CN109066688A - Based on the Probabilistic Load Flow data capture method under renewable energy uncertainty - Google Patents
Based on the Probabilistic Load Flow data capture method under renewable energy uncertainty Download PDFInfo
- Publication number
- CN109066688A CN109066688A CN201811034650.2A CN201811034650A CN109066688A CN 109066688 A CN109066688 A CN 109066688A CN 201811034650 A CN201811034650 A CN 201811034650A CN 109066688 A CN109066688 A CN 109066688A
- Authority
- CN
- China
- Prior art keywords
- load flow
- probability
- variable
- power
- probabilistic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
Present invention discloses the Probabilistic Load Flow data capture method under a kind of uncertainty based on renewable energy, S1 inputs the data file of power grid electric parameter, configures power grid random parameter；S2 carries out AC network Load flow calculation and probabilistic load flow, and output MCS samples scale, bus independent variable parameter, route independent variable parameter；S3 configures grid equipment section, including bus voltage amplitude section, line load rate section, transformer load rate section；S4 obtains electrical equipment utilization rate assessment result using Density Estimator；The present invention can fully consider many uncertain factors of power grid, Probabilistic Load Flow method based on Monte-Carlo Simulation, establish the calculation procedure of estimation electric power networks trend, and it is applied to practical transmission system, electric power networks capacity level and load factor etc. are analyzed, the planning and assessment of power grid are preferably instructed.
Description
Technical field
Data, which are obtained, the present invention relates to the Probabilistic Load Flow under the renewable energy uncertainty based on computer technology takes method.
Background technique
Electric energy has extremely important status as conveniently secondary energy sources in social development and human lives, and
Electric energy mainly passes through what the non-renewable energy such as coal, petroleum converted, with economic continuous development, also gradually to energy demand
Increase, traditional fossil fuel based on coal since its non-renewable feature is faced with the predicament petered out, and
Persistently using fossil fuel, oneself warp generates serious pollution to environment.The shortage of traditional non-renewable energy resources and environment it is serious
Polluting oneself becomes the serious hindrance that electric power development is faced.Therefore, find clean and effective renewable energy replace it is traditional not
Renewable energy become current electric power development there is an urgent need to.Based on the urgent need of the renewable energy to clean and effective, with
Wind-power electricity generation, the renewable energy power generation that photovoltaic power generation is representative gradually attract attention, the major advantage of renewable energy power generation
Be it is pollution-free, renewable, exactly these advantages make the research to renewable energy gradually deeply.
With the continuous maturation of renewable energy power generation technology, using wind-power electricity generation and photovoltaic power generation as the renewable energy of representative
Specific gravity of the source power generation shared by power grid just gradually increases.However renewable energy power generation also has very big drawback, output work
Rate be it is uncontrollable, randomness is very strong, such as: the output power of wind-power electricity generation be it is related with the size of wind speed, and wind speed by
The influence of weather be it is uncertain, there is very strong randomness, thus it is very strong to cause the output power of wind-power electricity generation also to have
Randomness；The output power of photovoltaic power generation and the power of illumination are related, and illumination equally by weather influenced be it is uncertain,
With very strong randomness, thus cause the output power of photovoltaic power generation that also there is very strong randomness.
Output power just because of renewable energy power generation has very strong randomness, causes to renewable energy power generation
The prediction of output power is no longer determining value, but describes renewable energy power generation with the stochastic variable of form of probability
Output power electric power system tide calculates the content most basic as Power System Analysis, can be the static security of electric system
Analysis and operation of power networks state real-time analysis offer basic data, can be electric system reactive compensation and power grid it is optimal
Load flow calculation providing method can provide means for the planning and designing of power grid and the risk assessment of power grid.Traditional certainty tide
Flow calculation methodologies can be divided into Newton-Laphson method, algorithm quicksort and DC power flow algorithm etc., the output based on generator it is active and
The network topology parameters of reactive power, the active and reactive power of load and electric system are known and are under the premise of determining value
The amplitude and phase angle (or real and imaginary parts of busbar voltage), branch of busbar voltage can be obtained by certainty Load flow calculation
Active and reactive power and the output quantities such as line loss qualitative results really.It, can be true according to output quantity qualitative results really
Determine whether electric system is safely operated, it can be estimated that whether the method for operation of electric system is reasonable, it can be determined that electric system branch
Road trend whether be more than analysis needed for thermostabilization and the dynamic stability limit etc. electric state amount.
The biggish electric system of the renewable energy power generations proportion such as wind-power electricity generation, photovoltaic power generation is safely operated
When analysis, since the renewable energy power generations such as wind-power electricity generation, photovoltaic power generation output power is no longer determining value, but with probability point
The stochastic variable of cloth description, traditional certainty Load flow calculation cannot handle the case where input quantity is stochastic variable, thus tradition
Certainty Load flow calculation will be no longer appropriate for this biggish safe operation of power system of renewable energy power generation proportion
Analysis.In order to solve this problem, Probabilistic Load Flow comes into being, and probabilistic load flow can count and various uncertain factors,
The random fluctuation generated, the start and stop of generating set, load are influenced by electricity price by weather including renewable energy power generation output power
Generated random fluctuation, the random fault of network topology and variation of power system operation mode etc..These are uncertain
Factor leads to the injecting power of the node of system and the topological parameter of electric system becomes stochastic variable, and Probabilistic Load Flow is according to random
The probability characteristics of variable, such as expectation of a random variable and variance, probability density function and cumulative distribution function etc., obtain
The amplitude and phase angle (or real and imaginary parts of node voltage) of node voltage, the active and reactive power of branch and route damage
The probability characteristics of these output stochastic variables such as consumption.Probabilistic load flow has expanded the application model of traditional certainty Load flow calculation
It encloses, more specific and comprehensive information can be provided for the planning and designing of electric system and Safety Operation Analysis, comprising: node voltage
Amplitude and the desired value of phase angle, variance, probability density function and cumulative distribution function, node voltage amplitude and phase angle it is out-of-limit general
Rate, branch active and idle desired value, variance, probability density function and cumulative distribution function, branch is active to be got over idle
Probability etc. is limited, these information are to the reasonability of planning and design of power system, the optimization and power grid of grid operation mode arrangement
Weak link and the assessment of potential risk etc. are extremely beneficial.
