1

2006 IEEE PES Transmission and Distribution Conference and Exposition Latin America, Venezuela

A Time-Domain Simulation Framework of an

IPR-Based Shipboard Integrated Power System
Christian A. Feliciano-Bonilla, and Bienvenido Vlez-Rivera

Index Terms Intelligent Power Router, IPR, power system

control, power system modeling, reconfiguration, shipboard
power system.

I. INTRODUCTION

E live in an era when the U.S. Navy has as some of its

main goals the reduction of manning on board its ships
and the increase of the survivability of its vessels. The
automatic reconfiguration of the electrical network in the
Navys shipboard power systems is critical for quickly
restoring service to those sections of the power system
severed by failure or battle damage [1]. The complexity of the
ships electric system, structure, time of response and many
other predominant factors will define the behavior of the other
systems interconnected to it. These interdependent and
automated systems will require a level of control that will
allow them to adapt and reconfigure the power network in
response to dynamically changing conditions as required by
the specific demands of different missions. Existing ship
protection systems are often deficient in supplying critical
devices and systems during battle or fault conditions and
within a practical response time frame. The control strategies
that are implemented in these cases are not sufficiently
effective to recover the system; therefore human intervention
is often necessary to restore the system manually. The system
must be capable of seamlessly transferring power during fault
conditions since a small glitch in power to a vital system may
have catastrophic results [2]. Many similarities exist between
This work was supported by the National Science Foundation thru EPNSE
award number 0224743 as well as by the University of Puerto Rico Mayagez Campus.

1-4244-0288-3/06/$20.00 2006 IEEE

a navy ship trying to survive to a missile attack and a hospital

or police station trying to survive to a natural disaster or
terrorist attack. So an attempt to automate the electric power
systems will benefit both the military and civilian sectors.
A real time simulation framework for the next generation
shipboard integrated power system network was developed
using a distributed concept based on scalable coordination by
an Intelligent Power Router (IPR). We hypothesize tht by
distributing the network intelligence and control functions
across multiple independent IPRs, the survivability, security,
reliability and re-configurability of the ship power system will
be improved significantly.
The fundamental engineering design principle behind the
IPR system is modular decentralized control. We propose a
system in which the control can be detached from central
control sites, and delegated to intelligent power routers
strategically distributed over the entire Electric Energy
Processing Network. Each IPR exploits its embedded
intelligence to switch power lines, shed load based on a
priority scheme, activate auxiliary or distributed generation,
isolate power regions of the energy delivery network to
prevent system cascade failures and receive/broadcast local
state variable information to and from other routers [3].
Fig. 1 shows that each IPR controls a series of power input
lines coming from either a power generator or another source,
possibly under the control of a different IPR. Also each IPR
controls a series of output lines that provide power to loads or
act as power sources to other power nodes routers which in
turn may be controlled by an independent power router.
Together the routers comprise a collaborative network
providing multiple redundant power paths between generators
and loads.

Fig. 1. Power system model with Intelligent Power Routers.

II. IPR ARCHITECTURE

The architecture of an intelligent power router is shown in
Fig. 2. Intelligent power routers are modular building blocks
that collaborate to reconfigure the power network in response
to failures and other contingencies. The router consists of
energy sensors and flow control devices that move power
between input and output lines. The flow control devices
operate under the control of a Programmable Intelligent
Communication and Control Unit (ICCU), which has the
necessary logic to determine how to route power from the
input lines to the output lines. Each IPR maintains information
on the power flowing through each of its connecting power
lines. This information is used to make local decisions on how
to reroute power in the event of changes in the flow of power
through the lines, which might be caused by faults or changes
in power generation or demand.
The simulation tool used to develop the IPR model and
all the components of the shipboard power system was
SimPowerSystems, a simulation platform that permits to
design detailed power systems models on a user-friendly
environment. SimPowerSystems uses the Matlab-Simulink
engine, permitting to build systems models using click and
drag procedures. Because Simulink uses MATLAB as its
computational engine, it is possible to integrate other
MATLAB toolboxes and Simulink blocksets to extend our
framework.

