CN106855909A - A kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic - Google Patents

A kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic Download PDF

Info

Publication number
CN106855909A
CN106855909A CN201710027392.4A CN201710027392A CN106855909A CN 106855909 A CN106855909 A CN 106855909A CN 201710027392 A CN201710027392 A CN 201710027392A CN 106855909 A CN106855909 A CN 106855909A
Authority
CN
China
Prior art keywords
distribution network
power distribution
active power
stochastic
emulation
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
Application number
CN201710027392.4A
Other languages
Chinese (zh)
Other versions
CN106855909B (en
Inventor
宋毅
孙充勃
原凯
赵金利
范朕宁
王成山
李鹏
靳夏宁
薛振宇
宋关羽
吴志力
崔凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin University
State Grid Economic and Technological Research Institute
Original Assignee
Tianjin University
State Grid Economic and Technological Research Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tianjin University, State Grid Economic and Technological Research Institute filed Critical Tianjin University
Priority to CN201710027392.4A priority Critical patent/CN106855909B/en
Publication of CN106855909A publication Critical patent/CN106855909A/en
Application granted granted Critical
Publication of CN106855909B publication Critical patent/CN106855909B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/30—Circuit design
    • G06F30/36—Circuit design at the analogue level
    • G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/18—Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic of the invention, belongs to active power distribution network emulation field.Consider the stochastic behaviour and Multiple Time Scales feature in active power distribution network running, the method is based on showing implicit hybrid solving thought, the Disturbance model for constituting active power distribution network stochastic dynamic model is integrated using progressive failure, the deterministic models for constituting active power distribution network stochastic dynamic model are integrated using implicit integration algorithm.The method can give full play to progressive failure and the numerical value of implicit integration algorithm resolves advantage, and can be adapted to stochastic and dynamic simulation calculation of the system under failure or operating condition.Compared with traditional explicit integration method, while numerical precision needed for meeting emulation, computation efficiency and numerical stability are improve, realize the active power distribution network stochastic and dynamic emulation considered in the case of photovoltaic, wind generator system Integrated access.