Renewable energy technologies serve not only as the supplement and standby of bulk power grid concentration supply power, and for making full use of point
Scattered renewable energy reduces disposal of pollutants, and economic sustainable development is promoted to all have significance.But for application
For widest wind energy and solar energy renewable energy technologies, mostly to access based on power distribution network, can make after grid-connected power distribution network by
One passive network is transformed into active electric network, this can largely influence the size and Orientation of network trend and electrical member
The voltage of part and user terminal；Secondly as the dynamic wave of the external environment conditions such as wind farm wind velocity or photovoltaic plant intensity of illumination
Dynamic variation, so that there are larger random fluctuations for the output power of grid-connected rear renewable energy, and intermittent power swing will be right
The power quality of bulk power grid adversely affects；In addition, renewable energy generally use power electronic equipment implement it is grid-connected, and it is electric
The negative sequence voltage that voltage, the current harmonics of power electronic device generation are also inevitable or even unbalanced grid faults generate
And the voltage harmonic of power grid itself and renewable energy current transformer interact, and will lead to current transformer and generate additional harmonic electricity
Stream.When bulk power grid has enough spare appearances and regulating power, need not generally consider caused by renewable energy power swing
Frequency departure, and mainly consider voltage fluctuation and flicker caused by power swing.Therefore, after renewable energy access electric system
Influence for power quality is concentrated mainly on the side such as voltage deviation and voltage fluctuation, three-phase imbalance and harmonic current influence
Face.
Due to the randomness of renewable energy, for the safe and stable operation for guaranteeing system, when the renewable energy of access system
When source is larger, should to after its access system grid stability and adaptability analyze.Since renewable energy is electrical
Control model and rotary inertia etc. are different from conventional power unit, will change power grid to a certain extent after extensive renewable energy access
Transient stability.It is whether influential on electrical power system transient power-angle stability after grid-connected for extensive renewable energy to ask
It is existing that topic depends on grid operation mode, topological structure of electric and the Wind turbines technology taken, the access of renewable energy
The transient stability that system may be deteriorated is also possible to the transient stability of improvement system, this must be in conjunction with the actual motion of power grid
Feature can determine to carry out simulation calculation.Therefore, it is connect for the extensive renewable energy of power grid practical operation situation research
Transient stability problem after entering is highly desirable, and can be run for dispatching of power netwoks and be provided certain practical guidance foundation.
Currently, Probabilistic Load Flow research conditions are as follows both at home and abroad:
Probabilistic Load Flow is to be proposed by Borkowska in 1974 first.Later, research day of the scholars to Probabilistic Load Flow
Benefit is deeply.Two aspects of computation model and calculation method are concentrated mainly on to the research of Probabilistic Load Flow.The calculating mould of Probabilistic Load Flow
Type is broadly divided into DC Model, linearisation AC model, piece-wise linearization AC model and retains nonlinear exchange mould
Four class of type.Such as Borkowska, in the DC Model of the simplification of proposition in 1974, the linearisation that Allan was proposed in 1976 is handed over
Flow model, Allan in order to further increase computational accuracy 1981 propose piece-wise linearization AC model,
The precision that Sokierajski was proposed in 1978 is higher to retain nonlinear exchange mould etc., these probabilistic load flow models
As main computation model in current probability load flow calculation method, wherein the widest to linearize AC Ioad flow model application
It is general.The research of probability load flow calculation method is that the emphasis in probabilistic load flow research occurs by the effort of domestic and foreign scholars
A large amount of probability load flow calculation method, these probability load flow calculation methods are broadly divided into simulation, approximation method and parsing
Method three classes can be carried out according to following several aspects in order to assess the superiority and inferiority of probability load flow calculation method performance: 1) can be obtained
The probability characteristics of output variable；2) computational accuracy wants sufficiently high；3) calculating the time cannot be too long；4) it is capable of handling stochastic variable
Correlation；5) robustness will be met the requirements.
1, simulation
Simulation probabilistic load flow is sampled input variable according to the probability-distribution function of input variable (takes out first
Quadrat method includes arbitrary sampling method and the improvement methods of sampling etc.), the sample matrix of the input variable met the requirements is established, then
Corresponding electric power system tide computation model is combined to repeat certainty trend according to the input variable sample matrix established
It calculates, finally output variable is fitted according to the discrete results of the resulting output variable of certainty Load flow calculation either general
Rate analysis, to obtain the probability-distribution function or probability characteristics of output variable.In simulation probabilistic load flow, according to defeated
It is input variable that is the most key, being built that the probability-distribution function for entering variable, which establishes the input variable sample matrix met the requirements,
The sample size and sample availability of sample matrix directly affect the precision of output variable result, if sample size is not big enough and sample
This validity not from when, the result precision of output variable will be very low, conversely, if sample size is sufficiently large and validity is also very high
When, the result precision of output variable also can be very high.In the case where the standard deviation of sample remains unchanged, selection increases number of samples
It is the mode that simulation more commonly improves sampling precision, however which will increase the number of certainty Load flow calculation, lead
It causes to calculate overlong time.Currently, simulation include Monte Carlo simulation approach based on random sampling, Latin hypercube,
Quasi Monte Carlo method and importance sampling technique.
Monte Carlo method is consistent with the purpose of Latin hypercube, also for illiteracy of the improvement based on random sampling
Special Monte Carlo Simulation of Ions Inside method sample drawn excessively causes the shortcomings that calculating overlong time, covers the methods of sampling by effective space come generation
For method of random sampling, quasi-Monte Carlo simulation is sampled stochastic variable using low diversity sequence.Low diversity sequence and puppet
Random number series are same concepts, and low diversity sequence is the number in the section that series of values determines, it is assumed that in certain dimension variable
Oneself has n-1 numerical value to low diversity sequence, by maximum blank space in n-1 numerical value of insertion to obtain the n-th of low diversity sequence
A numerical value, the purpose for the arrangement is that in order to avoid the numerical value of low diversity sequence is gathered on local space, to guarantee limited number
It is worth all standing in space.
2, approximation method
Another method of the approximation method as probabilistic load flow, as the term suggests its principle is the probability based on input variable
Feature approximate calculation goes out the probability characteristics of output variable, and this method, which does not need to extract great amount of samples as simulation, to be repeated
Certainty Load flow calculation, so the calculating speed of approximation method is faster than simulation speed, another advantage of approximation method is closely
It can also be by the correlation between input variable in view of in probabilistic load flow like method.Currently, approximation method includes point estimation
Method ,-secondary second order moments method and state transformation method.
Point estimations probabilistic load flow can be divided into following four step:
(1) according to the probability-distribution function of input variable each in electric system, it is corresponding to calculate each input variable
Preceding 2m-1 rank center away from；
(2) according to the preceding 2m-1 rank center of the resulting each input variable of calculating away from the m for constructing each input variable is a
Discrete state, wherein the information that m discrete state of each input variable is covered includes the preceding 2m-1 rank of each input variable
Center is away from all information covered；
(3) according to m discrete state of each input variable and its expectation, the pass of input variable and output variable is utilized
It is the m discrete state that formula calculates output variable；
(4) desired value and variance of output variable are calculated according to m discrete state of calculated output variable.
If the estimation point of input variable is more, the estimation point of output variable is also more, and computational accuracy is also higher, but calculates
Amount can also increased dramatically, and common point estimations include two o'clock and three point estimations, that is, m=2 or 3.Although point estimations
Calculate that simple, speed is fast, program is relatively easy to realize, but point estimations calculated output variable High Order Moment error compared with
Greatly, it is difficult to obtain the probability-distribution function of output variable, and upper relatively complicated in the processing of the correlation of input variable, this is also
The disadvantage of point estimations.