breaker block included in the SimPowerSystems model

library. The control logic signals will be sent by the ICCU of
each IPR after having determined a new reconfiguration
topology for the electric network. Series resistance and
capacitance snubber circuits are included in the model. They
can be optionally connected to the three individual breaker
components of the three-phase breaker model. If the block is
in series with an inductive circuit, an open circuit or a current
source, the snubber circuits have to be connected.
To obtain a multi-directional control of the power at each
IPR, a four port power input/output model was developed.
Each one of these ports will have a real and reactive power
sensor, a breaker and the interfacing circuits necessary for the
information exchange between sensors, switches and the
ICCU. Fig. 3 shows all the interconnected components in one
of these ports. The power flow can either enter or leave
through the connection ports or the breaker terminals. There is
a power sensor connected in series with a display to show the
value at each instant in time. The connected interfacing circuit
block transmits the port information to the ICCU of the
corresponding IPR. Fig. 4 shows the schematic of the model
obtained when four ports of this type are connected. The
power flow can enter through any of the four ports and can
leave through any of the remaining ports. The result is a
modular device that can control the flow of power in several
directions.
Connecting this energy flow control device (EFCD)
module in different system zones will allow the power from
different generators to flow to the connected loads in the
system using any of the available transmission lines. In the
next step we add to the module the intelligent communication
and control unit (ICCU) that will gather the sensor data and
the state of the switches as input. Then, it will calculate a new
configuration state for the network topology depending on the
input variables mentioned before and information that
neighbor IPR's offer or be required from them. The ICCU,
responsible for these algorithms will return the new switches
state (on/off) to the energy flow control module. The new
system topology will reconfigure the system and in this way
supply the critical or vital loads as quickly as possible taking
into account the physical constraints of the system.

Fig. 2. Intelligent Power Router architecture.

A. Sensors and Flow Control Devices

In our first generation IPR, the electric system parameters
measured by the sensors connected to the device are the phase
to ground voltages and the phase currents. With that
information the real and reactive power flowing through the
lines can be calculated. Power flowing from the generators to
the IPRs, power through the lines connecting two IPRs and
power flowing from the IPRs to the loads can be obtained. In
summary, the power at any point of the power system can be
measured.
The flow control devices for this first generation of the
IPR are simulated using a logical signal controlled three-phase

Fig. 3. Power input/output port with energy sensor and breaker.

Port 1
Port 3

Port 2

Port 4

Fig. 5. Intelligent Communication and Control Unit (ICCU) schematic.

Fig. 4. Four port energy sensors, breakers and interfacing circuits module.

B. Intelligent Communication and Control Module

The Intelligent Communication and Control Unit (ICCU) is
responsible for collecting all the information measured by
sensors, and depending on the IPR parameters and the status
of its switches, calculate a new topology to maximize amount
of load served giving priority to vital loads. Also, it has the
capacity to communicate with neighboring IPRs to learn the
status of higher priority loads connected to them. It is
important to emphasize that all the IPRs have the same type
of ICCU, which linked with the sensors and switches module,
completes the modular design of the IPR device, a new
building block that can be used to construct complex energy
distribution networks.
The ICCU was constructed using using block models
provided by the Matlab-Simulink program libraries. An IPR
can establish communication only with their directly
connected neighbors so they can exchange information like
necessary power needed by load priority and also can serve as
power brokers for non-directly connected IPR's. Fig. 5 shows
the ICCU schematic and Table I describe the input and
outputs of the IPR main controller inside the ICCU.
C. Interconnecting the ICCU and the EFCD
The sensors and switches module model shown in detail on
Fig. 4 and the ICCU shown in Fig. 5 were encapsulated to
facilitate the handling of the modules in the shipboard power
system model. Fig. 6 shows the connection of the two
modules which completes the IPR model. The EFCD module
contains four input/output three-phase power ports, two
outputs containing the ICCU input data (sensed parameters,
flow devices status), and an input coming from the ICCU with
the switches logic control data. An important characteristic of
this IPR model is that it has the capacity to be connected in
series with other modules to provide more power ports in the
event that the system requires it.

TABLE I
INTELLIGENT COMMUNICATION AND CONTROL UNIT (ICCU) INPUT AND
OUTPUTS DESCRIPTION

ICCU
Input/Output
1
2
3
4
5
6
7
8
9
10
11

Description
Parameters sensor sampling time
IPR parameters
Measured system parameters
Breakers status
IPR communication data input
Line protection
IPR switch
Breakers control
IPR communication data output
IPR connected load
IPR main controller

Fig. 6. Intelligent Power Router (IPR) encapsulated systems.

III. DISTRIBUTED RECONFIGURATION ALGORITHM

The IPR distributed reconfiguration algorithm was
programmed using a graphic design tool called Stateflow that
also works with Matlab-Simulink. Stateflow is a simulation
environment for modeling the logic used in the control and
supervision of physical systems modeled in Simulink.
Visually model and simulates logical controls, providing
complex descriptions of the system behavior, using finite state
machines theory, flow charts, and states transition diagrams,
in a same window. In Simulink, a Stateflow block uses a state
diagram to represent an object with a set of discrete modes.

