On a new approach for the simulation of transients

in power systems
J. Mahseredjian1 S. Dennetière2 L. Dubé3 B. Khodabakhchian4 L. Gérin-Lajoie5

Abstract—This paper presents a new simulation tool named infinity when opened. This is a fundamental principle in such
EMTP-RV. EMTP-RV is a completely new program with a new a model to avoid superfluous natural frequencies and matrix
graphical user interface and a new computational engine. The
simulation uses a new matrix formulation for computing load- conditioning problems. That is why system (1) is a variable
flow, steady-state and time-domain solutions. Theoretical rank system where closing a switch eliminates a matrix
advantages are emphasized and demonstrated through practical column and the corresponding row. The disadvantage of such
examples. An open-architecture graphical user interface (GUI) is
manipulations is the computational effort, especially when the
developed to maximize flexibility and allow creating and
maintaining complex designs. number of switches and switching frequency become high.
The EMTP-V3 code was written using the legacy Fortran-
Keywords: simulation tools, numerical methods, EMTP 77 language and related methods. Such codes have significant
limitations in automatic memory management, in code
I. INTRODUCTION organization and interfacing with external programs. Modern
Since its initial concept presented in 1969 the basic EMTP computing languages and methods provide much more
type simulation approach remained unchanged. It is used in powerful and efficient environments for software development
various commercial (DCG-EMTP Version 3, EMTP-V3 [2]) while providing crucial advantages in code evolution and
and non-commercial packages. The main system of symmetric maintenance.
equations used in [1] is given by: Another major aspect in the simulation of transients is the
Yv = i (1) graphical user interface (GUI). It constitutes the first and the
This is referred to in the literature as the standard nodal main facet directly apparent to the user. It also represents a
analysis formulation. It assumes that all network components technological barrier into the design and simulation of more
can be given an admittance matrix model. Matrix Y is the complex cases. Although several GUI programming
admittance matrix, v is the vector of unknown voltages and i approaches are presented in the literature [4][5], the
is the vector of current sources combined with history current innovation path in this field is towards increased level of
sources for the trapezoidal integration method. Since v also automation, open-architecture and advanced visualization.
This paper presents a new simulation tool named EMTP-
includes known voltages sources, equation (1) is actually
RV with a new graphical user interface EMTPWorks. It
implemented taking only a submatrix Yn of Y for finding n
supersedes EMTP-V3 and presents several improvements in
unknown voltages. One of the important disadvantages of this computational engine, graphical user interface and software
formulation is the inability to incorporate ungrounded voltage development aspects. Although there are commercial aspects
sources. This has been corrected in [3] by using modified- in any industrial grade software development, this
nodal analysis, which incorporates an extra row for the presentation is kept at the technical contribution level.
voltage source equation and an extra column for the voltage EMTP-RV has been written from scratch without any
source current which becomes listed in the unknown’s vector. reverse-engineering or code recuperation from old EMTP-V3.
The assumption of admittance model existence for every It constitutes a historical moment in the complex evolution of
component is a significant limitation. An ideal transformer EMTP type software.
model, for example, does not have admittance matrix
formulation. This can be partially avoided by adding losses, II. NEW FORMULATION
but can then cause ill-conditioning problems. Other limitations
The development of EMTP-RV is based on the conception
are found in the representation of ideal switches. An ideal
of a computational core for the solution of its system of
switch has pure zero resistance when closed and becomes
equations. The main architecture is shown in Fig. 1. The core
has a solution step controller (Organizer) which is responsible
(1) École Polytechnique de Montréal, Canada for all data input and communication tasks. It schedules
jean.mahseredjian@polymtl.ca
(2) Électricité de France, Clamart, France various task calls to the simulated network devices (models
(3) DEI Technology, Montréal, Canada and options). These are request signals received by device
(4) TransÉnergie Technologies, Hydro-Québec, Montréal, Canada codes organized into encapsulated modules with device
(5) Hydro-Québec, Montréal, Canada
objects holding data and methods. Devices may respond back
Presented at the International Conference on Power Systems by participating into various requests. Participation is optional
Transients (IPST’05) in Montreal, Canada on June 19-23, 2005 since devices may have different equations and needs.
Paper No. IPST05 - 139
 system above provides further generalization and introduces
Data new non-zero submatrix possibilities.
A. Nonlinear functions
Core Code
