Transcript epnes2003-v2 - ECE - UPRM

EPNES: Intelligent Power Routers for
Distributed Coordination in Electric
Energy Processing Networks:
Report 1
Agustín Irizarry
Manuel Rodríguez
José Cedeño
Bienvenido Vélez
Miguel Vélez-Reyez
Efraín O’Neill
Carlos Torres
Idalides Vergara
Juan Jimenez
Marianela Santiago
Project Goal: Electrical Energy Networks Featuring
Intelligent Power Routers (IPRs)
Producers
P1
…
P2
…
P3
Pn
System
Reconfiguration
with
Minimal Human
Intervention
R2
…
Routers
R1
Rk
R3
R4
…
C1
September 25, 2003
C2
Consumers
EPNES: Intelligent Power Routers
C2
Cm
2
State-of-Art Power Delivery
Producers
P1
P2
P3
Pn
Consumers
C1
C2
C3
C4
Power systems with centralized control
September 25, 2003
EPNES: Intelligent Power Routers
3
Re-routing in Response to Failures
Producers
P1
P2
P3
Pn
System MTTR
Limited by
Operator
Response Time
x
x
Consumers
C1
September 25, 2003
C2
C3
EPNES: Intelligent Power Routers
C4
4
Re-routing in Response to Major
Disturbances
Producers
P1
P2
P3
Pn
Slow
Operator
Response
May Cause
Cascading
Failures
Consumers
C1
September 25, 2003
C2
C3
EPNES: Intelligent Power Routers
C4
5
Re-routing in Response to Major
Disturbances
Producers
P1
P2
P3
Pn
IPRS
Respond
Promptly
to Avoid
Further
Deterioration
Consumers
C1
September 25, 2003
C2
C3
EPNES: Intelligent Power Routers
C4
6
Our approach
• Decentralized control in response to major
disturbances
• Intelligent Power Routers (IPR):
–
–
–
–
modular building blocks
strategically distributed over entire network
embedded intelligence
information exchange allows neighboring IPRs to
coordinate network reconfiguration
– improve network survivability, security, reliability,
and re-configurability
September 25, 2003
EPNES: Intelligent Power Routers
7
Distributed Routing
Routers
Data
Consumer
S1
C1
Data
Servers
R1
R2
C3
Internet
R3
S2
R4
C2
Multiple redundant paths to move data between computers
September 25, 2003
EPNES: Intelligent Power Routers
8
Distributed Routing: Tradeoffs
• Advantages
– Highly reliable
• Multiple redundant paths to deliver the data
– Highly scalable
• Grow network by adding more routers incrementally
– Improved Performance
• Distributed and Parallel processing for data movement
• Disadvantages
– Complex Control: Requires intelligence!
• Continuously run routing algorithms to find possible
routes
– Complex Implementation
• Hardware and software not trivial to implement
September 25, 2003
EPNES: Intelligent Power Routers
9
Recovering from Failures
• Each router continuously monitors the
network
• When a broken link is detected by a router:
– Its routing table is updated to reflect unavailable
link
– Update notice is propagated to near neighbors
– Neighboring routers react accordingly
• Update their tables
• Propagate their updates to their own neighbors
• Idea is to find new paths to move the data
– Avoid routes that use broken link
September 25, 2003
EPNES: Intelligent Power Routers
10
Distributed routing for
power delivery systems ?
• We believe possible to use the
concept of distributed control and
coordination to obtain:
– Greater reliability
– Scalability
– Improved survivability
September 25, 2003
EPNES: Intelligent Power Routers
11
How are power delivery systems
different from computer networks?
– Energy (not data) is transmitted
– Must match generation to demand at all times
– No buffers
– Its a bit hard to get rid of excess energy
We must deal with the laws of Physics!