First-order second moment method is a kind of method of approximation probability simulation, to the non-linear power flow algorithm of electric system into
Row Taylor series expansion takes first order Taylor series to calculate the expectation of input variable as the computation model of first-order second moment method
Value and covariance, obtain the desired value and variance of output variable according to the computation model for the first-order second moment method established.Once
Second order moments method calculating speed is fast, and principle is also relatively simple, and program is also easy to accomplish, and can handle the correlation between input variable
Property, it can be difficult to obtaining the probability-distribution function of output variable.
State transformation can be divided into linear transformation, polynomial transformation and Unscented transform according to the difference of transform method.If electric
Force system input variable Normal Distribution, then output variable can be obtained according to the power flow equation after linearisation, due to trend
Equation is linearized, so output variable may be considered the linear combination of input variable, and output variable is still obeyed
Normal distribution.Unscented transform by Fitted probability be distributed in the way of replace nonlinear transformation calculating process, only need a small amount of sample and
Its corresponding weight can be obtained by input variable probability characteristics, using nonlinear function calculate output variable desired value and
Variance state transformation method calculating speed is fast, and the theory of conversion process is simple, but state transformation method requires input variable necessary
Normal Distribution, which also limits the application scenarios of state transformation method.
3, analytic method
The basis of analytic method is convolutional calculation, obtains output variable by convolutional calculation according to the probability distribution of input variable
Probability distribution.Analytic method can be divided into convolution method and Cumulants method according to the processing mode difference to convolutional calculation.Convolution method
Using traditional convolutional calculation formula, traditional convolutional calculation is carried out under conditions of stochastic variable is mutually indepedent.
It is close according to the probability of input variable if input variable is independent from each other in the probabilistic load flow of electric system
The probability density function that function carries out the available output variable of convolutional calculation is spent, but if there is correlation between input variable
Property, then just needing also to need to handle the input variable with correlation before convolutional calculation, pass through transfer principle
Input variable with correlation is converted to the linear combination of mutually independent random variables, carries out convolutional calculation then to obtain
To the probability density function of output variable.
Convolutional calculation is more complicated, if the model of system is very big, the number of input variable is more, then convolution meter
Calculation will be fairly time consuming, thus in order to reduce the time of traditional convolutional calculation, there is scholar to propose a series of transform method and is used for
Traditional convolutional calculation is improved, such as Laplace transform, Fourier transformation and Fast Fourier Transform (FFT), although utilizing these transformation
Method improves traditional convolutional calculation, improves calculating speed to a certain extent, but computational efficiency is still not
It is high.
Cumulants method is most widely used method in analytic method, this method avoid complicated convolutional calculation, according to
Cumulant characteristic, in each rank cumulant of known input variable, each rank cumulant of available output variable.Half
Invariant is alternatively referred to as cumulant, and the two is same concept.Cumulant probabilistic load flow process is as follows:
(1) each rank square that input variable is calculated according to the probability-distribution function of input variable, can be moment of the orign, can also
Be center away from；
(2) each rank cumulant of input variable is calculated according to each rank square of input variable；
(3) according to the functional relation of input variable and output variable, output is calculated using the computation performance of cumulant
Each rank cumulant of variable；
(4) the general of output variable is calculated using corresponding series expansion principle according to each rank cumulant of output variable
Rate distribution function.
In fact, calculating each rank square of input variable can also ask in the case where the probability-distribution function of input variable is unknown
Out, in the case where oneself knows a large amount of discrete datas of input variable, each rank of input variable can be obtained by approximate method
Square.Cumulants method is actually or based on convolutional calculation principle, just with the computation performance of cumulant instead of complexity
Convolutional calculation, so the premise of Cumulants method probabilistic load flow is also built upon the mutually independent scene of input variable
Under, if there are correlations between input variable, just need before Cumulants method probabilistic load flow to phase
The input variable of closing property is handled, and the input variable with correlation is converted into independent random variable by transfer principle
Linear combination send kind a conversion purpose to be consistent with convolution method, by the input variable of correlation be converted to it is mutually independent with
After the linear combination of machine variable, so that it may carry out probabilistic load flow, premise is also built upon on the basis of Cumulants method.
Summary of the invention
The technical problem to be solved by the present invention is to realize a kind of basis given electric network composition, parameter and generator, negative
The service condition of the elements such as lotus, the method for determining electric system each section steady-state operating condition parameter calculate active power, idle
The distribution of power and voltage in power network.Usually given service condition has each power supply and the power of load point, pivot in system
Knot voltage, the voltage of equalization point and phase angle.Running state parameters to be asked include the voltage magnitude of each bus nodes of power grid
And the power distribution of phase angle and each route, power loss of network etc..
To achieve the goals above, the technical solution adopted by the present invention are as follows:
Based on the Probabilistic Load Flow data capture method under renewable energy uncertainty, it is characterised in that:
S1 inputs the data file of power grid electric parameter, configures power grid random parameter；
S2 carries out AC network Load flow calculation and probabilistic load flow, output MCS sampling scale, bus independent variable ginseng
Number, route independent variable parameter；
S3 configures grid equipment section, including bus voltage amplitude section, line load rate section, transformer load rate area
Between；
S4 obtains electrical equipment utilization rate assessment result using Density Estimator.
If the power grid electric parameter and grid equipment section in S1 and S3 change, the number in real-time update S1 and S3
According to.
In the S2, AC network tidal current computing method is based on one pressgang of newton is inferior or Fa Liewen Burger-Ma Kuaerte
The AC network tidal current computing method of method includes:
1) input stochastic variable probability-distribution function is obtained；
2) input stochastic variable sample matrix is obtained；
3) Load flow calculation is carried out based on sample matrix, and saves result；
4) sample matrix of output variable is obtained.
In the S2, probability load flow calculation method includes: based on Monte Carlo simulation approach
1) Load flow calculation data are read, input variable information is read, determines probabilistic load flow parameter；
2) input variable sample matrix is obtained based on Monte Carlo sampling techniques；
3) Load flow calculation is carried out using sample matrix, and saves output variable information；
4) probability distribution and statistics of analysis node voltage magnitude phase angle and Line Flow.
Probability-distribution function approximating method in the S4 based on Density Estimator includes:
1) probabilistic load flow structure is read；
2) node voltage data or Line Flow data are selected；
3) Density Estimator function is selected；
4) it calculates cuclear density and estimates width；
5) probability density function of data is obtained；
6) the accumulation probability distribution of data is obtained.
The present invention can fully consider many uncertain factors of power grid, the Probabilistic Load Flow side based on Monte-Carlo Simulation
Method, establishes the calculation procedure of estimation electric power networks trend, and is applied to practical transmission system, analyzes electric power networks capacity water
Gentle load factor etc. preferably instructs the planning and assessment of power grid.