These modes are known as states. A finite state machine in

Stateflow reacts to events, changing states for the controlled
object. The object behavior depends on which state it is and in
how the object changes from one state to another.
Fig. 7 shows the proposed finite state machine for the IPR
controller. It consists of four super-states called: steady state,
contingency detected, attending request and request broker.
Each IPR has a controller with an embedded finite state
machine of this type. The Stateflow blocks in the shipboard
power system model are found inside the ICCUs, as can be
observed in Fig. 5. During a simulation each IPR react
independently of the other IPRs. For example if an IPR
detects a problem in one of its lines the controller state
machine of that IPR will pass to the contingency detected
super-state to request the power needed to supply at least it
vital loads. The neighbor IPRs receiving the power request
will pass to the attending request state if they have a
generation source connected, if not they will pass to the
request broker super-state to act as power brokers between the
IPRs not connected directly. The request message will
continue passing through the IPRs until a positive response is
received by the petitioner IPR. If the power request cannot be
supplied the state machines will pass to the steady-state superstate to wait for another event.

decide where will flow the power, permit to divide the ship in
electric zones. Fig. 8 shows the shipboard power system
divided into five zones. Ten IPRs are integrated into the
system identified in the figure with a dashed circle around
them.

Fig. 8. Shipboard power system divided in five zones.

Fig. 9 demonstrates the complete power system with IPRs

used for the testing of the model under various scenarios. All
of the system components are identified including the loads
priorities. Priority one loads are classified as vital, priority two
are classified semi-vital and priority three as non-vital loads.
Also included in the figure are the main and auxiliary
generators, the propulsion motors and the load transformers.
As can be seen some of the IPRs have generation sources
connected and others not. Some only have loads and others
IPRs connected. For these reasons the ICCU needs to know
what things it has connected and their characteristics, like the
generators capacities, the lines power capacities, the
magnitudes of the connected loads and it neighbor IPRs.
V. EXPERIMENTAL RESULTS

Fig. 7. Intelligent and Communication Unit (ICCU) proposed state machine.

IV. SHIPBOARD POWER SYSTEM MODEL

The platform used to integrate the IPR's to an electric
network and verify its functionality is based on an integrated
power system architecture for future naval ships. The power
system original model was developed by the Center for
Electromechanics (CEM) at the University of Texas with the
purpose of simulating real operational scenarios, and makes a
power system evaluation of the ship performance during
different conditions [4].
The power system under consideration reflects the notional
DD power system architecture and is developed on the
SimPowerSystems Matlab/Simulink environment. The model
permits to analyze reconfiguration settings, energy storage and
power quality problems. Also permits new technologies
introduction into the system components like generators and
propulsion motors to be able to achieve the required power
density level. The advantage of having four controlled ports to

The experimental results were carried out creating various

contingencies scenarios in which the ships electric network
suffers power losses in loads or generators of different zones.
Because the only system parameter measured in this first
version is the real power, an attack or some system internal
fault is simulated disconnecting the lines supplying an IPR.
Switches connected in lines between generators and IPRs and
in lines between IPRs are closed in normal state. For each
contingency scenario some of them are programmed to open
at defined time intervals for simulating in this way an attack or
fault. When the switches are opened some loads will loss
power and the ICCU of the corresponding IPR will react to
take action.
A. Non-Simultaneous Contingencies Scenario
Fig. 10 show the normal state power flows of the shipboard
power system generators. The main and auxiliary generators
supply the propulsion motors and the load service zones
respectively. In the presented scenario all the service zone
loads lose their power because the auxiliary generators are
disconnected by the protection system in response to four nonsimultaneous contingencies. These contingencies occur in

Fig. 9. Shipboard power system simulation model components.

different zones of the ship at 1.3, 1.8, 2.4 and 3.0s of

simulation time as shown in Fig. 11. The affected lines and the
loads without power after the contingencies are also shown in
the figure. For this scenario the backup generator has a
generation capacity of 120 kW while the main generators have
a capacity of 400 kW each one. All the service zone loads has
a fixed magnitude of 20 kW.

five of the service loads, being at its maximum capacity. The

main generators supply two of the priority one loads and the
two propulsion motors, remaining almost at their maximum
capacities. The power flow of each generator after executing
the IPR's reconfiguration scheme can be observed in Fig. 12.
TABLE II
CURRENT MAGNITUDE PLOTS OF SYSTEM DEVICES DURING THE SIMULATION

Fig. 10. Generators power flows in normal state.

TABLE III
Fig. 11. Disconnection times and affected loads.