Description

A kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic
Technical field
The present invention relates to a kind of active power distribution network emulation mode.It is more particularly to a kind of to consider photovoltaic, wind generator system And the aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic of load random walk characteristic.
Background technology
As distributed power source (distributed generator, DG) largely accesses distribution network, and Demand-side rings Continuing to develop for advanced multiplexe electric technology should be waited, the random walk characteristic that distributed power source and load have is to the safety with electricity consumption side Reliability service brings new challenge.The random walk characteristic of system can directly affect the dynamic response characteristic of distribution network, enter And the dynamic characteristic of whole system is influenceed, particularly fault characteristic.But in Practical Project field, often cannot be directly true Tested in system to analyze and study relevant issues, therefore generally ground as main using effective digital simulation method Study carefully means.
With in active power distribution network, enchancement factor is on the increase, emulation caused by random perturbation in system operation Result difference also more substantially, has strong stochastic behaviour based on certainty power distribution network Stability Modeling and emulation mode in treatment Had some limitations during Modern power distribution network simulation.Active power distribution network certainty Dynamic Simulation Model lost a part of system Information in true running, such as variable dynamic behaviour of functional relation and active power distribution network etc..These systems are transported Disturbance during row mostlys come from light irradiance, the random fluctuation of wind speed in distributed generation system, and Demand Side Response is led The load change at random of cause, the measurement error of transient-upset and control element when synchronous generator rotor is rotated etc..Therefore, As the increasingly complication of power distribution network, the random perturbation from system can not be ignored, need to build active power distribution network with motor-driven State simulation model.
Therefore, can be by one group of stochastic differential-algebraic equation (stochastic differential-algebraic Equation, SDAE) system is modeled：
In formula, f () is the differential equation, for the dynamic process of descriptive system；G () is that algebraic equation is used to describe system The trend constraint of system；Vector x represents all state variables in system, such as synchronous machine generator rotor angle, rotating speed, power electronic equipment control Variable etc.；Vectorial y represents all algebraic variables in system, such as rotor electric current, each node voltage amplitude of system, phase angle Deng；Vectorial η represents the random disturbance quantity in system, such as light irradiance, wind speed, temperature etc.；ξ represents white Gaussian noise process, For the change at random process of simulated disturbance variable.
Different from certainty dynamic simulation, the emulation of active power distribution network stochastic and dynamic is main by a plurality of of simulation system variable Dynamic trajectory, the corresponding situation with analysis system operation characteristic under system random perturbation.Therefore, Monte-Carlo Simulation (Monte Carlo simulation) numerical solution becomes the most frequently used method for solving of SDAE, and the essence of the method is based on substantial amounts of Simulating scenes, carry out Multi simulation running, error convergence situation and the simulation times correlation of emulation.To ensure numerical value essence Degree, generally requires by substantial amounts of simulation calculation.Further, since different types of distributed power source largely accesses power distribution network, both It is related to, including the static direct current link including photovoltaic array, batteries to store energy, further relate to include small size asynchronous blower fan, permanent-magnet synchronous Blower fan is in interior exchange link so that active power distribution network running is increasingly complex, not of the same race compared in conventional electric power system The component arrangement of class, a large amount of distributed electrical source control systems in active power distribution network need to realize having by power electronic equipment Faster dynamic response characteristic, making the gap of each interelement dynamic time constant in system becomes more notable so that active to match somebody with somebody Power network in the process of running, with more obvious Multiple Time Scales characteristic.The Multiple Time Scales of active power distribution network dynamic simulation are special Levying can mathematically be attributed to stiff problem, be the precision and stabilization for effectively solving the numerical algorithm that this problem need to be to being used Property propose requirement higher.
Active power distribution network dynamic simulation algorithm can be divided into explicit integration side according to the resolving form difference to system model Method and the major class of implicit integration method two.In explicit numerical algorithm, calculating process is simple, and computation efficiency is high, but due to it Numerical stability is poor, process Multiple Time Scales characteristic problem when, easily there is numerical value unstability, be typically only possible by use compared with Small step is long to be emulated, and computational efficiency is limited by larger.And Implicit Numerical Algorithm has good numerical stability, at place During reason stiff problem, larger simulation step length, but implicit algorithm may be selected, compared to explicit, per time step under calculating it is more complicated, Need iteration to be solved, hamper the further lifting of computational efficiency.
In the significant active power distribution network dynamic simulation of consideration stochastic behaviour, on the one hand need before simulation accuracy is met Put, reduce the emulation used time；On the other hand, it is necessary to consider system model Multiple Time Scales characteristic, it is to avoid numerical value unstability occur.
The content of the invention
The technical problems to be solved by the invention be to provide it is a kind of while simulation accuracy and numerical stability is ensured, Realize the active power distribution network stochastic and dynamic emulation in the case of consideration photovoltaic, wind generator system Integrated access is suitable to active matching somebody with somebody The aobvious hidden mixed integrating method method of power network stochastic and dynamic emulation