September 25, 2003
EPNES: Intelligent Power Routers
12
Project Organization
Distributed
Control
Models
IPR
Architecture
Economics
Education
Restoration
Models
September 25, 2003
IPR
PROTOCOLS
EPNES: Intelligent Power Routers
13
Potential architecture of the Intelligent
Power Router
Power System
Energy Flow
Control Devices
Sensor
Input
Switching
Commands
Interfacing
Circuits
ICCU
Intelligent Power Router
September 25, 2003
EPNES: Intelligent Power Routers
14
IPRs Design
• Basic Functionality of IPR
Take the role of
controlling the
routing of power
over the lines.
September 25, 2003
EPNES: Intelligent Power Routers
15
Simulation Tool
• Understand how to model physical components for
power system
• Creating self-defined models
September 25, 2003
EPNES: Intelligent Power Routers
16
Simulating the IPR
• Simulating
basic
functionality
of IPR
– Load Priority
– Line Priority
September 25, 2003
EPNES: Intelligent Power Routers
17
Power System Restoration
 Overview: Improvement of security and reliability of the electric
power system operation.
 Researchers: Juan J. Jiménez, Graduate Student UPRM
José R. Cedeño, Assistant Professor UPRM
 Research: Formulate the Power System Restoration (PSR)
problem and solve it with an Evolutionary Computation
technique.
 Approach: Use particle swarm optimization for solving the PSR
problem. Formulate the PSR problem as a multi-stage,
combinatorial, nonlinear, constrained optimization problem with
binary and continuous variables.
September 25, 2003
EPNES: Intelligent Power Routers
18
Power System Restoration
Problem formulation in terms of penalty functions:
Nn
Nk
NL
Nj
Nj


2
2
2
2
f i  min  AL  S L  M L  X L   AST  S   APG  Pj   AQG j  Q j   AVn  Vn   APFk  PFk2 
j
n 1
k 1
L 1
j 1
j 1


st,:
PGTotal  PDTotal  0
QGTotal  QDTotal  0
where,
  NL
  L 1







S =     X L     X k    X j   i 
Nk
Nj
k 1
j 1
The objective of the formulation is to minimize the
unserved load while satisfying the operating
constraints of the system. Also, at each stage of the
restoration process only one switching operation is
allowed.
Pj = PG j  PGlimj
Q j = QG j  QGlimj
Vn = Vn  V lim
PFk = PFk  PFklim
September 25, 2003
EPNES: Intelligent Power Routers
19
Power System Restoration
Particle swarm optimization (PSO) Approach:
•
•
•
•
PSO is one of the Evolutionary Computation techniques.
PSO was originally developed in 1995 by a social-psychologist (James Kennedy) and an
electrical engineer (Russell Eberhart).
PSO emerged from earlier experiments with algorithms that modeled the "flocking behavior"
seen in many species of birds.
PSO consists of a number of particles (possible solutions) moving around in the search space
looking for the best solution.
PSO Model:



vik 1  vik  c1  rand ()1  pbest i  sik  c2  rand () 2  gbest i  sik
sik 1  sik  vik 1
Continuous variables
IF rand () i  S vi 
k 1
i
THEN
s
ELSE
k 1
i
s
= 1,
vik
Binary variables
v

k 1
i
sik 1
vgbest
v pbest
=0
sik
September 25, 2003
EPNES: Intelligent Power Routers
20
Power System Restoration
Total load served increase through the stages.
In each stage all the control and stage variables
were within their limits and the power balance
equations were met.
The restoration path was established and all loads
were served.