Table II shows the behavior of the current magnitude in

different devices of the shipboard power system during the
simulation time. Table III summarizes the affected loads
reconfiguration times. As can be seen 100% of the loads were
supplied in an acceptable time. The backup generator supplied

AFFECTED LOADS RECONFIGURATION TIME DATA

6
[10] A. Ouroua, L. Domashk and J.H. Beno, Electric Ship Power System
Integration Analyses through Modeling and Simulation, in 2005 IEEE
Electric Ship Technologies Symposium, pp. 70-74, July 25-27 2005.
[11] K. Davey and R. Hebner, Reconfiguration of Shipboard Power
Systems, in IASME Transactions, Vol.1, pp. 241-246, April 2004.
[12] S. Khushalani and N.N. Schulz, Optimized Restoration of Shipboard
Power Systems with IPS Architecture and Distributed Generation, in
Reconfiguration and Survivability Symposium, 2005.
[13] Naval Sea Systems Command Engineering Directorate - Electrical
Engineering Group, NAVSEA Design Practice and Criteria Manual for
Electrical Systems for Surface Ships, Chapter 300.

VIII. BIOGRAPHIES
Fig. 12. Generators power flow scheme after reconfiguration.

VI. CONCLUSIONS
We have developed and verified a first generation model of
the IPR, a new modular, embedded intelligence device to
achieve decentralized control of a future naval ship power
system. The IPR building block permits a more organized
arrangement of power distribution complex networks,
complying with the proposed design objectives of:
survivability, fault tolerance, scalability, cost-effectiveness
and unattended 24/7 operation. The creation of this new
simulation platform permits the real-time detailed analysis of
the devices behavior that composes the network and the IPRs.
Our next objectives are: the design of a second generation IPR
control software model, sensing of others parameters
(frequency, reactive power) and the addition of power
electronics flow control devices to the IPR module.
VII. REFERENCES
[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

K.L. Butler-Purry and N.D.R. Sarma McCoy, Intelligent Network

Reconfiguration of Shipboard Power Systems, in Power Engineering
Society General Meeting, July 2003
J.G. Ciezki and R.W. Ashton, Selection and Stability Issues Associated
with a Navy Shipboard DC Zonal Electric Distribution System, in IEEE
Transactions on Power Delivery, Vol.15, No.2, pp. 665-669, April 2000.
Agustn A. Irizarry-Rivera, Manuel Rodrguez, Miguel Vlez-Reyes,
Jos R. Cedeo, Bienvenido Vlez, Efran O'Neill-Carrillo, and
Alberto Ramrez, "Intelligent Power Routers for Distributed
Coordination in Electric Energy Processing Networks", in Proceedings
of the 2003 EPNES Workshop, Orlando, Florida, 2003.
J. Beno, A. Ouroua, M.Flynn, Effect of EM weapons requirements on
the electric ship power system, presented in IMAREST Engine as a
Weapon 2004, June 2004.
E.L Zivi, Integrated Shipboard Power and Automation Control
Challenge Problem, in Power Engineering Society Summer Meeting,
Vol. 1, pp. 325-330, July 2002.
H. Zhang, K.L. Butler, N.D.R. Sarma, H. DoCarmo, S. Gopalakrishnan
and A. Adediran, Analysis of tools for simulation of Shipboard Electric
Power Systems, in Electric Power Systems Research, pp. 111-122,
February 2001.
S. Khushalani and N.N. Schulz, Optimized Restoration of Shipboard
Power Systems with IPS Architecture and Distributed Generation, in
Reconfiguration and Survivability Symposium, 2005.
T. Nagata, Y. Tao, H. Sasaki and H. Fujita, A Multi-agent approach to
Distribution System Restoration, in The 2004 47th Midwest Symposium
on Circuits and Systems, Vol.2, pp. 655-660, July 2004.
K.L. Butler, N.D.R. Sarma and V.R. Prasad, Network Reconfiguration
for Service Restoration Shipboard Power Distribution Systems, in
IEEE Transactions on Power Systems, Vol.16, pp. 653-661, November
2001.

Christian Feliciano-Bonilla was born in Mayagez,

Puerto Rico on December 20, 1981. He holds a BS
degree in electrical engineering obtained in 2004
from the University of Puerto Rico Mayagez
Campus. He is currently working an MS degree in
electrical engineering from the same university. His
work interest is in power systems and power
electronics. His present work focus is on modeling
and simulation of reconfiguration schemes for
electric ship power systems.
Bienvenido Vlez-Rivera was born in Ponce Puerto
Rico, on March 19, 1964. In 1986 he obtained a BS
degree with a double major in Computer Science and
Mathematics from Cornell University. He completed
an MS in Computer Science from the University of
California Berkeley in 1988 working in the area of
databases. He returned to his natal Island in 1989
where he lectured computer science until 1993,
when he began working towards a doctoral degree in
Computer Science which he completed in 2000 at
the Massachusetts Institute of Technology in the area of Information Retrieval
and Distributed Systems under the guidance of Dr. David K. Gifford. Dr.
Vlez-Rivera is currently an Associate Professor of computer science and
engineering at the University of Puerto Rico Mayagez campus and leads
several research projects in the areas of Digital Government, Distributed
Information Systems and Web Application Frameworks.