It can be proven that if nonlinear device equations are
request
Organizer Device models linearized at each solution time-point, matrix A becomes the
participation data, methods Jacobian matrix of the Newton solution. This approach
provides a true-nonlinear solution to all nonlinear devices. It
Solver
eliminates the less precise pseudo-nonlinear devices and all
topological restrictions (see [7]) encountered in the
Scopes, external interaction compensation method used in [2]. Machine models are also
Fig. 1 EMTP-RV main architecture classified as nonlinear functions and solved through the same
iterative solver for obtaining a simultaneous solution with all
The key aspect in this architecture is the ability to easily
nonlinear devices and in any topological configuration.
accommodate device equations into the main system of
Further details on this approach will become available in a
network equations. A new system of modified-augmented-
separate paper.
nodal analysis is proposed to eliminate various classical nodal
or modified-nodal analysis limitations: B. Control system equations
⎡ Yn Vc Dc Sc ⎤ ⎡ v n ⎤ ⎡ i n ⎤ Improvements in the solution of control system equations
⎢V ⎥⎢ ⎥ ⎢ ⎥ are presented in a separate paper [8]. Numerical delays that
⎢ r Vd DVD SVS ⎥ ⎢ iV ⎥ = ⎢ v b ⎥ (2)
⎢ Dr D DV Dd S DS ⎥ ⎢ i D ⎥ ⎢db ⎥ existed in [2] between control blocks due to nonlinear
⎢ ⎥⎢ ⎥ ⎢ ⎥ feedback loops, are eliminated using a linearization technique.
⎣ S r S SV S SD S d ⎦ ⎣ i S ⎦ ⎣ sb ⎦
This is an augmented formulation which is keeping only the C. Steady-state solution
Yn part from equation (1). Although submatrices are The main purpose of a steady-state solution in EMTP is to
identified due to their typical contents and for symbolic initialize the network state variables for minimizing the
explanations given below, equation (2) can be viewed as a natural response at startup in time-domain. The formulation
generic Ax = b system. Matrix A is not necessarily method is the same as in equation (2), only now all variables
become complex and devices must provide steady-state
symmetric which provides another advantage over equation
equivalents for each frequency in the generic case of harmonic
(1) for some device models.
steady-state.
For a voltage source connected between two nodes k and
m , the source equation is given by: vk − vm = vb km . It is D. Load-Flow solution
directly inserted into the main system by placing a 1 and a -1 The steady-state solution can provide acceptable operating
in columns k and m respectively of Vr . The source voltage conditions (power-flow) when initialized from a load-flow
is vb km . The source current condition is accounted by solution. The augmented matrix concept is expanded to
include load-flow constraint equations and the solved
transposition in the submatrix Vc . If the source is nonlinear function becomes:
disconnected, then only its diagonal cell in Vd is set to 1 and ⎡ A A I ⎤ ⎡ ∆x ⎤
vb km = 0 . A similar approach is used for entering other ⎢L ⎥⎢ ⎥=F=0 (3)
⎣ LA L d ⎦ ⎣∆x LF ⎦
models. An ideal transformer with secondary nodes k - m , The matrix A is the linear network matrix (left-hand side
primary nodes i - j and a transformation ratio g is modeled matrix in equation (2)), A I is a connectivity matrix for
with the equation: vk − vm − gvi + gv j = 0 . This equation is accounting for load-flow devices, L LA and L d provide load-
again directly accommodated in the submatrix Dr with the flow device constraint equations. The unknowns are the
standard network variables ∆x = [ ∆v n ∆iV ∆i D ∆i S ] and
t
equivalent row transposition appearing in Dc . Ideal switches
are included straightforwardly using submatrices S . When a the load-flow device currents I and internal voltages E :
switch (between nodes k - m ) is closed: vk − vm = 0 . When t
∆x LF = ⎡⎣ ∆I L ∆I PQ ∆I PV ∆I SL ∆E PQ ∆E PV ∆E SL ⎤⎦
the same switch is open, its current becomes 0 by setting the
The subscripts are: L for loads, PQ for PQ control sources,
corresponding diaogonal cell in S d to 1. The solved system is
PV for PV control and SL for slack bus.
now of fixed rank and quickly reformulated at each switch
Equation (3) presents a new approach for multiphase load-
status change. Adding a resistance and a fixed voltage drop
flow solution. It is different from earlier presentations [9] by
(diode effect) to the switch, is achieved using
means of a concatenation of load-flow equations with the
vk − vm − Rikm = vdc . This equation goes into S r , S d and sb .
readily available network equations. It can inherently
A system similar to (2) was initially proposed in [6]. The accommodate all EMTP device models and provide the most
precise initialization. Non classical load-flow devices, such as Scripting is extended to drawing using EMTPWorks
frequency-dependent transmission lines, become automatically methods (extensions) added to JavaScript. Drawings can be
included. There is no need to perform separate and automatically updated through scripts for reflecting data
cumbersome Jacobian derivations [9] for network models, but changes. This is demonstrated in Fig. 2 for the ideal
only for load-flow device constraint equations. transformer unit when updating the number of secondary
windings.
III. IMPLEMENTATION
132.8kV Tr0_1 441kV Tr0_1
The EMTP-RV design is object-oriented. Such a design has + + + +