The technical solution adopted in the present invention is：A kind of aobvious hidden mixed integrating method for being suitable to the emulation of active power distribution network stochastic and dynamic Method, comprises the following steps：
1) stochastic differential that active power distribution network topological connection relation, dynamic element parameter are set, characterize random perturbation variable Equation parameter and simulation calculation parameter, and time of origin, the type of failure or operation are set, wherein simulation calculation parameter includes Emulation termination time T, simulation step length Δ t, Monte-Carlo Simulation total degree I；
2) reading active power distribution network topological relation carries out Load flow calculation, obtains system load flow result of calculation；
3) according to system load flow result of calculation, simulation initialisation is carried out to the dynamic element in active power distribution network；
4) the initial number of times i=1 of Monte-Carlo Simulation is set；
5) this emulation initial time t=0 is set；
6) judge whether current emulation moment t is emulation initial time, if t=0, need setting active power distribution network to become at random Amount initial value simultaneously skips to step 8), otherwise into next step；
7) t is maden-1~tnPeriod calculates, and is had to constituting active power distribution network stochastic dynamic model using progressive failure Source power distribution network Disturbance model integrates 1 step-length and obtains tnThe system stochastic variable at moment is simultaneously preserved, using Euler-ball mountain Method is solved to the stochastic differential equation in active power distribution network stochastic dynamic model；
8) in tnAt the moment, random perturbation variable is transmitted, realize Disturbance model and active power distribution network deterministic models Between data transfer；
9) t is carried outn~tn+1Moment simulation calculation, wherein emulation moment t=t+ Δ t, Δ t is simulation step length, using implicit Integral algorithm integrates 1 step-length to the active power distribution network deterministic models for constituting active power distribution network stochastic dynamic model, obtains tn+1 Moment system state variables and algebraic variable are simultaneously preserved, and differential-algebraic equation group is asked using implicit simultaneous solution method Solution, using implicit backward Euler method to the differential equation in differential-algebraic equation group；
10) according to step 1) to judge whether system occurs in the case where moment t is emulated former for the simulated fault that sets and Action Events Barrier is operated, if not occurring, skips to step 12) continue executing with, otherwise into next step；
11) when breaking down or operating, setting system state variables is constant, according to active power distribution network deterministic models side Journey, recalculates system algebraic variable, by system variable result of calculation again assignment, return to step 7)；
12) judge whether emulation moment t reaches emulation end of a period moment T, if not up to emulation ends moment, return to step 7), otherwise into next step；
13) Monte-Carlo Simulation number of times i=i+1 is set, judges simulation times i whether more than Monte-Carlo Simulation total degree I, if being not more than, i.e. i≤I, then return to step 5), otherwise into next step；
14) I Monte-Carlo Simulation result is collected, simulation result is exported according to emulation demand and is drawn, terminated Artificial tasks.
Step 7) described in active power distribution network Disturbance model be to be characterized by one group of stochastic differential equation：
In formula, η ∈ RKFor the vector that K n-dimensional random variable ns are constituted；α () and β () is scalar equation, represents random respectively The side-play amount and diffusing capacity of the differential equation；x∈RNFor the vector that N-dimensional state variable is constituted；y∈RMIt is M dimension algebraic variable compositions Vector；ξ is white Gaussian noise process；
1 step-length is integrated to Disturbance model using progressive failure, stochastic differential equation therein is solved Using Euler-ball mountain method, wherein, step-length is Δ t, and the time is from tn-1To tn, by ηn-1Obtain ηn, recurrence formula is as follows：
ηn=ηn-1+α(xn-1,yn-1,ηn-1)Δt+β(xn-1,yn-1,ηn-1)ξn-1。
Step 9) described in active power distribution network deterministic models be to be characterized by one group of differential-algebraic equation：
In formula, x ∈ RNFor the vector that N-dimensional state variable is constituted；y∈RMIt is the vector that M dimension algebraic variables are constituted, due to step It is rapid 8) to deliver active power distribution network random perturbation variable, therefore η is given value in equation；
Described use implicit integration algorithm integrates 1 to the deterministic models for constituting active power distribution network stochastic dynamic model Step-length, is solved using implicit simultaneous solution method to differential-algebraic equation group, using implicit backward Euler method to differential- Differential equation in Algebraic Equation set, specific solution procedure is as follows：
(1) 1 step is integrated using backward Euler method, step-length is Δ t, and the time is from tnTo tn+1, by xnObtain xn+1, recurrence formula For
(2) by differential equation f () in differential-algebraic equation through backward Euler method differencing after, bring algebraic equation into Group g () simultaneous uses Newton iteration method to become with identical frequency solving state into a complete Groebner Basis Amount x and algebraic variable y, when iteration difference meets error requirements twice, iteration terminates.
A kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic of the invention, is considering system Multiple Time Scales characteristic on the premise of, can effectively process the significant active power distribution network dynamic simulation problem of stochastic behaviour.This Inventive method has taken into account that progressive failure calculating speed is fast, the advantage of implicit integration algorithm better numerical value stability simultaneously, has 0.5 rank Strong Convergence, 1 rank weak disparity, can efficiently accomplish many times while simulation accuracy is ensured with numerical stability The significant active power distribution network stochastic and dynamic artificial tasks of dimensional properties, particularly suitable for active containing various distributed power sources and load Power distribution network random walk is simulated and analytical calculation.The present invention can give full play to the number of progressive failure and implicit integration algorithm Value resolving advantage, while simulation accuracy and numerical stability is ensured, realizes consideration photovoltaic, wind generator system and comprehensively connects Active power distribution network stochastic and dynamic emulation in the case of entering.
