Power Systems Modelling

Power Systems Modelling

  • Bassel Hassan asked a question:
    How do I choose a "configuration parameter of the powergui" to simulate(simulink) a model of distribution network consist of PV and residential load ?


    I am simulating (by simulink) a distribution network feeder which consists of PV generation site and load. my simulation is too long at time domain so it requires a phasor simulation type (in the configuration parameter of the powergui). However, some of the elements in the model require a discrete simulation type. Running the simulation for the feeder model using a phasor simulation type gave me an error.

    May I know how do I go about solving this problem? Any help or advice would really be greatly appreciated!

  • Gautham Ashokkumar asked a question:
    How can I scale a voltage regulator in the original IEEE 34 bus system to use in a scaled version of the IEEE 34 test system in PSCAD?

    The IEEE 34 distribution test feeder data is publicly available; the PSCAD model for the original IEEE 34 test system is also found online.  

    However, we have scaled our system to a 12.47 kV system instead of the original 24.9 kV system.  How should the voltage regulator of the 24.9 kV original model be modified accordingly? 

    A picture of the voltage regulator block and its control block is attached; the original model in PSCAD is attached; I can send you any more info you need

    + 2 more attachments

  • U. Dreher added an answer:
    In PSCAD, how can I take the value of phase angle in radians from the output of a PLL block and shift it by 90 degrees (or pi/2 radians) leading?

    I have the output in radians, and just need a block or a logic to shift that phase angle output by 90 degrees leading to use it in another unit.

    U. Dreher

    You should be able to place an adder node - having the angle of your choice at the 2nd input.


  • Srete Nikolovski added an answer:
    In Digsilent if the power factor is used to determine MVAR and use in the software. Will it converge in its load flow calculation ?

    I have a network that is failing to converge due to reactive limits . I used the power triangle calculation to determine the MVAR after having the power factor. 

    For example, 10MVA  generator with 0.9 pf means the maximum power is 9MW and 4.35 MVAR. If I use this calculation then the reactive power will be very high for the network. 

    When I check other models that don't have convergence problem , the total reactive power is very small about a third of the total reactive power. What are the limits that can be used for the reactive power to be used for calculation if there is no data for the machines? 

    Srete Nikolovski

    In DIgSILENT Powerfactory software you have option for generator modeling in whcih you can set min and max reactive power limits of capacitive and inductive production of generator. Generator can work  to maintan the voltage on its local bus or work in P, cos phi mode to maintain power factor and deliver active power P.

    You can also add capability cure of generator if you have it from producer.

    So the amount of reactive power, most be inductive and depends of loads and lines and transformer parameters. So check line lenghts and uk % of transformers and ractive power of loads.. You also need to have another bus with Utility called slack bus and set it as SL node.

  • Mansour Moradi added an answer:
    Microgrid test system in Digsilent or matlab simpower?
    I want work on my project in the field of "Microgrid." I need a test system of microgrid in digsilent or matlab simulink. If anyone here has this file, please give me.
    Thanks guys!
    Mansour Moradi

    Hi, Please see: simpowersystems examples in Matlab R2015.
    for to open attached file, use of MATLAB R2015.

  • Kishor Babu Gunti added an answer:
    What is wrong with my current waveform?


    This question pertains to Matlab simulation. I will be grateful in case anyone knows the answer.

    Attached .docx file is a Simulink circuit of a inverter simulation. In order to implement close loop control of inverter, I first make it work as rectifier and trying to do its PLL and current control loop implementation. In this quest, I am applying a voltage through a grid and modify it to another level by transformer. I am not able to figure out that after putting any Kp and Ki values, why does the input current waveform shape is not symmetrical? Can any one help me to find the problem? In another file, I have attached Voltage (Upper) and current (Down) waveform taken from grid. The corresponding scope is marked in .docx file.



    + 1 more attachment

    Kishor Babu Gunti

    you can apply a voltage source from grid and pass through zero-crossing detector

  • Rajanivedha Ramakrishnan added an answer:
    Explanation on the model of REGC-A and REEC-B model in the power world simulator?

    can some one help me lending hands to under the blocks which means the operation of the control system model of generators and exciters. I am in need of urgent help

    + 1 more attachment

    Rajanivedha Ramakrishnan

    I need to know the operation of the block diagram of these blocks and how it is helpful in integrating it to a pv system

  • Fatemeh Jafari Naeeni added an answer:
    In this case, what advantage have miniature keys and phase and null disconnecting rather than simple phase for LVAC circuits in the substation panel?