several benefits in program development and maintenance.

Modern programming concepts, such as data hiding and 699.1238 34
+

encapsulation, eliminate data corruption problems caused by

device models and allow establishing a clear programming 200

approach by gathering all model data and methods through a Fig. 2 Automatic redrawing
single object and module. Model code legibility is
dramatically increased in comparison with old Fortran-77 In addition to scriptable drawing, a Symbol Editor is
coding used in EMTP-V3 while improving computational available for complete customization. The example of Fig. 3
speed. Although the core of the code is programmed using [11] makes the substation drawing appear as the actual
Fortran-95 (new object oriented language, see [10]) for substation using the Symbol Editor and subcircuits. An
achieving best computational speed, mixed language assembly of devices can be placed into a subcircuit with
programming is allowed specially for establishing connections programmable masking. Subcircuits can contain subcircuits.
with external functions and the graphical user interface. This is a hierarchical approach with unlimited number of
Another major benefit over EMTP-V3 is case-tailored and levels. In EMTPWorks subcircuits are created automatically
automatic memory allocation. It is a key issue in the from user selections on the screen and pin interfaces.
software’s ability to solve extremely large cases and/or usage Hierarchical designs allow visualizing study cases at a
of very small time-steps with transmission line models. higher level. In the wind park study of Fig. 4, the details of the
Modern programming methods and tools can significantly wind generator control circuits, the wind generator model, the
alleviate the software development process and contribute to wind effect representation, the static compensator (Statcom)
the implementation of more complex algorithms. and the load are hidden in subcircuits to provide a high level
visualization to the actual simulation. These are multilevel
IV. NEW GRAPHICAL USER INTERFACE subcircuits with masks for allowing the user to specify internal
parameters. The masks allow entering extensive script codes
The innovations in this field are in the development of an
or calls to external functions for calculating primitive device
open architecture and generic circuit oriented interface with
data at any subcircuit level.
extended single-line-diagram handling capability. Signals (not
Incoming 420 kV Transmission line
nodes), devices and connectivity pins are the constitutive
elements. EMTPWorks is designed for manipulating
extremely large networks. 1
+

Line CVT
2 400/63 kV
Almost all aspects of the GUI named EMTPWorks are 120 m P Q
Load2
120 m
scriptable and modifiable by the user. The scripting language 20 m 20 m 20 m 20 m 20 m 20 m
+

is JavaScript with EMTPWorks extensions. Connectivity with + + + + + +

other languages and ActiveX usage is readily available. Data

input is based on web pages. Each device has a local web 65 m 1 65 m
2 400/63 kV
120 m
address. This approach also allows data exchange and device P Q

20 m 20 m 20 m
addressing through the Internet. It also provides quasi 65 m
+

+
+

+ + + +
b

unlimited capabilities in the development of data forms

+

1 Ph.
CVT 24m
24m
through dynamic HTML programming. The fact that device
20 m 20 m 20 m 20 m 20 m 20 m
+

data and symbol methods are not hard-coded provides users

+

+ + + + + +
b

with maximum flexibility for creating customized devices.