Brief description of the drawings
Fig. 1 is the overall flow figure of the aobvious hidden mixed integrating method method that the present invention is suitable to the emulation of active power distribution network stochastic and dynamic；
Fig. 2 is mesolow active power distribution network example topological diagram；
Fig. 3 is the active power stochastic and dynamic list simulation track of photovoltaic generating system 1 to when partial enlarged drawing；
Fig. 4 be under different step-lengths L10 busbar voltage stochastic and dynamic list simulation tracks to when partial enlarged drawing；
Fig. 5 is the active power stochastic and dynamic simulation track of photovoltaic generating system 1；
Fig. 6 is the active power stochastic and dynamic simulation track of wind generator system 1；
Fig. 7 is L10 busbar voltage stochastic and dynamic simulation tracks.
Specific embodiment
A kind of the aobvious hidden of active power distribution network stochastic and dynamic emulation that be suitable to of the invention is mixed with reference to embodiment and accompanying drawing Integration method is closed to be described in detail.
A kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic proposed by the present invention, belongs to electric power System emulation field, is particularly well-suited to containing photovoltaic, the active power distribution network dynamic simulation of wind generator system.The inventive method is pin A kind of fixed step size stochastic and dynamic that stochastic behaviour and the significant active power distribution network dynamic simulation of Multiple Time Scales characteristic are proposed is imitated True method.Wherein, calculation procedure during solution wall scroll simulation track under one integration step of the inventive method is：First, at random Disturbance term model part carries out numerical value resolving using Euler-ball mountain method, and secondly, certainty part is asked using implicit simultaneous method Solution, wherein it is determined that the differential equation of property model carries out difference processing by backward Euler method, differencing equation joins with algebraic equation It is vertical, by Newton iteration method solving system variable.Be in the embodiment of the present invention using mesolow active power distribution network as test example.
As shown in figure 1, a kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic of the invention, bag Include following steps：
1) stochastic differential that active power distribution network topological connection relation, dynamic element parameter are set, characterize random perturbation variable Equation parameter and simulation calculation parameter, and time of origin, the type of failure or operation are set, wherein simulation calculation parameter includes Emulation termination time T, simulation step length Δ t, Monte-Carlo Simulation total degree I；
2) reading active power distribution network topological relation carries out Load flow calculation, obtains system load flow result of calculation；
3) according to system load flow result of calculation, simulation initialisation is carried out to the dynamic element in active power distribution network；
4) the initial number of times i=1 of Monte-Carlo Simulation is set；
5) this emulation initial time t=0 is set；
6) judge whether current emulation moment t is emulation initial time, if t=0, need setting active power distribution network to become at random Amount initial value simultaneously skips to step 8), otherwise into next step；
7) t is maden-1~tnPeriod calculates, and n is greater than the integer equal to 1, using progressive failure to constituting active distribution The active power distribution network Disturbance model of net stochastic dynamic model integrates 1 step-length and obtains tnThe system stochastic variable at moment is simultaneously Preserve, the stochastic differential equation in active power distribution network stochastic dynamic model is solved using Euler-ball mountain method；
Described active power distribution network Disturbance model is characterized by one group of stochastic differential equation：
In formula, η ∈ RKFor the vector that K n-dimensional random variable ns are constituted；α () and β () is scalar equation, represents random respectively The side-play amount and diffusing capacity of the differential equation；x∈RNFor the vector that N-dimensional state variable is constituted；y∈RMIt is M dimension algebraic variable compositions Vector；ξ is white Gaussian noise process；
1 step-length is integrated to Disturbance model using progressive failure, stochastic differential equation therein is solved Using Euler-ball mountain method, wherein, step-length is Δ t, and the time is from tn-1To tn, by ηn-1Obtain ηn, recurrence formula is as follows：
ηn=ηn-1+α(xn-1,yn-1,ηn-1)Δt+β(xn-1,yn-1,ηn-1)ξn-1(3)；
8) in tnAt the moment, random perturbation variable is transmitted, realize Disturbance model and active power distribution network deterministic models Between data transfer；
9) t is carried outn~tn+1Moment simulation calculation, wherein emulation moment t=t+ Δ t, Δ t is simulation step length, using implicit Integral algorithm integrates 1 step-length to the active power distribution network deterministic models for constituting active power distribution network stochastic dynamic model, obtains tn+1 Moment system state variables and algebraic variable are simultaneously preserved, and differential-algebraic equation group is asked using implicit simultaneous solution method Solution, using implicit backward Euler method to the differential equation in differential-algebraic equation group；
Described active power distribution network deterministic models are characterized by one group of differential-algebraic equation：
In formula, x ∈ RNFor the vector that N-dimensional state variable is constituted；y∈RMIt is the vector that M dimension algebraic variables are constituted, due to step It is rapid 8) to deliver active power distribution network random perturbation variable, therefore η is given value in equation；