Test System and Results:
September 25, 2003
Stage
Restoration Path
Generation
Units and
Transformers
0
1
2
3
4
5
6
7
8
9
10
11
G1 & G2 & G3 &
T1-4 T2-7 T3-9
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Transmission Lines
L4-5
L4-6
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
EPNES: Intelligent Power Routers
L5-7
X
X
X
X
X
X
X
L6-9
X
L7-8
X
X
X
X
X
Loads
L8-9
X
X
X
L5
L6
L8
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
21
De-Centralized Communication &
Control Protocols
• Objectives
– De-centralized System Restoration Algorithm
– Maximize number of high-priority loads restored
• Approach
– Model as Network of IPRs (Graph Model)
– Design Communication Protocols and Routing
messages algorithms
– Design Objective Function
MAX  kR Lk * yk *(  Prk ),   max Prk
•
•
•
•
Prk : Priority of load k , range [1,N], N is the lowest priority
Lk : each of the loads in the system (power required/load)
Yk : Variable decision ( yk = 1 : Restored, yk = 0 : no restored)
R: set of de-energized loads
September 25, 2003
EPNES: Intelligent Power Routers
22
Restoration in Electrical Energy Network
Featuring Intelligent Power Routers (IPRs)
System
Restoration
going
Process
down
Normal
State
Src 1
Link 1
Bus 1
PR 1
Link 4
Table 1. Priority and Realibility
Src 2
Link 2
Src 3
Link 3
Bus 2
PR 2
Link 5
PR
Link
Priority
Reliability
Pr1
1
-
1
4
1
-
2
-
1
3
-
2
5
2
-
6
1
-
4
-
1
5
-
2
7
1
-
6
-
1
8
1
-
Pr2
Link 6
Pr3
Bus 3
PR 3
Link 7
Snk 1
— Normal State Message
— Request Power
September 25, 2003
Bus 4
PR 4
Link 8
Pr4
Snk 2
— Deny Request
— Request Status
— Response Status
— Affirmative Response
EPNES: Intelligent Power Routers
23
Global considerations
• Graphs approach
– Tree approach
• IPR Classification
– Sink IPR, PR, Source IPR
• Link Organization
– Reliability , Priority
September 25, 2003
EPNES: Intelligent Power Routers
24
Risk Assessment
• What do we want to do?
– Measure the change in reliability of the system
when is operated with and without IPRs.
• How to measure it?
– Adequacy
– Security
• Well-Being indices
• Risk Framework
• What influences reliability ?
– Effect on system’s reliability of adding IPRs
September 25, 2003
EPNES: Intelligent Power Routers
25
Well Being indices
• What are they?
• How do they capture
changes in the
network?
September 25, 2003
Example:
two 3 MW units, one 5 MW unit, 2% FOR each
Capacity
Out (MW)
Probability
0
.98×.98×.98
.941192
3
.02×.98×.98 +
.98×.02×.98
.038416
5
.98×.98×.02
.019208
6
.02×.02×.98
.000392
8
.02×.98×.02 +
.98×.02×.02
.000784
11
.02×.02×.02
.000008
EPNES: Intelligent Power Routers
26
Failure mechanism
• We need the IPR
failure probability
– No data available on
IPR’s failure modes
or probability (They
have not being built
yet !)
– Data Routers info
may be useful to
make an
approximation.
Data Router
Comp Hardware
Intelligence
Switch
Power Hardware
• How does it fail?
–Software
–Router
–Switch
September 25, 2003
EPNES: Intelligent Power Routers
27
Validation TestBed:
DC Zonal Electric Distribution System
By: Lida Jáuregui-Rivera, Ph.D. Student
Advisor: Dr. Miguel Vélez-Reyes
September 25, 2003
EPNES: Intelligent Power Routers
28
DCZEDS: Simplified Model
September 25, 2003
EPNES: Intelligent Power Routers
29
Starboard and Port Power Supplies
 3-phase input Voltage : 480-560 V line-line rms
Power Supply Voltages and Currents
 Regulates an output of 500 V dc for loads
up to 15KW
Starboard - Power Supply - Voltage
500
0
-500
0
100
0.1
0.2
0.3
0.4
Starboard - Power Supply - Current
0.5
0.6
0
0.1
0.2
0.3
0.4
Port - Power Supply - Voltage
0.5
0.6
0
0.1
0.2
0.3
0.4
Port - Power Supply - Current
0.5
0.6
0
-100
600
500
400
200
0
100
0
-100
September 25, 2003
0
EPNES: Intelligent
Power0.1 Routers0.2
0.3
Time/sec
0.4
0.5
30
0.6
Zone 1 Subsystem
Components of Zone1
 Two Ship Service Converter
(SSCM).