Detailed description of the invention
The content of width attached drawing every in description of the invention expression is briefly explained below:
Fig. 1 is based on the Probabilistic Load Flow data capture method flow chart under renewable energy uncertainty；
Fig. 2 is probabilistic load flow flow chart；
Fig. 3 is that LHS samples schematic diagram；
Fig. 4 is the probability-distribution function fitting flow chart for calculating data；
Specific embodiment
It is handed over using above-mentioned probability load flow calculation method extensive in the case where considering the access of extensive renewable energy
Galvanic electricity net calculates bus voltage amplitude, line load rate (including line and transformer) institute value range that may be present, carries out
Statistical analysis assessment.
Comprehensively consider all kinds of uncertain factors, including network topology, node load, generator output etc. in AC network
The randomness of variable obtains bus electricity using mentioned-above probability load flow calculation method and probability-distribution function approximating method
Pressure amplitude value, line load rate given utilization rate (transformer is the inverse of capacity-load ratio, and route is load factor) section confidence level,
To calculate electric network swim distribution situation.
As shown in Figure 1, calculation method is made of following three main modulars:
AC power flow in S2 calculates (SerialPF) module: being based on newton-La Fuxun (Newton-Raphson, NR)
Method or the AC network Load flow calculation for arranging literary Burger-Ma Kuaerte (Levenberg-Marqardt, LM) method；
Probabilistic load flow (Probabilistic Load Plow, PLF) module in S2: it is based on Monte Carlo simulation
The probabilistic load flow of (Monte Carlo Simulation, MCS) method considers the randomness of node load, generator output
And the failure rate of route；
S4 electric power networks evaluation module: Density Estimator is carried out to the calculated result of PLF module, obtains busbar voltage width
Value, line load rate are in the confidence level for giving section.
Wherein, SerialPF module is the core of PLF module, and PLF module provides the necessary number needed for calculating for TCA module
According to.Illustrate the data flow and implementation procedure of probabilistic load flow below:
Input data: dat file, independent variable configuration file and the equipment utilization section configuration file of PSD-BPA.
S1 input data: the dat file of PSD-BPA: the data file of record power grid electric parameter；
Independent variable configuration file: the text file of description AC network randomness includes 3 data chapters and sections, is followed successively by
MCS samples scale, bus independent variable parameter, route independent variable parameter；
S3 input data: equipment utilization section configuration file gives the text file in network equipment section, includes 3 numbers
According to chapters and sections, it is followed successively by bus voltage amplitude section, line load rate section, transformer load rate section.
Output data: AC power flow calculated result, probabilistic load flow result and utilization rate assessment result.
AC power flow calculated result: the calculation of tidal current under base regime；
Probabilistic load flow result: the probabilistic load flow result based on MCS method；
Network evaluation result: the electrical equipment utilization rate assessment result based on KDE.
It is the probability load flow calculation method based on stochastical sampling that AC power flow in S2, which calculates (SerialPF) module,；
Probabilistic Load calculation formula are as follows:
In formula: X is input stochastic variable, the fluctuation including node load, generator failure etc.；W and Z be output with
Machine variable, W represent node voltage amplitude and phase angle, and it is active and idle that Z represents Line Flow；F is node power equilibrium equation, g
For Line Flow equation.The probability distribution of given input stochastic variable X, the result of probabilistic load flow are output stochastic variable W
With the probability distribution and statistics of Z.
Probability load flow calculation method based on stochastical sampling is as shown in Fig. 2, include the following steps:
1) input stochastic variable probability-distribution function is obtained
2) input stochastic variable sample matrix is obtained
3) Load flow calculation is carried out based on sample matrix, and saves result
4) sample matrix of output variable is obtained
The particular content of each step is as follows:
1) probability-distribution function of input stochastic variable is obtained
Power system load has the characteristics that time variation, randomness, in the analysis of electric system stochastic problem, it is generally recognized that
Load is the stochastic variable of Normal Distribution.Assuming that it is mi that the active pi of the load of node i, which obeys desired value, standard deviation is σ i's
Normal distribution, the then probability density function of stochastic variable pi are as follows:
The reactive load of node i is it can also be assumed that be the stochastic variable of Normal Distribution, or the change active with load
Change and changes and remained unchanged with the power factor for guaranteeing node i.
Equipment fault will lead to topological structure of electric and change, and the trend of system can also change correspondingly.In traditional tide
Flow point analysis in, usually choose critical circuits perhaps generator analysis N-1 or N-2 after system load flow be distributed situation of change.
The consequence that this kind of method analysis failure generates, does not account for a possibility that failure occurs but, and do not account for systematic influence more
Big higher-dimension failure.A kind of more reasonable analysis method should consider the consequence of the probability that failure occurs and generation simultaneously, from risk
Angle carrys out the influence that comprehensive measurement failure generates system.In the project, route or power generation are described using binomial distribution
The failure of machine.Bi-distribution be t it is independent be/it is non-test in successful number discrete probability distribution, it is assumed that test every time
Successful probability is r, then just obtains k successful probability in t test and provided by following probability mass functions:
In formula: (﹒)！For factorial function.
In generator model, it is assumed that a generating set has the identical generator of t platform, and every generator failure probability is r,
In circuit model, t=1 is usually assumed that, it is that Bernoulli Jacob is distributed that binomial distribution, which is degenerated, at this time.
2) input stochastic variable sample matrix is obtained
It is random for k-th of input to become after the probability-distribution function of given load fluctuation, generator and line fault
Measure Xk(k=1,2 ..., n), it is assumed that its probability-distribution function is Yk=Fk(Xk).Setting sampling scale is N, then takes equal on 0-1
The pseudo random number y of even distributionk,i, X is calculated according to the following formulakSample xk,i:
The sampled value of each stochastic variable is in line, the sampling matrix X of n × N rank is formed.
3) Load flow calculation is carried out based on sample matrix
I-th (i=1,2 ..., N) row successively based on sample matrix X, using one pressgang of newton is inferior or Fa Liewen Burger-
The AC network Load flow calculation of Ma Kuaertefa carries out Load flow calculation.
4) sample matrix of output variable is obtained
The calculation of tidal current for saving i-th (i=1, the 2 ..., N) row based on sample matrix X, obtains output stochastic variable
The sample matrix of (node voltage amplitude and phase angle, Line Flow are active and idle), the expectation of analysis output stochastic variable, variance
Etc. each rank square.
Probability load flow calculation method in S2 includes: using Monte Carlo simulation approach
1) Load flow calculation data are read, input variable information is read, determines probabilistic load flow parameter；
2) input variable sample matrix is obtained based on Monte Carlo sampling techniques；
3) Load flow calculation is carried out using sample matrix, and saves output variable information；
4) probability distribution and statistics of analysis node voltage magnitude phase angle and Line Flow.