    If in upper distributed substation and transmission, inner consumption feeds from Aux transformers while it’s located nearby power transformer, we can feel that the transformer is under of more tensions and having much short current,

    LVAC circuits in the substation panel

    What are the other solutions in this case?

    Fatemeh Jafari Naeeni

    maybe i can help you but need more explanation what you want exactly!

  • Fatemeh Jafari Naeeni added an answer:
    I am trying to catch the electrolyser, storage an compressor cost even from quote but for a bit large case and also if i move from size to bigger ?

    I am trying to catch the electrolyser, storage and compressor cost even from quote but for a bit large case and also if I move from size to bigger  size how I can scale the cost based on the big size?

    Fatemeh Jafari Naeeni

    maybe i can help you but need more explanation what you want exactly!

  • Olivier Barre added an answer:
    What type of modification can be made using simulations in FEMM software under the title of calculation of air gap?

    I am working on optimization of transformer parameters and need expert's advice on this one specifically please guide me.My team has introductory level experience on this FEMM tool.We  need to present simulations for any type of transformer modification.

    Olivier Barre

    Dear Pranav Vyas,,

    FEMM allows scripting which are able to modify or automate simulation. I have sent you two types of files. There is nothing to do if you want to observe FEMM with scripting. Open FEMM and go to "file", in opening such box, you will see "Open LUA script". Go to this item and a dialog box appears. Load the file Position_2.lua and let FEMM doing the simulation. In FEMM, The LUA console must be open. You will see in this box, the calculated torque when the motor is rotating. With such a simulation, the cogging torque can be calculated and visualised.
    This file is just an example which show you one method in using scripting.

    If you have a FEMM file which is a simulation of a transformer, I can do a LUA script file which is able to modify an air-gap and do some calculations which are linked to the thickness of the air-gap.

  • Birender Singh added an answer:
    Does anyone have a Matlab/PSAT algorithm to identify buses for reactive power injection to improve voltages in unbalanced radial distribution system?

    I have an IEEE 34 bus distribution system, with unbalanced loads and PV injection at all load buses. Please guide me on identifying buses to improve voltage profile. 

    Birender Singh

    HI Gautham

    Especially when solving unbalanced systems with three-phase representations, heuristic algorithms are widely used in reconfiguration studies because of their fast computational speed and good accuracy, branch-exchange method is a classical heuristic method that is widely used in reconfiguration study.

    Am attachcing a figurein which the components are to be attached in case of unbalanced load.

  • Amlan Chakrabarti added an answer:
    What is the range of sampling frequency of transmission line protecion in VSC-HVDC system in practice?

    Dose sampling frequency of  100 kHz achievable in  pratical engineering?

    Amlan Chakrabarti

    The sampling frequency should be greater than twice of nominal frequency. In HVDC transmission line, the nominal frequency is zero. So, a sampling frequency of 10 Hz with a time period of 100 milliseconds may serve the purpose. This will provide 10 samples per second.

  • Yasaman Majd added an answer:
    Does anybody know what is the effect of increasing the stages of ring oscillator on power consumption and frequency?

    does anybody know what is the effect of increasing the stages of ring oscillator on power consumption and frequency?if we increase the stages power must be increased or not?and also about frequency

    please do me a favor and help me  

    Yasaman Majd

    yes I found it,Thanks

  • Ganesh Prajapat added an answer:
    Could anybody share the data of 5MW DFIG generator?

    I am searching [pu] data of a 5 MW DFIG generator for my simulation which i am going to perform on NREL 5MW baseline wind power plant.

    However, i have the pu data of a 1.5MW DFIG generator and the simulation is running well with some controllers.

    plz provide the data sheet if anybody have.

    Thanks and regards,


    Ganesh Prajapat

    hi Mohamed A. Elgenedy sir, i am using matlab script for the simulation and that is why i need the data. however, i have some of the papers with some useful data but some data are missing in it.

  • Sill Fsxa added an answer:
    Is it possible to model the line outage in MATPOWER ?

    Specially when the network will be separated islands after line outage (radial distribution networks) . 

    Sill Fsxa


    I think it's a little late, but I hope it could be useful for someone.

    For running PF in a power system with line outages, if islanding phenomena happens, you could find the related buses using find_islands(YOUR_MPC) in matpower. it'll give you isolated buses and you could change their TYPE to isolated (YOUR_MPC.bus(isolated buses,BUS_TYPE) = 4) so that the Power flow solution could converge.