Modules
 A diode or’ing network
 One Ship Service Inverter Module (SSIM)
with a Load Bank
 The inputs to this subsystem block include
 on/off signals for the two SSCM’s and
the SSIM
 Voltage reference
SSCM’s.
setting
for
the
 The voltage reference setting controls the
output voltage of the SSCM.
September 25, 2003
EPNES: Intelligent Power Routers
31
Zone1 Ship Service Converter
Module
 The converter accepts 500 V dc and regulates
Voltages and Currents Waveforms
the output voltage to 400 dc for loads up to 20
A.
SSCM-stbd Vout
400
200
0
0
0.1
0.2
0.3
SSCM-port Vout
0.4
0.5
0.6
0
0.1
0.2
0.3
SSCM-stbd Iout
0.4
0.5
0.6
0
0.1
0.2
0.4
0.5
0.6
0.4
0.5
400
200
0
60
40
20
0
60
0.3
SSCM-port Iout
40
20
Block Diagram of the
SSCM
September
25,Control
2003
0
0 Power0.1Routers 0.2
EPNES: Intelligent
0.3
Time/sec
32
0.6
Zone1 Ship Services Inverter Module
 Accepts 380 – 440 V dc and Provides a 3-phase AC
voltage (380 – 440 V)
Voltages and Currents Waveforms of
the Three Phase Load
Load bank Voltage
600
400
200
0
-200
-400
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.4
0.45
0.5
Load Bank Current
SSIM Control
Diagram
40
20
0
-20
-40
0.2
September 25, 2003
0.25
EPNES: Intelligent Power Routers
0.3
0.35
Time/sec
33
Zone 2
 Two Ship Service Converter Modules
Subsystem
(SSCM)
 A diode or’ing network
Voltages and Currents Waveforms
Zone2 SSCM-stbd Vout
400
 Motor Controller Module
200
0
0
0.1
0.2
0.3
Zone2 SSCM-port Vout
0.4
0.5
0.6
0
0.1
0.2
0.3
Zone2 SSCM-stbd Iout
0.4
0.5
0.6
0
0.1
0.2
0.3
Zone2 SSCM-port Iout
0.4
0.5
0.6
0.3
0.4
0.5
0.6
400
200
0
6
4
2
0
6
4
2
0
September 25, 2003
0
0.1
EPNES: Intelligent
Power
Routers0.2
34
Inverter Topology of the Motor Controller
 Accepts 300 – 420 V dc. The ouput of the
inverter is connected to a inductio motor
Torque, Speed, Voltages and Currents
Waveforms
Electromagnetic Torque of the Induction Motor
30
20
10
0
0
0.1
0.2
Rotor0.3
Speed
0.4
0.5
0.6
0
0.1
0.2
0.3
0.4
Induction Motor Stator Voltages
0.5
0.6
-100
0
20
0.1
0.2
0.3
0.4
Induction Motor Stator Currents
0.5
0.6
0.1
0.2
0.5
0.6
200
100
0
Block Diagram of the
Drive Control
100
0
0
-20
September 25, 2003
0
EPNES: Intelligent Power Routers
0.3
0.4
35
Zone 3 Components
 Two Ship Service Converter Modules (SSCM)
Output Voltages and Currents Waveforms of the
SSCM’s
 A diode or’ing network
 Constant Power Load Module
Zone3 SSCM-stbd V-out
400
200
0
0
0.1
0.2
0.3
Zone3 SSCM-port
V-out
0.4
0.5
0.6
0
0.1
0.2
0.3
Zone3 SSCM-stbd
i-out
0.4
0.5
0.6
0
0.1
0.2
0.3
Zone3 SSCM-port
i-out
0.4
0.5
0.6
0
0.1
0.2
0.3
Time/sec
0.4
0.5
0.6
400
200
0
3
2
1
0
10
5
0
September 25, 2003
EPNES: Intelligent Power Routers
36
Constant Power Load Module
 The topology is based on a buck converter.