+

110 m
110 m 1 Ph.
EMTPWorks is an object-oriented program written in BR_9832 BR_9432 CVT
+

C/C++ and based on Microsoft Foundation Class library. Each +

Line CVT &

+

CVT 100 MX
network device is an object with attributes. Attributes are 50 MX Shunt Reactor
Shunt Reactor
available to the user and used for entering data and methods.
Outgoing 420 kV Transmission line
Device attributes are the customization entry point. Other
objects, such as device pins and signals are given attribute
Fig. 3 Example of complete redrawing with symbol editor
specification menus. This is an open-architecture approach.
 120kV /_17
Network
+

Equivalent 120 kV Network

Wind

v + +

0.2uF
69kV /_0
69/0.69

Q_net
75 75
+

+ +
scope
CP CP

Q
Fault
scope
+

1E15/1E15/0
P
P_net
?i

9/9.15/0
9/9.15/0
Active and
+

Passive +
Load SW_Fault
+

1
SW_Network ?i
-1/9.15/0

420 MW Load
+
Fig. 4 Wind park test case with wind behavioral model

C3

C7
+

32 MW Synchronous Motor Load

1Ohm
+

0.2uF

Load1
0.1
V. DEMONSTRATION EXAMPLES

380uF
1 2

Q
P
A. Case 1

YgD_1
120/26.4
VM +
Vnet
?v

DYg_SM2

13.8/122
2

SM_load
The diagram of this test case is shown in Fig. 3. The Omega_1

25.5/12
1

SM_load_control
simulation results can be found in [11]. It is a demonstration 1 2

1
SM

?m
of numerical stability and precision in solving extremely

f(u)
+ +

YgYg_np1
1

40MVA
1uF

12kV
nonlinear arc model problems. Handling of specific computer

?m
500MVA
13.8kV

SM2
SM
Pm

requirements, such as memory for the frequency dependent

models with very small time-step usage, are also confirmed.

AVR&Gov
Out

IN
B. Case 2

(pu)

YgYg_np2

25.5/6.6
1
AVR_Gov_SM2
The design view for this test case is twice the size of the
ASM2

11000hp
6.6kV
?
network shown in Fig. 4. The test case is explained in details S ASM
?m
in [12]. The Statcom is a 3-level Statcom with PWM control.

?i

SW_ASM1
1/1E15/0
+

Induction motors in steady-state

Its IGBT model is based on a nonlinear diode function for
ASM3

11000hp
Q_ASM1 P_ASM1

6.6kV
increased precision. All nonlinear functions, including the

P
scope S ASM

Q
scope ?m
asynchronous machines are solved simultaneously with

C4
network equations. The simulation time-step is 50 µs . The ASM4

11000hp
+

6.6kV
40uF
mean number of iterations per time-point is below 3. S ASM
Starting motor at 1 s

?m
ASM1_control

Tm

11000hp

C. Case 3
T
ASM
ASM1

ASM5
6.6kV

11000hp
S

6.6kV
Speed

In this example (Fig. 5) 5 asynchronous machines are S ASM

?m

initially in steady-state when a motor is started. A fault on the ?m

connected network bus is initiated at 9s and islanding occurs ASM6

11000hp
near 9.15 s . Local power is delivered by the synchronous S ASM 6.6kV

machine SM2. Fig. 6 presents connected asynchronous ?m

machine slips. Machine data is taken from [13]. Machines are +

automatically initialized using their slip specification. The

240uF

synchronous machines and the corresponding controls are also

Fig. 5 Fault and system islanding test case with machines
automatically initialized.
All nonlinear devices are solved simultaneously with 0.04