Specifically, it is assumed that x is the input stochastic variable of n dimension, and the relationship of output variable z and input variable x are
Z=h (x)
Monte Carlo simulation obtains the sample x of input variable x by sampling to input stochastic variable x{1},x{2},...,x{N}, and by sample x{i}It substitutes into, to obtain the sample z of output stochastic variable{i}:
z{i}=h (x{i})
Obtain sample x{1},x{2},...,x{N}Sample mode there are many, there are commonly simple random sampling (Simple
Random sampling, SRS) and Latin Hypercube Sampling (Latin hypercube sampling, LHS).In SRS, sample
This arbitrarily chooses from the valued space of stochastic variable；And LHS is then made of two steps that sample and sort, specific algorithm is such as
Under:
1) it samples.Assuming that stochastic variable xkThe cumulative distribution function of (k=1,2 ..., n) is Fk(xk).By FkTake
Value space average is divided into N number of section, and a number is arbitrarily chosen from each section as FkSampled value, then xkSampled value beThe sampled value of each stochastic variable is in line, the sampling matrix X, LHS of n × N rank are formed
It is as shown in Figure 3 to sample schematic diagram；
2) it arranges.When inputting stochastic variable independence, the correlation between stochastic variable sampled value is inputted to calculated result
There is certain influence, sequence can reduce the correlation between sampled value.There are many kinds of sort methods, such as Cholesky to decompose,
Columnwise-pairwise algorithm, Single-switch optimization method etc..Wherein, the arrangement side decomposed based on Cholesky
Method are as follows:
Assuming that L is the sequential matrix of n × N, every a line is arranged corresponding to 1, the 2 ..., N of sampling matrix X.L's is linear
Correlation matrix is the ρ of n × n dimensionL, according to definition ρLTo be poised for battle positive definite matrix, therefore can be by ρLCarry out Cholesky decomposition:
ρL=DDT
In formula, D is lower triangular matrix.
N × N-dimensional matrix G can be calculated by following formula:
G=D-1L
Different from matrix L, the element of matrix G is not necessarily positive integer.According to G generator matrix G', so that every a line of G'
It is the arrangement of 1,2 ..., N, and its arrangement for putting in order corresponding to row element each in G.According to the row of each row element of G'
Column update the arrangement of each row element of sampling matrix X.
As shown in figure 4, S4 is fitted output variable probability-distribution function using Density Estimator, kernel function K is selected as Gauss
Kernel function obtains node voltage amplitude and phase angle, the active and idle probability-distribution function of Line Flow.
Specifically, the probabilistic load flow based on simulation obtains the sample data of Line Flow, needs to sample number
According to being analyzed, the probability density function and cumulative distribution function of Line Flow power are obtained, to assess electric network security.
Although the specific parameter distribution of input variable obedience of Probabilistic Load Flow, such as normal distribution, binomial distribution etc., due to
The non-linear behavior of power flow equation, so the probability distribution of the output variable (node voltage and Line Flow) of Probabilistic Load Flow is multiple
It is miscellaneous, multimodal characteristic is usually showed, can not be fitted with simple parameter distribution.If therefore it is simple using normal distyribution function come
The probability distribution of Line Flow is described, error is often larger；One kind more reasonable manner is obtained using non-parametric estmation
The probability-distribution function of Line Flow.Density Estimator is typical Nonparametric Estimation, and theoretical basis is as follows:
Density Estimator (Kernel density estimation, KDE) is also referred to as the estimation of Parzen window, is 20 generation
It records a kind of density estimation method that five sixties proposed and grew up, it is a kind of effective Nonparametric Estimation.It is false
If x{1},x{2},...,x{N}For the sample of stochastic variable x, the probability density function of variable x is f (x), then the Density Estimator of f
It can indicate are as follows:
In formula: h is bandwidth, and N is sample size, and K () is kernel function, and meets following condition:
As N → ∞, h → ∞ and Nh → ∞,Convergence in (with)probability is in f.The shape and codomain of kernel function, which control, to be used to
The degree estimating the number of f data point used in the value of point x and utilizing, common kernel function includes homogeneous nucleus, nucleus vestibularis triangularis, Gauss
Core etc..This project selects gaussian kernel function, expression formula are as follows:
According to Density Estimator theory, compared to kernel function, influence of the bandwidth h to fitting result is bigger.In different bandwidth h
Under, Density Estimator result difference is larger.When bandwidth h is smaller, Density Estimator curveMore tortuous, slickness is very poor,
Show the unexistent multimodal characteristic of former probability density function f；When bandwidth h is larger, Density Estimator curveIt is more smooth,
But more details can be covered.Therefore it is extremely important and necessary for selecting suitable bandwidth h.Usually there are two class bandwidth selection sides
Method: cross-validation method (Cross-validation method, CV) and insertion (Plug-in method, PI).
1) cross-validation method
Density Estimator valueDifference with theoretical probability density function f may be expressed as:
In above formula, first item can be calculated by sample data, and Section 2 is represented by estimated valueExpectation, Section 3
It is unrelated with bandwidth h, it can be omitted, simplify are as follows:
The main task of cross-validation method is the Section 2 in accurate estimator.
3) insertion
Density Estimator valueWith the expectation of the difference of theoretical probability density function f are as follows:
In formula:
Define asymptotic mean square error (Asymptotic Mean Squared Error, AMISE) are as follows:
Then asymptotic optimum bandwidth are as follows:
In formula, only R (f ") is unknown, so the main task of insertion is the value of reasonable estimation R (f ").
For the estimation of R (f "), scholar Silverman proposes an empirical method: it is assumed that f is normal probability density letter
Number N (0, σ2), then R (f ")=(3 π-0.5σ-5)/8, if selecting gaussian kernel function, available optimum bandwidth h=1.06 at this time
σn-1/5.When data are close to when normal distribution, this method is a good selection.However, ought really be distributed as it is asymmetric or
When being multimodal, this method may cause excess smoothness.Therefore scholar proposes to use following bandwidth selection mode with good fit
Unimodal or bimodal probability function:
H=min { σ, R/1.34 } 1.06n-1/5
In formula: σ is the standard deviation of sample, and R is the interquartile-range IQR of sample.
The expression way of renewable resource of the invention is as follows:
1, photo-voltaic power supply equivalent model
The output of photovoltaic generating system is largely illuminated by the light the influence of intensity.Therefore, intensity of illumination is at some
Probability density function in period can be indicated with following formula.
In formula,PMIt is solar battery in certain period
Export general power；RMIt is solar battery in certain period Maximum Power Output；Γ is Gamma function.
The then active power output of entire photovoltaic system are as follows:
In formula, A is photovoltaic array area；η is efficiency；T, T' are ginseng related with geographical location and photovoltaic system itself
Number.
The Equivalent Model of photo-voltaic power supply generally there are two types of, one is be seen as invariable power current source；Another kind is
It is seen as voltage source.This project will take second of Equivalent Model, and photo-voltaic power supply is equivalent to a voltage source and impedance
It is connected in series.Photo-voltaic power supply will be handled as follows in calculation of fault: when distribution power system load flow calculation containing photo-voltaic power supply, by it
It is equivalent to the computation model of voltage source series impedance.
Distributed generation resource with inverter can usually keep node voltage amplitude due to the adjustment effect of inverter
It is constant, therefore such node can be seen as PV node.