  • Sanjoy Roy added an answer:
    How to model a battery in GAMS where power output is being curtailed from a wind farm?

    Hi all I'm looking to model power flow through a large battery in GAMS where the battery can either charge, discharge or store the energy given parameters of P = 1MW, E = 4MWh. I have a table of power output and corresponding SMP price of electricity for each time slot in order to maximize profit and minimize wind curtailment. 

    I currently am using the following constraints: 

    Constraint1(t) .. MaxFlow(t) =l= Charge(t,'Power') + Discharge(t);
    *Setting constraint that battery can either charge or discharge
    Constraint2(t) .. Charge(t,'Power') =l= StoragePower;
    *Power flow in to the battery will be less than the Power Capacity of the battery
    Constraint3(t) .. Discharge(t) =l= StoragePower;
    *Power flow out of the battery will be less than the Power Capacity of the battery
    Constraint4(t) .. StorageLevel(t) =e= Charge(t,'Power') *( 0.1667 * eff) - Discharge(t) * (0.1667 * eff);
    *Energy in the battery is equal to the energy in - energy out
    Constraint5(t) .. StorageLevel(t) =l= StorageEnergy;
    *Energy in battery cannot exceed the energy capacity
    Constraint6(t) .. Used(t) =e= Discharge(t) * 0.1667 * eff;
    *The energy used is what is discharged * time * efficiency
    Constraint7(t) .. HourlyRevenue(t) =e= Used(t) * Charge(t,'Price');
    *hourlyrevenue is equal to the price of electricity * energy

    ObjectiveFunction .. TotalRevenue =e= sum(t,HourlyRevenue(t));

    When I run this code the answer returned is stating it's infeasible. I'm currently unsure if I have too many or too little contraints for the problem to be solved. Any help/advice would be greatly appreciated.

    Kind regards,


    Sanjoy Roy

    Hi Ciaran,

    With the two inputs there, I don't think the constraints are too many for solution by any stretch of imagination.

    Whether they are too less or not depends on whether the solver can find a solution within the defined hyperspace. Just try solving the model once you think it is OK ! If GAMS exists saying Problem Unbounded, then that straight away tells you that more constraints are required.

    This is the situation where the optimisation hyperspace has not been properly closed by constraints, one side of the hyperspace is open, and therefore a solution for the model can not be found.

    Hope that helped.


  • Shrikant Mali added an answer:
    The PV inverter in the system is modeled as current source tied to the utility grid. How to control reactive power for maintaining grid voltage?

    The system has a current source as the PV inverter output. The aim is to inject reactive power from the PV inverter to maintain grid voltage within 95% to 105%. Most of the DQ control inverter techniques have voltage as output. But for my model, I need current (Ipvac, see diagram) as output, not voltage. Please refer the diagram attached for more clarity. VLL in the diagram is the voltage I must control.

    I already have a look-up table and algorithm to generate the Iq reference (reactive current), but don't know how to use it and achieve VAR control with final current output. If I suppy VARs from the PV inverter, the grid current drawn will reduce, and hence voltage drop will reduce.

    Shrikant Mali

    Please find below answers to above questions:

    1) Yes, Standard practice is to use RMS values of voltages. And no need to use arcsin function to get theta.

    2) Yes, Irefac *Sin(theta) will do (use modulator block).

    3) Yes, It is right.

    For freq, If you are simulating for frequency variation (i.e. active power supply is not equal to active power demand) then you should measure frequency using actual grid voltage. And if you are not interested in this type of study then providing fixed 60Hz value will work.

  • Debasish Mondal added an answer:
    Any advice on optimal location of facts and DG?

    if i have solve optimal power flow problem for following function
    1. fuel cost
    2. active power loss minimization
    3. reactive power loss
    4. voltage deviation minimization
    5.Lmax improvement

    how to find out where optimal location of DG-TCSC-SVC-UPFC-IPFC-STATCOM-DSTATCOM

    and which case improvement when find its optimal location

    Debasish Mondal

    You may follow this Article you will get about optimal location of FACTS device.

    D. Mondal, A. Chakrabarti and A. Sengupta, “Optimal placement and parameter setting of SVC and TCSC using PSO to mitigate small signal stability problem,” International Journal of Electric Power and Energy System, vol. 42, no. 1, pp. 334-340, April 2012,

  • Narasim Ramesh added an answer:
    I have reference Iq*, reference Id*, the Iq and Id from performing the DQ transform. How can I get the Ipeak current back, to use as current source?