network equations. Transformers are modeled with a 0.035

nonlinear magnetization branch. In EMTP-RV machines are 0.03
Slip_ASM2
also classified as nonlinear devices and solved through the 0.025 Slip_ASM3
Slip_ASM4
iterative process. All nonlinear devices are true-nonlinear and Slip_ASM5
Slip_ASM6
0.02
can coexist in the same network without artificial insertions
0.015
such as stub lines used in EMTP-V3. The test case assembled
0.01
in Fig. 5 could not be simulated in EMTP-V3.
0.005
Despite the strongly connected network of several
machines, the mean number of iterations per time-point is 3. 0
0 2 4 6 8 10 12 14 16
time (s)
Only 55 s (!) of CPU time are required to simulate an interval
of 50 s with a time-step of 200 µs ( 2 GHz computer).
Fig. 6 Connected asynchronous machine slips
 0.00378S
avr_governor_pu
13.8kV

C1
AVR_Gov_2

SM4
200MVA
BUS: BUS5 BUS2
Out ?m BUS4 +

Load5
SM LF 45MW

0.0024S
YgYg_BUS4 21MVAR
IN 1 2
LF SubofBUS1
AVR&Gov

C4
BUS1 YgYgD_1
(pu) SM_BUS4
P=132MW 13.8/230 SUB_BUS1 + 2 500/230/50
V=13.8kVRMSLL
1
SM:SM4

+
BUS10 Power exchange

1.60934E+01
TLM_10mi
TLM_120mi
13.8kV 3 with another network
SM6

CP
AVR_Gov_3 550MVA + CP

+
193.1213 Slack
BUS:
Out YgYgD_2 + ?i
?m BUS6 BUS7
500kVRMSLL /_0
SM 2
DYg_BUS6 LF Slack:Slack
1
IN 1 2 Slack
AVR&Gov TLM_90mi BUS12 BUS13
YgD_BUS13 Slack: 500kVRMSLL/_12.96
(pu) LF TLM_15mi

+
SM_BUS6 13.8/230
+
144.8410
CP 2 1 3
P=520MW + CP AVR_Gov_4
V=13.8kVRMSLL 2.41400E+01 69/230 SM11
BUS1
SM:SM6 13.8/230 BUS11 Out
?m

a
c

18MVAR

20MVAR

14MVAR
2 1 SM
+

50MW

45MW

30MW
9.65606E+01 IN

LF

LF

LF
LF 13.8kV
TLM_60mi

AVR&Gov
125MVA
Load13c Load13b Load13a SM_BUS11 BUS:
(pu)
P=76MW
CP

290
V=14kVRMSLL
SM:SM11
BUS2
13.8kV

CP2
50MVA
avr_governor_pu PVbus:SM_BUS8
BUS9 TLM_180mi
+

AVR_Gov_1 SM8
BUS8
Out ?m
DYg_BUS8
SM 1 2
IN
Load9

LF 240MW
AVR&Gov
LF 13.8/230 60MVAR
(pu)
SM_BUS8
P=40MW
BUS8
V=13kVRMSLL + VM
SM:SM8 ?v

Fig. 7 Load-flow demonstration test case

30
SM11

D. Case 4 25
Omega

The 230 kV network case shown in Fig. 7 starts from a 20 SM6

load-flow solution and automatically initializes network state SM4 SubofBUS1/SM3

variables and machine controls for almost immediate 15
SM8
establishement of steady-state operating conditions in the
10
time-domain simulation. The network has unbalanced 0 50 100 150 200 250
t (ms)
300 350 400 450 500

transmission lines and unbalanced loading at BUS13. Fig. 8 Machine angular velocity (rad/s)
Multiphase load-flow constraint data is entered through load-
600
flow (LF) devices and is layered over the network simulated
in time-domain. Any network can be augmented with load- 500 SM6

flow devices without any other manipulations. All machine

400
voltage amplitudes and angles are initially arbitrarily set to
positive sequence data with phase-a at zero degrees. The load- 300
flow algorithm is started first. It converges within 6 iterations
and generates load-flow solution data which automatically 200 SubofBUS1/SM3

initializes the steady-state solution of the actual network with SM4

100 SM11
synchronous machines. All synchronous machines and
corresponding control circuits (AVR and Governor) are 0
SM8
0 50 100 150 200 250 300 350 400 450 500
automatically initialized at the time-domain simulation startup. t (ms)