Described use implicit integration algorithm integrates 1 to the deterministic models for constituting active power distribution network stochastic dynamic model Step-length, is solved using implicit simultaneous solution method to differential-algebraic equation group, using implicit backward Euler method to differential- Differential equation in Algebraic Equation set, specific solution procedure is as follows：
(1) 1 step is integrated using backward Euler method, step-length is Δ t, and the time is from tnTo tn+1, by xnObtain xn+1, recurrence formula For
(2) by differential equation f () in differential-algebraic equation through backward Euler method differencing after, bring algebraic equation into Group g () simultaneous uses Newton iteration method to become with identical frequency solving state into a complete Groebner Basis Amount x and algebraic variable y, when iteration difference meets error requirements twice, iteration terminates.
10) according to step 1) to judge whether system occurs in the case where moment t is emulated former for the simulated fault that sets and Action Events Barrier is operated, if not occurring, skips to step 12) continue executing with, otherwise into next step；
11) when breaking down or operating, setting system state variables is constant, according to active power distribution network deterministic models side Journey, recalculates system algebraic variable, by system variable result of calculation again assignment, return to step 7)；
12) judge whether emulation moment t reaches emulation end of a period moment T, if not up to emulation ends moment, return to step 7), otherwise into next step；
13) Monte-Carlo Simulation number of times i=i+1 is set, judges simulation times i whether more than Monte-Carlo Simulation total degree I, if being not more than, i.e. i≤I, then return to step 5), otherwise into next step；
14) I Monte-Carlo Simulation result is collected, simulation result is exported according to emulation demand and is drawn, terminated Artificial tasks.
Instantiation is given below：
This example is realized and is suitable to the aobvious of active power distribution network stochastic and dynamic emulation based on Matlab programming language environment Hidden mixed integrating method method, by containing photovoltaic, wind generator system mesolow active power distribution network example (see Fig. 2) to of the invention Method is verified and analyzed, and is compared with traditional explicit random simulation method simulation result.Using traditional explicit with During machine emulation mode, Disturbance model uses Euler-ball mountain method, and certainty part uses explicit over-over mode, wherein micro- Equation is divided to be solved by explicit improved Euler method.The hardware platform of emulation testing is Intel Xeon E5-2603 v3 CPU@ The 6 core PCs of 1.60GHz, 8GB RAM；Software environment is 64 operating systems of Windows 7.
Mesolow active power distribution network example is divided into two parts of middle-voltage network and secondary network, and system frequency is 50Hz. Wherein, middle-voltage network voltage class is 10kV, and main feeder is connected to external network by 10kV/110kV transformers；External network is adopted Simulated with the ideal voltage source that output voltage is 1.03p.u., secondary network voltage class is 0.4kV, main feeder passes through 0.4kV/ The transformer of 10kV is connected to middle-voltage network M5 nodes, interconnection switch S1, S2 closure.In addition, power distribution network uses three-phase symmetrical line Road and load, specific line parameter circuit value as in Fig. 2 mark shown in, and accessed three permanent magnet synchronous direct-drive wind force generating systems (in Pressure subnet in), two photovoltaic generating systems.The power electronic equipment control mode of each distributed power source, access capacity and active Power initial value refers to table 1.
In consideration light irradiance, in the case of the stochastic and dynamic characteristic of wind speed and load, the random perturbation of load fluctuation is special Property is simulated by Ornstein-Uhlenbeck processes, and each load active power initial value and stochastic behaviour parameter refer to table 2； The random fluctuation of light irradiance is by with amplitude coefficient swWiener process simulations, photovoltaic generating system stochastic behaviour parameter Refer to table 3；Wind speed is then modeled by the continuous Wind speed model of meter and Weibull distribution characters, and wind generator system is random Characterisitic parameter refers to table 4.
The change at random process of simulation load active power consumption is shown below：
In formula, η (t) is Ornstein-Uhlenbeck processes；PL0It is load active power initial value；QL0It is reactive load Power initial value；PLT () is t load active power, QLT () is t reactive load power.
Consider that the photovoltaic generating system of light irradiance S stochastic behaviours is shown below：
In formula, NsAnd NpRespectively series connection and parallel photovoltaic cell number；RsAnd RshIt is series connection and parallel resistance value (Ω)；Q is Electron charge constant (C)；K is Boltzmann constant (J/K)；A is diode characteristic fitting coefficient, is taken as 2；U is photovoltaic array Output voltage (V)；I and IsIt is photovoltaic array output current and diode saturation current (A)；T and TrefFor photovoltaic array work is exhausted To the operating temperature (K) of battery under temperature value and standard conditions；S and SrefIt is the light spoke under actual light irradiance and standard conditions Illumination (W/m2)；IphrefAnd IsrefIt is the photogenerated current under standard conditions light irradiance and diode saturation current (A)；CTIt is temperature Degree coefficient, (A/K) is provided by producer；EgIt is energy gap (eV), it is relevant with photovoltaic cell material；swIt is disturbance amplitude coefficient.
Consider that the wind generator system of wind speed v stochastic behaviours is shown below：
In formula, η (t) is the Ornstein-Uhlenbeck processes after simplifying；Φ [] is Gauss cumulative distribution function, can It is expressed as：Erf [] is Gauss error function (Gaussian error function), can It is expressed as：Weibull cumulative distribution function is represented, i.e.,
Simulating scenes set as follows：
Based on mesolow active power distribution network example, simulation time t=5s, simulation step length Δ t=2ms, M6 during 0.9s are set There is three-phase through low resistance grounding short trouble in bus, wherein short-circuit resistance is 2 Ω, 1.0s moment fault clearances.Random delta Discrete steps δ t be set to 2ms, t is consistent with discrete steps for simulation step length Δ.In power distribution network steady-state operation, pseudorandom is set Number function mutually in the same time, is generating identical stochastic variable in different simulation algorithms, now i.e. comparable different emulation are calculated The single-track simulation result in stochastic and dynamic emulation under method.