2, Wind turbines equivalent model
Know Wind turbines active power of output and reactive power are as follows:
Wherein, P is Wind turbines active power, and Q is Wind turbines reactive power, and U is wind-powered electricity generation node voltage amplitude, and x is
The sum of generator unit stator reactance and rotor reactance, xmFor excitation reactance, r is rotor resistance, and s is revolutional slip.By Wind turbines with
PQ node considers.
3, miniature gas turbine equivalent model
Generator used in miniature gas turbine is High Speed Permanent Magnet Synchronous Generator, it and usually said synchronous generator
The difference of machine is magneto alternator using permanent magnet excitation, therefore, the model and synchronous generator of miniature gas turbine
Model it is similar, as long as energized circuit parameter is set as definite value.
Its power out-put characteristic such as following formula:
In formula, η is the efficiency that turbine power converts electrical power；ωfFor fuel flow rate；N is revolving speed.
The high-frequency alternating current that miniature gas turbine issues needs to be converted into industrial-frequency alternating current, can just be incorporated to power distribution network, have
The output of function power has the characteristic for being similar to centralization power generation:
Wherein, Vi' it is original machine power；X'fjFor output power of motor；Pm,in,Pm,outFor prime mover input, output work
Rate；ωR, ω be generator give, the angular speed of receiving end.
For miniature gas turbine during accessing power distribution network, Equivalent Model can regard synchronous generator and inverter as
The collective model that model combines, by the adjusting control action of inverter, miniature gas turbine can issue reactive power, protect
It is constant to hold output voltage, therefore, PV node can be regarded as during simulation calculation.
4, gird-connected inverter equivalent model
The electromotive force U of the equivalent model of inverter0And equivalent output impedance Z calculation formula is as follows:
By above two formula, the open-circuit voltage and output impedance of the inverse distributed power under power frequency can be calculated.
The probabilistic electrical network analysis probabilistic model expression way of the present invention is as follows:
1, Load Probability model
Regard load prediction results as a stochastic variable S, and assume its Normal Distribution, therefore, can be used just
The uncertainty of the approximate reflection load of state distribution.The Load Probability model is verified in long-term practice, in a large amount of probability tides
It flows and is used widely in relevant document.Assuming that some time is carved with the mean value of workload P and variance is respectively μPAnd σP, idle negative
The mean value and variance of lotus Q is respectively μQAnd σQ, then the active probability density function f (P) of load and idle probability density function f (Q)
Expression formula respectively indicates are as follows:
After obtaining each node load predicted value in future time section, its base can be established according to its mean value and variance parameter
In the Load Probability model of normal distribution.
2, generator outage probability model
Generally assuming that generator, there are two states: one is because of state of stopping transport caused by maintenance or failure, one is
Normal operating condition.The probability distribution P of Generator Status X obeys two o'clock point, is as follows:
Wherein, 0 indicate that stoppage in transit state, generated power power output are 0MW, 1 indicates normal operating condition, in normal operating condition
When, generated power is contributed between unit minimum load and maximum output.PFOR(PFOR< 1) indicate unit outage rate, be
Stoppage in transit hour and hour of stopping transport add the ratio of the sum of hours run.This model is using the index as generator outage probability to its shape
State is sampled.
When progress generator probabilistic model is established, some generator node may be made of multiple generating sets,
It can be then based on above-mentioned Two-point distribution model, be distributed using the outage probability that enumerative technique acquires generator node.
Assuming that certain generator node G is made of the unit (G1, G2, G3) of three 1000MW capacity, outage rate is
0.05, the available genset of the node and corresponding probability can be as shown in following tables:
3, transmission facility outage probability model
It is equal to the outage probability model of generator, the outage probability model of transmission facility also assumes that there are two states:
One is stoppage in transit states, and one is operating statuses, and Two-point distribution is obeyed in state distribution.
The transmission facility that the present invention is considered includes transmission line of electricity and transformer, when carrying out Load flow calculation and analysis, by
It stops transport in transmission facility, grid structure may change, and need to carry out isolated island judgement to research system.If there is isolated island
Situation need to carry out specially treated when calculating trend.
For the present invention using above-mentioned Probabilistic Load Flow algorithm model under the power grid normal operating mode of Wuhu Region, typical peak is negative
Lotus day data are calculated, and power-balance statistics is 2542.839+1.507i (MW/MVar) for total power generation, and total load is
2429.7-5421.33i (MW/MVar), total losses are 113.192+ 5422.837i (MW/MVar), and line loss per unit is
4.45132%, it can be restrained through Load flow calculation.
The present invention is exemplarily described above in conjunction with attached drawing, it is clear that the present invention implements not by aforesaid way
Limitation, as long as the improvement for the various unsubstantialities that the inventive concept and technical scheme of the present invention carry out is used, or without changing
It is within the scope of the present invention into the conception and technical scheme of the invention are directly applied to other occasions.
Claims (5)
1. based on the Probabilistic Load Flow data capture method under renewable energy uncertainty, it is characterised in that:
S1 inputs the data file of power grid electric parameter, configures power grid random parameter；
S2 carries out AC network Load flow calculation and probabilistic load flow, and output MCS samples scale, bus independent variable parameter, line
Road independent variable parameter；
S3 configures grid equipment section, including bus voltage amplitude section, line load rate section, transformer load rate section；
S4 obtains electrical equipment utilization rate assessment result using Density Estimator.
2. the Probabilistic Load Flow acquisition methods under the uncertainty according to claim 1 based on renewable energy, feature exist
In: if the power grid electric parameter and grid equipment section in S1 and S3 change, the data in real-time update S1 and S3.
3. the Probabilistic Load Flow acquisition methods under the uncertainty according to claim 1 based on renewable energy, feature exist
In: in the S2, AC network tidal current computing method is based on one pressgang of newton is inferior or the friendship of Fa Liewen Burger-Ma Kuaertefa
Flowing power grid load flow calculation method includes:
1) input stochastic variable probability-distribution function is obtained；
2) input stochastic variable sample matrix is obtained；
3) Load flow calculation is carried out based on sample matrix, and saves result；
4) sample matrix of output variable is obtained.
4. the Probabilistic Load Flow acquisition methods under the uncertainty according to claim 1 based on renewable energy, feature exist
In: in the S2, probability load flow calculation method includes: based on Monte Carlo simulation approach
1) Load flow calculation data are read, input variable information is read, determines probabilistic load flow parameter；
2) input variable sample matrix is obtained based on Monte Carlo sampling techniques；
3) Load flow calculation is carried out using sample matrix, and saves output variable information；
4) probability distribution and statistics of analysis node voltage magnitude phase angle and Line Flow.