    I have read papers and some books on DQ transform as this concept is new to me, to perform control of the reactive power that I'm injecting (by generating a value of Iq* reference from a look up table).

    Most of the papers have switching models explaining the use of Vq and Vd components to control switching and regulate voltage.

    But in my case, I have an averaged model, (hence no switches in model) with a current source output. How can I get output current from comparing Iq*reference, Iq and Id* reference and Id?

    Narasim Ramesh

    1. The d-q transform( after phase transformation say 3phase to two phase can be visualized as a conversion of phase quantities to synchronous  rotating reference frame (see e.g. Krause & Lipo  et al) say  rotating at  we .

    2.So reference IDref and IQref  is obtained from say Iaref sin(wt) and Ibref cos(wt) by setting w = we. and init angle=0 . Hopefully we know freq w.

    3 Now given id and iq  in dq frame we can obtain process wrt IDref and IQref in dq frame and use  inverted transformation matrix to obtain instantaneous phase values.

    Good Luck.


  • Anand Mp added an answer:
    Real measure high resolution solar data (15 mins - 30 mins) for Canada?

    Hello Researchers, 

    I need high resolution say, a measurement interval of 15 mins - 30 mins solar radiation data for my research. Most of the hourly data archives have average hourly solar radiation which don't accommodate the variation in horizontal direct beam radiation (looks like a smooth deterministic curve). I need to see the deviations in radiation due to cloud and other obstructions if any.    

    Please comments your suggestions. Thanks in advance!

    Anand Mp

     Hello Hasan,

    Thanks a lot. This is wat I was looking for. As my ongoing research is case specific (Manitoba system) I may have to get a similar data for Manitoba. Anyway, these set of data will serve the purpose now.

    Thank you

  • Yehia F. Khalil added an answer:
    If we have electricity power with high penetration of renewable energy and i wnt to use the excess power to produce hydrogen ?

    if we have electricity power with high penetration of renewable energy and i want to use the excess power to produce hydrogen. can anyone help me with new tariff for excess power since we want to encourage all electrolyser to produce hydrogen at off- peck periods?  

    Yehia F. Khalil

    If you have excess electricity from a renewable resource solar PV or wind turbines and you want to produce hydrogen gas, then you can direct this excess electric power to a water electrolyzer (some of some work at high pressure) where water is electrolyzed into hydrogen gas and oxygen gas. You can then the hydrogen gas for future use as a feedstock to operate a PEM fuel cell and produce electricity on demand.

    Hope this helps answer your question!

  • Sergio Honwana added an answer:
    Who can help to find IEEE test system of Radial low-voltage unbalanced distribution system as detailed model (R & X of lines & loads)?

    I need to valid test system of Radial low-voltage unbalanced distribution system. Each one of the Lines & loads in this system is an impedance (as R & X ).In this test system, the real & reactive power can be defined or not.The important cases at the required test system are:

    1-Radial Distribution system

    2-Low-voltage(220-240 V)

    3-unbalanced in load side

    4-Impedance values of lines & loads

    Sergio Honwana


    I presume that this is typical RDN(Radia Distribution Network) at  the secondary distribution feeders supply residential and commercial customers at 120/240/440 volts.

    In general, the distribution system is the electrical system between the substation fed by the transmission system and the consumers’ premises. As for  f the current case the voltage level  is set for 220-240 V, it is require to maintain requirements for such type distribution in order to guarantee consumers power rate requirement mainly the proper voltage, loading  and reliability issues.

    RDN unbalanced are prone for Dynamic change in imposed load, High R/X ratio, and contain also extremely large number of nodes and branches.

    You might find Ladder network methods implementation in some section of  the system in order to finds suitability and practical application for single source radial structure networks with high R/X ratio;

    You may also use  SVC or Series compensation at substation fed in order to regulate voltage and  Reactive power correction.

    i suggest to follow   the below link as well...


  • Anil BHARADWAJ Chivukula added an answer:
    How can I develop a grid model in PSCAD with low SCR (short circuit ratio) ?

    I am working on modeling of weak grid (with low short circuit ratio) in PSCAD, can anyone help to find a suitable reference to develop that model.

    Anil BHARADWAJ Chivukula

    Thank you Kailash Sir,

  • Saurav Raj added an answer:
    Need modified IEEE 30 bus system data?

    In modified IEEE 30 bus system data active power loss is 5.660 MW under base load condition without using optimization. Please provide me data.

    Saurav Raj

    Thank you sir

  • Srete Nikolovski added an answer:
    Thermal rating of cables, temperature computation around the media?