Stochastic and dynamic emulation single-track simulation result test：
From the figure 3, it may be seen that in the active power distribution network for considering stochastic behaviour, it is of the invention to show hidden mixed integrating method method with tradition The photovoltaic generating system active power single-track fit solution that explicit integral is obtained is good.Wherein, the relative error of track Less than 10-5.Illustrate that aobvious hidden mixed integrating method method of the invention can effectively process active power distribution network stochastic and dynamic simulation problems. In addition, deterministic simulation method is emulated compared to stochastic and dynamic, in system dynamic course, it is impossible to embody system variable operation During stochastic behaviour.
Further to compare a kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic of the invention With the simulation efficiency and numerical stability of traditional explicit integration method, respectively in Δ t=1ms, tri- kinds of Δ t=2ms, Δ t=5ms Under different simulation step lengths, different emulation modes are tested, wherein the emulating image of L10 busbar voltages is as shown in figure 4, by this Figure understands, of the invention to show hidden mixed integrating method method with traditional explicit integration method emulation rail in Δ t=1ms, Δ t=2ms Mark fitting is good, illustrates that the method disclosure satisfy that the numerical precision needed for emulating, but under the conventional step delta t=5ms of emulation, The inventive method can carry out simulation calculation, and traditional explicit integration method occurs numerical value unstability, it is impossible to complete artificial tasks. Additionally, the efficiency to different emulation modes is compared (be shown in Table 5), it is known that the inventive method and traditional explicit integration method phase Than the computational efficiency of emulation is improved.
Stochastic and dynamic emulation multi-trace simulation result test：
Through 400 Monte-Carlo Simulations, distributed power source active power output image is analyzed, simulation track such as Fig. 5, schemed Shown in 6.Because system occurs three-phase through low resistance grounding short trouble, during failure, each node voltage will occur wink When fall.Each distributed power source active power output will appear from vibration, then, be adjusted through control strategy, recover specified reference value. By image：1) due to the arbitrary excitation in distributed generation system running, and failure influence, distributed electrical Source is exerted oneself and obvious fluctuation occurs, after fault clearance, active output random wave can occur due to the influence of arbitrary excitation It is dynamic；2) 400 geometric locuses of Monte-Carlo Simulation are counted, obtaining its track average can be good with deterministic simulation track Fitting, this also illustrates that stochastic and dynamic emulation while system random process authenticity is embodied, is also maintained and deterministic simulation Identical accuracy.
Further, since the distributed power source, load in active power distribution network can cause network with the stochastic behaviour of time fluctuation The fluctuation of interior joint voltage, and the misoperation of protection device may be caused.One of wherein main performance is on high-tension side random spy The influence that property is brought to the safe and reliable operation of low-pressure side.Therefore only considering the stochastic behaviour of wind generator system, middle pressure load In the case of carry out fault test, choose the voltage out-of-limit situation before and after the L10 busbar voltage failures in secondary network and divided , there is voltage out-of-limit (U before and after the failure of statistical chart 7 in analysismin=0.95p.u.) random track number, as a result as shown in table 6.
When using deterministic simulation method, its simulation result does not exist voltage out-of-limit situation；And in stochastic and dynamic emulation During, the out-of-limit probability of L10 busbar voltages is 5.75%；As can be seen that the stochastic behaviour of medium voltage side distributed power source, load The node voltage fluctuation of low-pressure side can be directly resulted in, more there is certain probability to make low-voltage bus bar voltage out-of-limit, and may cause to protect The misoperation of protection unit, safe and reliable operation of the influence with electricity consumption side.It can be seen that using deterministic simulation method, its emulation knot The result of really more random emulation is too conservative, and the inventive method can solve the problem that the active power distribution network dynamic simulation for considering stochastic behaviour Problem, can more truly reflect the practical operation situation of system.
In sum, a kind of aobvious hidden mixed integrating method side for being suitable to the emulation of active power distribution network stochastic and dynamic proposed by the present invention Method, can efficiently accomplish stochastic behaviour and the significant active power distribution network dynamic simulation task of Multiple Time Scales characteristic, compared to true Qualitative dynamic simulation can the more truly actual motion state of simulation system.Additionally, the inventive method is integrated with traditional explicit Method is compared, and while numerical precision needed for meeting emulation, improves computation efficiency and numerical stability.
Each distributed power source control mode of table 1, access capacity and active initial value
The load active power random parameter of table 2
The photovoltaic generating system random parameter of table 3
Stochastic model parameter Initial light irradiance
Photovoltaic generating system 1 10
Photovoltaic generating system 2 10
The wind generator system random parameter of table 4
Stochastic model parameter k λ α β Initial wind speed
Wind generator system 1 1.77 10.67 0.0000265 0.0072801 9.08m/s
Wind generator system 2 2.16 9.65 0.0000345 0.0083066 9.37m/s
Wind generator system 3 1.89 10.21 0.0000298 0.0077201 10.46m/s
The different emulation mode efficiency comparisons of table 5
The voltage out-of-limit track number statistics of table 6
Emulation mode Voltage out-of-limit Simulation times Out-of-limit track number Accounting
Certainty dynamic simulation It is no 1 0 0%
Randomness dynamic simulation It is 400 23 5.75%