5. the Probabilistic Load Flow acquisition methods under the uncertainty according to claim 1 based on renewable energy, feature exist
In: the probability-distribution function approximating method in the S4 based on Density Estimator includes:
1) probabilistic load flow structure is read；
2) node voltage data or Line Flow data are selected；
3) Density Estimator function is selected；
4) it calculates cuclear density and estimates width；
5) probability density function of data is obtained；
6) the accumulation probability distribution of data is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811034650.2A CN109066688A (en) | 2018-09-06 | 2018-09-06 | Based on the Probabilistic Load Flow data capture method under renewable energy uncertainty |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811034650.2A CN109066688A (en) | 2018-09-06 | 2018-09-06 | Based on the Probabilistic Load Flow data capture method under renewable energy uncertainty |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109066688A true CN109066688A (en) | 2018-12-21 |
Family
ID=64759811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811034650.2A Pending CN109066688A (en) | 2018-09-06 | 2018-09-06 | Based on the Probabilistic Load Flow data capture method under renewable energy uncertainty |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109066688A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109802394A (en) * | 2019-04-01 | 2019-05-24 | 东北大学 | A kind of probability load flow calculation method counted and distributed generation resource and electric car access |
| CN109921474A (en) * | 2019-03-04 | 2019-06-21 | 江苏师范大学 | A kind of distribution network electric energy loss Stochastic Analysis Method considering uncertain factor |
| CN110112790A (en) * | 2019-06-06 | 2019-08-09 | 南方电网科学研究院有限责任公司 | Method, device, equipment and storage medium for evaluating operating efficiency of active power distribution network |
| CN110311427A (en) * | 2019-06-18 | 2019-10-08 | 华北电力大学 | The two stages N-K robust Fault of meter and probability of malfunction constrains Unit Combination method |
| CN110427659A (en) * | 2019-07-13 | 2019-11-08 | 潍坊学院 | A kind of threephase asynchronous active power based on clustering determines method |
| CN110428164A (en) * | 2019-07-29 | 2019-11-08 | 广东电网有限责任公司 | A kind of operation of power networks state monitoring method, device, equipment and readable storage medium storing program for executing |
| CN111125877A (en) * | 2019-11-19 | 2020-05-08 | 广西电网有限责任公司 | Active power distribution network reliability evaluation method based on Monte Carlo simulation |
| CN111224399A (en) * | 2020-02-25 | 2020-06-02 | 山东大学 | Method and system for analyzing power grid uncertainty trend |
| CN111525562A (en) * | 2020-05-16 | 2020-08-11 | 国网甘肃省电力公司兰州供电公司 | Voltage flicker rapid evaluation method and system for rail transit power supply |
| CN111682530A (en) * | 2020-06-11 | 2020-09-18 | 广东电网有限责任公司韶关供电局 | Method, device, equipment and medium for determining out-of-limit probability of voltage of power distribution network |
| CN111948574A (en) * | 2020-07-31 | 2020-11-17 | 电子科技大学 | Method for quickly positioning open-circuit fault of inverter |
| CN112018787A (en) * | 2020-08-25 | 2020-12-01 | 广东电网有限责任公司广州供电局 | Three-phase unbalanced time sequence simulation method of distributed power supply |
| CN112615386A (en) * | 2020-11-23 | 2021-04-06 | 国网浙江省电力有限公司台州供电公司 | Wind power consumption-oriented optimal constant volume method for electric heating hybrid energy storage system |
| CN113051829A (en) * | 2021-03-31 | 2021-06-29 | 西南大学 | Transformer Duval Pentagon1 fault diagnosis method improved by applying space analysis theory |
| CN113625074A (en) * | 2021-07-02 | 2021-11-09 | 深圳供电局有限公司 | Method and device for judging standard exceeding of long-time flicker caused by transient power quality disturbance |
| CN113642886A (en) * | 2021-08-12 | 2021-11-12 | 国网经济技术研究院有限公司 | Method, system, medium and computing device for analyzing utilization rate of planned power grid line |
| CN113884820A (en) * | 2021-11-19 | 2022-01-04 | 安徽南瑞中天电力电子有限公司 | Line impedance measuring method and line fault type analyzing method based on electric energy meter |
| CN113987416A (en) * | 2021-11-02 | 2022-01-28 | 东北石油大学 | Oil-gas resource amount calculation method and system based on confidence level |
| US11349306B2 (en) | 2019-06-19 | 2022-05-31 | King Fahd University Of Petroleum And Minerals | Distribution grid fault analysis under load and renewable energy uncertainties |
| CN116613759A (en) * | 2023-05-18 | 2023-08-18 | 广东电网有限责任公司 | Power distribution network tide calculation method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104079003A (en) * | 2014-07-21 | 2014-10-01 | 国家电网公司 | Probability load flow calculation method for photovoltaic power contained distribution network |
| CN104682387A (en) * | 2015-03-10 | 2015-06-03 | 东南大学 | Probability load flow calculation method based on multi-zone interactive iteration |
| CN107730111A (en) * | 2017-10-12 | 2018-02-23 | 国网浙江省电力公司绍兴供电公司 | A kind of distribution voltage risk evaluation model for considering customer charge and new energy access |
-
2018
- 2018-09-06 CN CN201811034650.2A patent/CN109066688A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104079003A (en) * | 2014-07-21 | 2014-10-01 | 国家电网公司 | Probability load flow calculation method for photovoltaic power contained distribution network |
| CN104682387A (en) * | 2015-03-10 | 2015-06-03 | 东南大学 | Probability load flow calculation method based on multi-zone interactive iteration |
| CN107730111A (en) * | 2017-10-12 | 2018-02-23 | 国网浙江省电力公司绍兴供电公司 | A kind of distribution voltage risk evaluation model for considering customer charge and new energy access |
Non-Patent Citations (1)
| Title |
|---|
| 宋元峰等: "基于核密度估计的概率分布函数拟合方法", 《电网与清洁能源》 * |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109921474A (en) * | 2019-03-04 | 2019-06-21 | 江苏师范大学 | A kind of distribution network electric energy loss Stochastic Analysis Method considering uncertain factor |
| CN109802394B (en) * | 2019-04-01 | 2022-05-17 | 东北大学 | Probability load flow calculation method considering distributed power supply and electric vehicle access |
| CN109802394A (en) * | 2019-04-01 | 2019-05-24 | 东北大学 | A kind of probability load flow calculation method counted and distributed generation resource and electric car access |
| CN110112790B (en) * | 2019-06-06 | 2023-10-20 | 南方电网科学研究院有限责任公司 | Method, device, equipment and storage medium for evaluating operation efficiency of active power distribution network |
| CN110112790A (en) * | 2019-06-06 | 2019-08-09 | 南方电网科学研究院有限责任公司 | Method, device, equipment and storage medium for evaluating operating efficiency of active power distribution network |
| CN110311427A (en) * | 2019-06-18 | 2019-10-08 | 华北电力大学 | The two stages N-K robust Fault of meter and probability of malfunction constrains Unit Combination method |
| CN110311427B (en) * | 2019-06-18 | 2020-12-01 | 华北电力大学 | Two-stage N-K robust fault constraint unit combination method considering fault probability |