This is demonstrated by the waveforms for speed and power Fig. 9 Machine power (MW)
shown in Fig. 8 and Fig. 9. The initialization of synchronous
machine models is particularly more difficult under E. Case 5
unbalanced operating conditions and requires special
An exceedingly complex software engineering matter with
treatments in the steady-state solution process.
the graphical user interface design and the computational
engine is the capability to show and simulate very large
networks within acceptable clock time. EMTP-RV has been
tested on very large networks. The largest network assembled
so far is taken from the Hydro-Québec system. The summary
printout on primitive devices indicates: 25 synchronous
machines, 63 ideal sources with impedance, 1248 transformer
units, 3704 RLC branches, 610 coupled branch sections, 365 [6] J. Mahseredjian and F. Alvarado: “Creating an Electromagnetic
Transients Program in MATLAB: MatEMTP”. IEEE Transactions on
transmission lines, 1600 load models, 1500 switches, 378
Power Delivery, January 1997, Vol. 12, Issue 1, pages 380-388.
nonlinear devices (arresters, nonlinear inductances…) and [7] J. Mahseredjian, S. Lefebvre and X.-D. Do: “A new method for time-
6838 control system devices. In several cases primitive domain modeling of nonlinear circuits in large linear networks”. 11th
devices are used for building specialized devices such as Power Systems Computation conference (PSCC), Proceedings Vol. 2,
August 1993, pp. 915-922.
SVCs. The total number of devices is above 16000 for
[8] J. Mahseredjian, L. Dubé, M. Zou, S. Dennetière and G. Joos:
approximately 7500 nodes. The entire design can be shown in “Elimination of numerical delays in the solution of control systems in
EMTPWorks using 6 pages and has a load time (from open EMTP”, submitted to IPST 2005 (International conference on power
command to the complete appearance) of less than 14 s on a system transients), June 19-23, 2005, Montréal
[9] J. J. Allemong, R. J. Bennon, P. W. Selent: ”Multiphase Power Flow
2 GHz computer. It requires slightly less than 40 minutes per Solutions using EMTP and Newtons Method”, IEEE Trans. on Power
actual real-time second of simulation with an integration time- Systems, Vol. 8, No. 4, November 1993, pp. 1455-1462
step of 50 µs . [10] J. Mahseredjian, B. Bressac, A. Xémard, P. J. Lagacé and L. Gérin-
Lajoie: “A Fortran-95 implementation of EMTP algorithms”.
There are several advantages in the ease of maintaining Proceedings of IPST’01 (International conference on power system
such a network: unique location of network data, transients), Brazil, June 2001, pages 686-691.
exclude/include functions for partial simulations and [11] B. Khodabakhchian, J. Mahseredjian, M.-R. Sehati and M. Mir-
Hosseini: ”Potential Risk of Failures in Switching EHV Shunt Reactors
capability to perform tests on the actual network and/or in in Some one-and-a-half Breaker Scheme Substations”. Proceeding of
several locations. EMTPWorks provides scripting methods International Conference on Power Systems Transients, IPST 2003 in
and table-browser editing functions for manipulating data New Orleans, September 28th – October 2nd, 2003.
without clicking on devices one-by-one. [12] C. Abbey, B. Khodabakhchian, F. Zhou, S. Dennetière, J. Mahseredjian
and G. Joos: “Transient Modeling and Comparison of Wind Generator
Topologies”, submitted to IPST 2005 (International conference on power
VI. CONCLUSIONS system transients), June 19-23, 2005, Montréal
[13] G. J. Rogers, D. Shirmohammadi: “Induction machine modeling for
This paper presents an overview on a new simulation tool
electromagnetic transient program”, IEEE Trans. On energy Conversion,
named EMTP-RV. It is the new generation of the DCG-EMTP Vol. EC-2, No. 4, December 1987, pp. 622-628
software. It constitutes a new contribution in the history of