Claims (3)

1. it is a kind of to be suitable to the aobvious hidden mixed integrating method method that active power distribution network stochastic and dynamic is emulated, it is characterised in that including following step Suddenly：
1) stochastic differential equation that active power distribution network topological connection relation, dynamic element parameter are set, characterize random perturbation variable Parameter and simulation calculation parameter, and time of origin, the type of failure or operation are set, wherein simulation calculation parameter includes emulation Termination time T, simulation step length Δ t, Monte-Carlo Simulation total degree I；
2) reading active power distribution network topological relation carries out Load flow calculation, obtains system load flow result of calculation；
3) according to system load flow result of calculation, simulation initialisation is carried out to the dynamic element in active power distribution network；
4) the initial number of times i=1 of Monte-Carlo Simulation is set；
5) this emulation initial time t=0 is set；
6) judge whether current emulation moment t is emulation initial time, if t=0, at the beginning of needing to set active power distribution network stochastic variable It is worth and skips to step 8), otherwise into next step；
7) t is maden-1~tnPeriod calculate, using progressive failure to constitute active power distribution network stochastic dynamic model active distribution Net Disturbance model integrates 1 step-length and obtains tnThe system stochastic variable at moment is simultaneously preserved, using Euler-ball mountain method pair Stochastic differential equation in active power distribution network stochastic dynamic model is solved；
8) in tnAt the moment, random perturbation variable is transmitted, realized between Disturbance model and active power distribution network deterministic models Data transfer；
9) t is carried outn~tn+1Moment simulation calculation, wherein emulation moment t=t+ Δ t, Δ t is simulation step length, using implicit integration Algorithm integrates 1 step-length to the active power distribution network deterministic models for constituting active power distribution network stochastic dynamic model, obtains tn+1Moment System state variables and algebraic variable are simultaneously preserved, and differential-algebraic equation group is solved using implicit simultaneous solution method, are adopted With implicit backward Euler method to the differential equation in differential-algebraic equation group；
10) according to step 1) set simulated fault and Action Events judge system emulation moment t under whether break down or Operation, if not occurring, skips to step 12) continue executing with, otherwise into next step；
11) when breaking down or operating, setting system state variables is constant, according to active power distribution network deterministic models equation, System algebraic variable is recalculated, by system variable result of calculation again assignment, return to step 7)；
12) judge whether emulation moment t reaches emulation end of a period moment T, if not up to emulation ends moment, return to step 7), Otherwise enter next step；
13) whether Monte-Carlo Simulation number of times i=i+1 is set, simulation times i is judged more than Monte-Carlo Simulation total degree I, if No more than, i.e. i≤I, then return to step 5), otherwise into next step；
14) I Monte-Carlo Simulation result is collected, simulation result is exported according to emulation demand and is drawn, terminate emulation Task.
2. it is according to claim 1 it is a kind of be suitable to active power distribution network stochastic and dynamic emulation aobvious hidden mixed integrating method method, its Be characterised by, step 7) described in active power distribution network Disturbance model be to be characterized by one group of stochastic differential equation：
η · = α ( x , y , η ) + β ( x , y , η ) ξ
In formula, η ∈ RKFor the vector that K n-dimensional random variable ns are constituted；α () and β () is scalar equation, and stochastic differential is represented respectively The side-play amount and diffusing capacity of equation；x∈RNFor the vector that N-dimensional state variable is constituted；y∈RMFor M dimension algebraic variable constitute to Amount；ξ is white Gaussian noise process；
1 step-length is integrated to Disturbance model using progressive failure, stochastic differential equation therein is solved and is used Euler-ball mountain method, wherein, step-length is Δ t, and the time is from tn-1To tn, by ηn-1Obtain ηn, recurrence formula is as follows：
ηn=ηn-1+α(xn-1,yn-1,ηn-1)Δt+β(xn-1,yn-1,ηn-1)ξn-1。
3. it is according to claim 1 it is a kind of be suitable to active power distribution network stochastic and dynamic emulation aobvious hidden mixed integrating method method, its Be characterised by, step 9) described in active power distribution network deterministic models be to be characterized by one group of differential-algebraic equation：
x · = f ( x , y , η ) 0 = g ( x , y , η )
In formula, x ∈ RNFor the vector that N-dimensional state variable is constituted；y∈RMIt is the vector that M dimension algebraic variables are constituted, due to step 8) Active power distribution network random perturbation variable is delivered, therefore η is given value in equation；
Described use implicit integration algorithm integrates 1 step to the deterministic models for constituting active power distribution network stochastic dynamic model It is long, differential-algebraic equation group is solved using implicit simultaneous solution method, using implicit backward Euler method to differential-generation Differential equation in number equation group, specific solution procedure is as follows：
(1) 1 step is integrated using backward Euler method, step-length is Δ t, and the time is from tnTo tn+1, by xnObtain xn+1, recurrence formula is
(2) by differential equation f () in differential-algebraic equation through backward Euler method differencing after, bring Algebraic Equation set g into () simultaneous uses Newton iteration method with identical frequency solving state variable x into a complete Groebner Basis With algebraic variable y, when iteration difference meets error requirements twice, iteration terminates.
CN201710027392.4A 2017-01-13 2017-01-13 Explicit-implicit hybrid integration method suitable for random dynamic simulation of active power distribution network Active CN106855909B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710027392.4A CN106855909B (en) 2017-01-13 2017-01-13 Explicit-implicit hybrid integration method suitable for random dynamic simulation of active power distribution network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710027392.4A CN106855909B (en) 2017-01-13 2017-01-13 Explicit-implicit hybrid integration method suitable for random dynamic simulation of active power distribution network