| US11349306B2 (en) | 2019-06-19 | 2022-05-31 | King Fahd University Of Petroleum And Minerals | Distribution grid fault analysis under load and renewable energy uncertainties |
| CN110427659B (en) * | 2019-07-13 | 2022-11-25 | 潍坊学院 | Three-phase asynchronous motor active power determination method based on cluster analysis |
| CN110427659A (en) * | 2019-07-13 | 2019-11-08 | 潍坊学院 | A kind of threephase asynchronous active power based on clustering determines method |
| CN110428164B (en) * | 2019-07-29 | 2022-04-29 | 广东电网有限责任公司 | Power grid running state monitoring method, device and equipment and readable storage medium |
| CN110428164A (en) * | 2019-07-29 | 2019-11-08 | 广东电网有限责任公司 | A kind of operation of power networks state monitoring method, device, equipment and readable storage medium storing program for executing |
| CN111125877A (en) * | 2019-11-19 | 2020-05-08 | 广西电网有限责任公司 | Active power distribution network reliability evaluation method based on Monte Carlo simulation |
| CN111224399A (en) * | 2020-02-25 | 2020-06-02 | 山东大学 | Method and system for analyzing power grid uncertainty trend |
| CN111525562A (en) * | 2020-05-16 | 2020-08-11 | 国网甘肃省电力公司兰州供电公司 | Voltage flicker rapid evaluation method and system for rail transit power supply |
| CN111525562B (en) * | 2020-05-16 | 2023-07-21 | 国网甘肃省电力公司兰州供电公司 | Voltage flicker rapid evaluation method and system for rail transit power supply |
| CN111682530B (en) * | 2020-06-11 | 2022-06-28 | 广东电网有限责任公司韶关供电局 | Method, device, equipment and medium for determining out-of-limit probability of voltage of power distribution network |
| CN111682530A (en) * | 2020-06-11 | 2020-09-18 | 广东电网有限责任公司韶关供电局 | Method, device, equipment and medium for determining out-of-limit probability of voltage of power distribution network |
| CN111948574A (en) * | 2020-07-31 | 2020-11-17 | 电子科技大学 | Method for quickly positioning open-circuit fault of inverter |
| CN111948574B (en) * | 2020-07-31 | 2022-01-11 | 电子科技大学 | Method for quickly positioning open-circuit fault of inverter |
| CN112018787A (en) * | 2020-08-25 | 2020-12-01 | 广东电网有限责任公司广州供电局 | Three-phase unbalanced time sequence simulation method of distributed power supply |
| CN112018787B (en) * | 2020-08-25 | 2022-12-16 | 广东电网有限责任公司广州供电局 | Three-phase unbalanced time sequence simulation method of distributed power supply |
| CN112615386A (en) * | 2020-11-23 | 2021-04-06 | 国网浙江省电力有限公司台州供电公司 | Wind power consumption-oriented optimal constant volume method for electric heating hybrid energy storage system |
| CN112615386B (en) * | 2020-11-23 | 2023-04-07 | 国网浙江省电力有限公司台州供电公司 | Wind power consumption-oriented optimal constant volume method for electric heating hybrid energy storage system |
| CN113051829A (en) * | 2021-03-31 | 2021-06-29 | 西南大学 | Transformer Duval Pentagon1 fault diagnosis method improved by applying space analysis theory |
| CN113625074A (en) * | 2021-07-02 | 2021-11-09 | 深圳供电局有限公司 | Method and device for judging standard exceeding of long-time flicker caused by transient power quality disturbance |
| CN113642886A (en) * | 2021-08-12 | 2021-11-12 | 国网经济技术研究院有限公司 | Method, system, medium and computing device for analyzing utilization rate of planned power grid line |
| CN113642886B (en) * | 2021-08-12 | 2024-05-28 | 国网经济技术研究院有限公司 | Method, system, medium and computing equipment for analyzing utilization rate of planned power grid line |
| CN113987416A (en) * | 2021-11-02 | 2022-01-28 | 东北石油大学 | Oil-gas resource amount calculation method and system based on confidence level |
| CN113884820A (en) * | 2021-11-19 | 2022-01-04 | 安徽南瑞中天电力电子有限公司 | Line impedance measuring method and line fault type analyzing method based on electric energy meter |
| CN116613759A (en) * | 2023-05-18 | 2023-08-18 | 广东电网有限责任公司 | Power distribution network tide calculation method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109066688A (en) | Based on the Probabilistic Load Flow data capture method under renewable energy uncertainty | |
| CN102012956A (en) | Wind farm equivalent method based on wind farm input wind speed and wind direction chance fluctuation | |
| CN110854884B (en) | Wind power collection area subsynchronous oscillation risk online assessment and early warning method | |
| Weng et al. | Effects of wind power variability and intermittency on power flow | |
| CN106712032A (en) | Optimal power flow model construction method considering active power voltage regulation capacity of wind turbine generator set | |
| Mei et al. | Recent advances on smart grid technology and renewable energy integration | |
| CN103761574A (en) | Distributed power supply and region load matched feature matching method | |
| CN116937685A (en) | Renewable energy power grid time series production simulation method based on security domain | |
| Liu et al. | Increasing wind power penetration level based on hybrid wind and photovoltaic generation | |
| Ma et al. | Integrated strategy of the output planning and economic operation of the combined system of wind turbines-pumped-storage-thermal power units | |
| Ren et al. | Voltage Stability Analysis of Front‐End Speed Controlled Wind Turbine Integrated into Regional Power Grid Based on Bifurcation Theory | |
| Wang et al. | Optimal recovery strategy of DERs integrated distribution network based on scheduling rationality | |
| Yang et al. | Research on the reliability and capacity allocation of wind power-solar power-pumped storage hybrid power system | |
| Nanda et al. | Classification of faults in a wind connected power system using artificial neural network | |
| Wu et al. | Cyber‐Enabled Intelligence Control and Security Optimization for Complex Microgrid Networks Transient Frequency Stability Analysis of Power Systems considering Photovoltaic Grid Connection | |
| He et al. | Damping characteristics improvement of a wind-PV-thermal-bundled power systems by coordinating optimization of controller parameters | |
| Su et al. | Planning optimization for the photovoltaic/storage/network system considering photovoltaic consumption level and grid supporting capacity | |
| Fan et al. | Temporal and Spatial Distribution of Power System Voltage based on Generalized Regression Neural Network | |
| Tan et al. | Stochastic optimal power flow with wind generator based on stochastic response surface method (SRSM) and interior point methods | |
| Singh et al. | Classification of Faults in a Distributed Generator Connected Power System Using Artificial Neural Network | |
| Panwar et al. | Significance of dynamic and transient analysis in the design and operation of hybrid energy systems | |
| Yang et al. | Interval State Estimation Considering Randomness of Multiple Distributed Generations in Active Distribution Networks | |
| Cao et al. | Study on reliability assessment of composite generation and transmission system integrated wind farm | |
| Brown | Evaluation of integration of pumped storage units in an isolated network | |
| Bostani et al. | Damping sub-synchronous resonance in DFIG wind farms: An innovative controller utilizing wide-area measurement systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181221 |
|
| RJ01 | Rejection of invention patent application after publication |