电力科学研究院 | Method for distinguishing parameter of power generator speed regulating system |
| CN106169747A (en) * | 2016-07-20 | 2016-11-30 | 河海大学 | A kind of double fed induction generators parameter identification method |
| CN106788092A (en) * | 2017-02-28 | 2017-05-31 | 南京工程学院 | A kind of PARAMETER IDENTIFICATION OF SYNCHRONOUS MACHINE method based on atom decomposition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109981011B (en) | 2021-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106058937B (en) | A kind of power distribution network broad sense load modeling method of the wind power plant containing direct-drive permanent-magnetism | |
| CN110032812A (en) | A kind of dynamic state estimator method based on adaptive volume Kalman filtering | |
| CN111293686A (en) | ARMAX system identification-based real-time evaluation method for inertia of power system | |
| CN104392056B (en) | A kind of directly driven wind-powered unit modeling method and device | |
| CN106383947B (en) | The fast acquiring method of wind power plant current collection network dynamic equivalent parameters | |
| CN105741188B (en) | Doubly-fed wind turbine group's electro-magnetic transient coherence grouping method | |
| CN110350535A (en) | A kind of meter and large-scale wind power and the grid-connected distribution network voltage probability of stability appraisal procedure of electric car | |
| CN108549962A (en) | Wind power forecasting method based on history fragment sequence search and sequential rarefaction | |
| CN109921426A (en) | Wind-electricity integration system probability load flow calculation method based on CV-KDE | |
| CN105576654B (en) | Directly driven wind-powered field equivalence method and system | |
| CN108090244A (en) | A kind of parallel type lithium ion battery system modeling method | |
| CN110222309A (en) | A kind of generator method for dynamic estimation based on robust volume Kalman filtering | |
| CN109412203A (en) | A kind of more feed-in electric system grid connection capacity Limit Calculation Methods of wind-powered electricity generation | |
| CN106443496A (en) | Battery charge state estimation method with improved noise estimator | |
| CN110674864A (en) | Wind power abnormal data identification method with synchronous phasor measurement device | |
| CN115392133A (en) | Wind power plant optimal clustering equivalence method adopting Gaussian mixture model | |
| CN105701265B (en) | A kind of double-fed fan motor unit modeling method and device | |
| CN106410862A (en) | Wind power plant single machine equivalent method based on active recovery slope correction | |
| CN105303319A (en) | Wind power plant dynamic clustering equivalence method based on operating data | |
| CN109981011A (en) | A kind of generator parameter identification method | |
| CN110571794B (en) | Transient model equivalent calculation method suitable for doubly-fed wind power plant | |
| Wang et al. | Evaluation method of wind turbine group classification based on calinski harabasz | |
| CN111293693A (en) | Doubly-fed wind turbine converter control parameter identification method based on extended Kalman filtering | |
| CN109638892A (en) | A kind of photovoltaic plant equivalent modeling method based on improvement fuzzy clustering algorithm | |
| CN109256773A (en) | Power system state estimation method of the noise reduction from coding and depth support vector machines |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210916 Address after: 20 Chegongzhuang West Road, Haidian District, Beijing 100038 Patentee after: China Institute of Water Resources and Hydropower Research Patentee after: Tianjin Shuike electromechanical Co.,Ltd. Address before: 20 Chegongzhuang West Road, Haidian District, Beijing 100038 Patentee before: China Institute of Water Resources and Hydropower Research |
|
| TR01 | Transfer of patent right |