Publications (2)

Publication Number Publication Date
CN106855909A true CN106855909A (en) 2017-06-16
CN106855909B CN106855909B (en) 2020-03-20

Family

ID=59127208

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710027392.4A Active CN106855909B (en) 2017-01-13 2017-01-13 Explicit-implicit hybrid integration method suitable for random dynamic simulation of active power distribution network

Country Status (1)

Country Link
CN (1) CN106855909B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109886523A (en) * 2018-12-25 2019-06-14 清华大学 A kind of comprehensive energy net dynamic model multi tate calculation method
WO2023005700A1 (en) * 2021-07-27 2023-02-02 天津大学 Model-data-hybrid-driven electrical grid reliability rapid calculation method and device

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102545263A (en) * 2012-01-19 2012-07-04 浙江大学 Power system transient stability simulation method based on explicit numerical integration
CN102609575A (en) * 2012-01-19 2012-07-25 浙江大学 Power system transient stability simulating method based on implicit numerical integration
CN103440374A (en) * 2013-08-23 2013-12-11 中国南方电网有限责任公司电网技术研究中心 Electromagnetic transient simulation modeling method based on state variable nodal analysis combination frame
CN103793562A (en) * 2014-01-05 2014-05-14 天津大学 Active power distribution network transient state real-time simulation system designing method based on FPGA
CN104751246A (en) * 2015-04-09 2015-07-01 国网宁夏电力公司经济技术研究院 Active distribution network planning method based on stochastic chance constraint
CN105808825A (en) * 2016-02-29 2016-07-27 天津大学 Rolling type stochastic projection integration method suitable for dynamic simulation of active power distribution networks

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102545263A (en) * 2012-01-19 2012-07-04 浙江大学 Power system transient stability simulation method based on explicit numerical integration
CN102609575A (en) * 2012-01-19 2012-07-25 浙江大学 Power system transient stability simulating method based on implicit numerical integration
CN103440374A (en) * 2013-08-23 2013-12-11 中国南方电网有限责任公司电网技术研究中心 Electromagnetic transient simulation modeling method based on state variable nodal analysis combination frame
CN103793562A (en) * 2014-01-05 2014-05-14 天津大学 Active power distribution network transient state real-time simulation system designing method based on FPGA
CN104751246A (en) * 2015-04-09 2015-07-01 国网宁夏电力公司经济技术研究院 Active distribution network planning method based on stochastic chance constraint
CN105808825A (en) * 2016-02-29 2016-07-27 天津大学 Rolling type stochastic projection integration method suitable for dynamic simulation of active power distribution networks

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
SEKINE T, ASAI H: "Unconditionally stable explicit method for the fast 3-D simulation of on-chip power distribution network with through silicon via", 《ELECTRONIC COMPONENTS & TECHNOLOGY CONFERENCE》 *
孙充勃, 李鹏, 王成山,等: "含多直流环节的混合结构有源配电网潮流计算方法", 《电力系统自动化》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109886523A (en) * 2018-12-25 2019-06-14 清华大学 A kind of comprehensive energy net dynamic model multi tate calculation method
CN109886523B (en) * 2018-12-25 2021-03-02 清华大学 Multi-rate calculation method for dynamic model of comprehensive energy network
WO2023005700A1 (en) * 2021-07-27 2023-02-02 天津大学 Model-data-hybrid-driven electrical grid reliability rapid calculation method and device

Also Published As

Publication number Publication date
CN106855909B (en) 2020-03-20

Similar Documents

Publication Publication Date Title
Huang et al. Integrated transmission and distribution system power flow and dynamic simulation using mixed three-sequence/three-phase modeling
CN102682358B (en) A kind of assessment wind-electricity integration scale and Net Frame of Electric Network adaptive planning simulation method
Joo et al. Coherency and aggregation techniques incorporating rotor and voltage dynamics
Gupta Probabilistic load flow with detailed wind generator models considering correlated wind generation and correlated loads
CN103700036B (en) A kind of transient stability projecting integral method being suitable to power system Multiple Time Scales
CN107968409A (en) A kind of probability load flow calculation method and system for considering imbalance power distribution
Sun et al. A co-simulation platform for smart grid considering interaction between information and power systems
CN110034581A (en) The electrical betweenness vulnerability assessment method in the section of electric system under wind-electricity integration
CN104992017A (en) Wind power random fluctuation based generalized load modeling method and apparatus
CN104156542A (en) Implicit-projection-based method for simulating stability of active power distribution system
Cepeda et al. Bulk power system availability assessment with multiple wind power plants.
CN106855909A (en) A kind of aobvious hidden mixed integrating method method for being suitable to the emulation of active power distribution network stochastic and dynamic
CN105808825B (en) A kind of roller accidental projection integration method suitable for active power distribution network dynamic simulation
Sufyan et al. Implementation of PMU-based distributed wide area monitoring in smart grid
Chen et al. Power system on-line transient stability prediction by margin indices and random forests
CN104158190B (en) Electric network swim safe prediction device and method
Bai et al. Event-triggered forecasting-aided state estimation for active distribution system with distributed generations
Bagchi et al. Studying the impacts of incorporating energy storage devices into an aggregated probabilistic model of a virtual power plant
Bian et al. Identification and improvement of probabilistic voltage instability modes of power system with wind power integration
Cokkinides et al. Visualization and characterization of stability swings via GPS-synchronized data
Guo et al. Online model-free estimation of the dynamic system model for a power system with renewables in ambient conditions
Li et al. Two-point estimate method for probabilistic optimal power flow computation including wind farms with correlated parameters
Narayan et al. Impact of wind correlation and load correlation on probabilistic load flow of radial distribution systems
Liu et al. Impact of electric vehicles on microgrids considering multiple correlations
CN104699950A (en) Method for calculating stochastic load flow with correlation problem of stochastic power units

Legal Events

Date Code Title Description
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