industrial grade EMTP tools. EMTP-RV has several original VIII. BIOGRAPHIES
solution methods that contribute to the elimination of Jean Mahseredjian (M'87) graduated from École Polytechnique de Montréal
topological restrictions, to improving the solution of with M.A.Sc. (1985) and Ph.D. (1991). From 1987 to 2004 he worked at
nonlinearities, to the flawless simulation of control systems IREQ (Hydro-Québec) on research and development activities related to the
and to the automatic steady-state initialization methods. simulation and analysis of electromagnetic transients. In December 2004 he
joined the faculty of electrical engineering at École Polytechnique de
EMTP-RV is also based on a new graphical user interface
Montréal.
with innovations in its open-architecture approach and Sébastien Dennetière (M'04) graduated from École Supérieure d'Electricité
scripting methods. (Supélec) in France in 2002. He received the M.A.Sc. degree from École
The recoding from scratch of major scientific software such Polytechnique de Montréal in 2003. From 2002 to 2004 he worked at IREQ
as EMTP, constituted an enormous challenge. The stressed (Hydro-Québec) on research and development activities related to the
simulation and analysis of electromagnetic transients. In October 2004 he
points were research and development on improved solution joined the research center of EDF where his interests are currently in the field
methods and software engineering approaches. Modern of insulation coordination and power system simulations.
programming methods and tools can significantly alleviate the Laurent Dubé (M'77) obtained the B.A. (1967) and B.Eng. (1972) from
software development process and result into more powerful Sherbrooke University and M.Eng. (1973) from École Polytechnique de
and user-friendly programs even within a highly scientific Montréal. He developed the TACS program of EMTP and the MODELS
simulation language and solver. He recently worked with IREQ (Hydro-
context. Québec) on the development of user-defined modeling for EMTP-RV. His
consulting activities are in the field of user-defined modeling methods and
VII. REFERENCES languages.
[1] H. W. Dommel: “Digital Computer Solution of Electromagnetic Bahram Khodabakhchian holds BS and MS (École Polytechnique de
Transients in Single- and Multiphase Networks”, IEEE Trans. on Power Montréal) degrees in electrical engineering. He has over 23 years of
Apparatus and Systems, vol. 88, no. 4, April 1969. experiences in power system studies at Hydro-Quebec. He is currently in
[2] EMTP Development Coordination Group, Electromagnetic Transients charge of the marketing and sales of EMTP-RV, the new Restructured Version
Program Version 3. of the DCG-EMTP.
[3] C. W. Ho, A. E. Ruehli and P. A. Brennan: “The modified nodal Luc Gérin-Lajoie received his B.Sc.A in Electrical Engineering from École
approach to network analysis”. Proc. 1974 International symposium on Polytechnique de Montréal in 1982. From 1982 to 1985, he worked for Hydro-
circuits and systems, San Francisco, pp. 505-509, April 1974 Québec in System Planning. In 1985 he joined the Control and Protection
[4] H. K. Hoidalen, L. Prickler, J. L. Hall: “ATPDraw- Graphical team in the Systems Studies Department. His responsibilities include
Preprocessor to ATP. Windows Version.”. Proceedings of International protection specification, modal analysis, optimization and control
Conference on Power Systems Transients, IPST 1999, Hungary, June coordination, as well as electromagnetic phenomena and analytic studies
20-24, 1999, pp. 7-12. related to the performance of SVC and synchronous machines on the bulk
[5] O. Nayak, G. Irwin and A. Neufeld: “GUI enhances electromagnetic transmission system. He also contributed for Hydro-Québec, the validation of
transients simulation tools”, IEEE Computer Applications in Power, Vol. the new EMTP-RV software.
8, No. 1, pp. 17-22, January 1995.