    Dear All

    Does anyone know and can sent me a link for free software which can be used for thermal analysis of cable - ampacity and temperature computation around the cables in different environment (ground, water, air, concrete pipe.) I know that there are many commercial software from CDEGS, EATON-CYME..... but I need easy free software to eliminate manual computation.

    Thanks in advance

    Srete Nikolovski

    Super, valja ga proučiti !!

    U svakom slučaju veliko hvala.


  • Neil Ruedlinger added an answer:
    If I know one single device performance along time, how can I predict million of devices performance along time?

    for example, if I know current vs time curve for one single device; now I have one million device deliver current, each behaves independently, what theory or equation or simulation can I use to predict the million device behavior(overall current) along time?



    Neil Ruedlinger

    Hi Tao,

    I did not intend to convey a negative response; my response is a qualified maybe.

    I doubt, given enough dedication, time and resources that anything is impossible; we Engineers have and will continue to accomplish the seemingly impossible because I think resolving the 'impossible' is part of any Engineer's job description.

    There are other videos on YouTube where investigators have delivered the interim results of their ongoing research projects on applying Fractal and Chaos Mathematics to artificial Neural Networks, and (I humbly assume) to biological Neural Networks to illuminate the path towards the ultimate understanding of how the human brain actually functions.

    Complex systems, such as an ant/bee/termite colony or Neural Networks have been discovered, through Fractal and Chaos Mathematics studies, to possess a collective property known as 'emergent behavior', in which whilst the individual insects or neurons by themselves may not seem to be very smart, as a group they are able to solve problems that are intractable using conventional analytical Mathematical tools based on linear algorithms. One classic case is the Travelling Salesman Problem, described by Dr. Robert Sapolsky of Stanford University.

    In one of the YouTube links posted in a previous post, Dr. Sapolsky discusses how applying the complex system dynamics of a simulated ant colony, produces remarkably elegant emergent behavior, that very efficiently solves the Travelling Salesman Problem; a problem that has baffled Mathematicians and Computer Scientists for decades, prior to the application of Fractal and Chaos Mathematics, and the related field of cellular automata simulation.

    I have a suspicion that using these approaches to the study of an electrical network with a million current sources, may yield some interesting results. I also suspect that in addition to using Fractal, Cellular Automata and Chaos Mathematics, the Monte Carlo method, suggested by Odair would be very valuable to approximate real world variations in the parameters of each current source, including the simulation of different fault conditions that would have an impact on the emergent behavior of the electrical network simulation.

    The above paragraph may suggest this is a huge project; one discovery that may help to alleviate the challenges in constructing a simulation comes from experiments with natural complex systems; these systems will always bifurcate, i.e. they will always evolve along one of two paths; the path of evolution the complex system takes, depends on the initial conditions applied to the complex system. For electrical current sources, these initial conditions may be introduced through individual source variations in the wave mechanics (frequency, amplitude, phase), harmonics or impedance parameters.

  • Samreen Sid added an answer:
    To input measured variable wind speeds for mathematical modelling in simulink?
    How to input variable wind velocity to the simulink model correspondingly to get the output for variable speed, using the lookup table in simulink is giving an error. I want to give wind speed measured for every 10mins and to dump/interface those in simulink model.
    Samreen Sid

    Did you guys find a solution for that? I need to input variable wind speed to my simulink model.

  • Ankit Singh added an answer:
    How can we sense the input current in bridgeless SEPIC converter in CCM mode for hardware implementation ?

    How can we sense the input current in bridgeless SEPIC converter in CCM mode for hardware implementation ?

    Ankit Singh
    Thank You Sir...
  • Kandalam Giridhar added an answer:
    How to reduce the amount of force applied on wind generator rotor?

    I build 3-phase 4-wire wind generator; At no-load it is rotating freely using hand,but when i apply a small load(app 10w 1-phase load) it is not rotating freely,it acquiring huge amount of energy so,is there any possiblity to isolate the load and alternator,or is there any chances to rotate freely.

    Kandalam Giridhar

    thanks sanjay,i will check it.

  • Adrián Ramírez added an answer:
    How to model a 220kV, 42km long transmission line?

    For Voltage Drop and power loss calculations.Can it be modelled as 'T' or 'Pie'?

    The Total Inductance of the Line is measured as 0.04H and Capacitance as 0.4 microFarad.

About Power Systems Modelling

Simulation of Power Systems as a whole and of each separate component

Topic followers (983) See all