 Accepts 120 – 600 V dc and regulates the
output
Output Voltage and Current Waveforms of the
CPL
voltage to 100 V dc
 The converter is loaded with a 2-Ohm
resistor
CPL Output Voltage
100
80
60
40
20
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.4
0.5
0.6
CPL Output Voltage
50
40
CPL Control Diagram
30
20
10
0
September 25, 2003
0
0.1
EPNES: Intelligent Power Routers
0.2
0.3
Time/sec
37
Simulation of Fault Conditions
Fault in Zone 2 Bus at 0.4 sec. of
operation
Port - Zone1 - Current
100
Starboard - Zone1 - Voltage
1000
50
0
500
-50
0
-100
0
0
0.1
0.2
0.3
0.4
Starboard - Zone2 - Voltage
0.5
0.1
0.2
0.3
Port - Zone2 - Current
0.4
0.5
0.6
0.1
0.2
0.3
Port - Zone3 - Current
0.4
0.5
0.6
0.1
0.2
0.3
Time/sec
0.4
0.5
0.6
100
1000
0
500
0
-100
-200
0
0
0.1
0.2
0.3
0.4
Starboard - Zone3 - Voltage
0.5
0.6
100
1000
50
0
500
-50
0
0
0.1
September 25, 2003
0.2
0.3
Time/sec
0.4
-100
0
EPNES: Intelligent Power Routers
0.5
0.6
38
0.6
Output Voltages and Currents of the
Zone 2 SSCMs
Torque, Speed, Voltages and Currents
of the Induction Motor
Zone2 SSCM-stbd V-out
Electromagnetic Torque of the Induction Motor
40
500
20
0
0
-500
0
500
0.1
0.2
0.3
Zone2 SSCM-port V-out
0.4
0.5
0.6
-20
0
0.1
0.2
0.4
0.5
0.6
-100
0
100
0.1
0.2
0.3
0.4
Induction Motor Stator Voltages
0.5
0.6
0.1
0.2
0.3
0.4
Induction Motor Stator Currents
0.5
0.6
0.1
0.2
0.5
0.6
200
0.3
Rotor Speed
100
0
0
-500
0
10
0.1
0.2
0.3
Zone2 SSCM-stbd i-out
0.4
0.5
0.6
5
0
0
0
0.1
0.2
10
0.3
Zone2 SSCM-port i-out
0.4
0.5
0.6
5
0
-100
0
20
0
0
0.1
0.2
0.3
Time/sec
September 25, 2003
0.4
0.5
0.6
-20
0
EPNES: Intelligent Power Routers
0.3
Time/sec
0.4
39
Final Comments
• We have familiarized ourselves with the
DC Zonal testbed developed by ONR
– Lida Jauregui left UPRM.
– New student started: Noel Figueroa
• Testbed will serve a model for control
system development.
September 25, 2003
EPNES: Intelligent Power Routers
40
What we promised for year 1
• Design of first IPR(v1.0) software module
• Integration of the IPR module into simulation
system or development of the programmatic
interface
• Experimentation with IPR(v1.0)
• Formulation of the risk assessment problem
for IPR controlled system
• Development of economics and ethics
modules (curriculum improvement)
September 25, 2003
EPNES: Intelligent Power Routers
41
Activities for year 2
•
•
•
•
•
•
Disseminate results from iteration 0
Design of alternative IPR control algorithms
Simulations and preliminary reliability assessment
Design of second IPR (v2.0) software module
Evaluation of alternative IPR control algorithms
Use of economics and ethics modules in electrical
engineering courses (use assessment tools)
• Development of short course for non-power
engineeering majors
September 25, 2003
EPNES: Intelligent Power Routers
42
Questions ?
September 25, 2003
EPNES: Intelligent Power Routers
43
Backup Slides
September 25, 2003
EPNES: Intelligent Power Routers
44