Integrating VTB

1
Integrating VTB JADE Agent Platform using
CORBA for Reconfiguration of Shipboard Power
System (SPS)

• Gilberto Morejon , Sanjeev Srivastava,
• Dave Cartes, Andron Creary

Center for Advanced Power Systems Florida State
University Tallahassee, Florida
2
Outline

• Objectives
• MBR Agents for SPS Reconfiguration in VTB
• CORBA Interface
• Test Case
• Conclusions Future Works
• Work being done at CAPS, FSU

3
Objectives

• Autonomous control for reconfiguration of
  shipboard power system
• Efficient control based on limited local
  information
• Truly decentralized system to avoid single point
  of failure
• Reduced human operation to avoid human mistakes

4
Project Tasks

• Development of MBR algorithm for reconfiguration
  of SPS
• Software implementation of Multi-agent system
  incorporating MBR
• PIL implementation with VTB
• Agent implementation on hand held PDA (HP-iPAQ)
• Development of CORBA interface
• Implementation of agents on iPAQ-CORBA
  interface-VTB
• Development of mesh algorithm for MAS
• Implementation of mesh algorithm for MAS
• Comprehensive testing

5
MBR Agents for SPS Reconfiguration in VTB
Agent layer
Power system layer
6
MBR Agents for SPS Reconfiguration in VTB
An MBR Agent Its State Variables

• In each interface
• contract
• Power flowing through the interface
• in/out Cost-load curve
• Budget_(in/out)
• Available power to/from the neighbor
• Cost_(in/out)
• Cost for the power to/from the neighbor
• nvload_(in/out)
• Total non-vital loads on both sides of the
  interface

7
MBR Agents for SPS Reconfiguration in VTB
CORBA Interface
Agent 1
Algorithm
LEAP JADE JAVA
Wireless communication
Agent 2
VTB
Algorithm
LEAP JADE JAVA
Shipboard Power System model
Agent 3
CORBA objects
Algorithm
LEAP JADE JAVA
Agent 4
Algorithm
LEAP JADE JAVA
Agent 5
Algorithm
LEAP JADE JAVA
8
CORBA Interface Diagram
JacORB Java
OmniORB C
IIOP
IDL Compilers
JacIDL
OmniIDL
IDL Stubs
IDL Skeleton
Interface Description
9
VTB Interface
Retrieve model list
Register COSNaming entry
Nameserver OmniNames
VTB Model
Attempt to establish connection
Validate compatibility, lock connection
Register callbacks to input lines
JADE Agent
10
Test Case

• Test System
• 4 generators
• 15 CBs
• 8 Loads
• 4 vital
• 4 non-vital

11
Test Case

• Multi-Agent System
• 31 agents
• 15 CB agents
• 4 Generator agents
• 8 Load agents
• 4 Bus agents
• 25 agents on a single computer
• 6 agents on HP-iPAQs

12
Experimental setup
13
Experimental setup
14
Connecting to VTB
15
Test Case
20 MW
25 MW
Vital loads
Non-Vital loads
30 MW
25MW

• Total generation 100 MW
• Total load 85 MW

• 30 MW
• Initially not turned on

16
Test Case
Non-vital loads shed Total load shed 23 MW
20 MW
25 MW
30 MW
25MW

• 30 MW
• turned on

• Total generation 100 MW
• Total load 90 MW

17
Conclusion Future Work

• CORBA objects to facilitate communication between
  agents and VTB simulation were developed
• Developed MBR algorithm for reconfiguration was
  validated using VTB
• Test cases show effective performance of VTB
• Future works
• Communication speed needs to be improved
• Implementation of mesh algorithm for MAS

18
Publications

• L. Sun, G. Morejon, and D. Cartes, Interfacing
  Software Agents with Power System Simulations,
  IASME Transactions, Vol. 1, No. 2, pp. 312-317,
  Apr. 2004.
• Gilberto Morejon, Li-Hsiang Sun, David A. Cartes,
  CORBA implementation for interfacing software
  agents with virtual test bed, Reconfiguration
  Survivability Symposium 2005, 16-18 Feb. 2005.
• S. K. Srivastava, G. Morejon, and D. A. Cartes,
  Agents for automated reconfiguration of
  Shipboard Power System using CORBA interface and
  virtual test bed, journal paper in progress

19
Reconfigurable Power ConversionL. Qian, D.
Cartes, and A. Creary

• Active Filter
• Non-linear loads generate current harmonics in
  shipboard power systems resulting in the
  deterioration of power quality. Certain loads may
  be sensitive to the changes in power system. This
  simulation solves this problem with the use of a
  re-configurable induction motor drive and minimal
  equipment that requires no additional space.

20
Simulation Results

• Active Filter Results
• VTB and PSIM comparison.
• The graphs from PSIM and VTB system response
  follows and the results are identical.
• VTB Simulation PSIM Simulation

21
Simulation Results

• Active Filter Results
• VTB Graph PSIM
  Graph

22
Publications

• Lewei Qian, David Cartes and Hui Li, A Current
  Harmonic Selective Active Filter with An Adaptive
  Method (Submitted to IEEE Transaction on
  Circuits and Systems II)
• Lewei Qian, David Cartes, etc., A Reconfigurable
  Induction Motor Drive with Harmonic Cancellation
  Feature, IEEE Electric Ship Technologies
  Symposium 2005, ESTS 2005
• Wei Ren, Lewei Qian,, etc, Real Time Digital
  Simulations Augmenting the Development of
  Functional Reconfiguration of PEBB and Universal
  Controller, accepted by 24th American Control
  Conference, ACC 2005.
• Lewei Qian, David Cartes and Hui Li, An Adaptive
  Method for Harmonic Selective Shunt Active Filter
  Control, Reconfiguration Survivability
  Symposium 2005, 16-18 Feb. 2005.

23
Fault Detection and DiagnosisL. Liu, D. Cartes,
and A. Creary

• Fault Identification
• This simulation uses a mathematical model for
  simulation of faults in a non-salient pole
  permanent magnet synchronous motor system (PMSM).
  The model can be used to continuously simulate
  the transients of a PMSM under different
  operating conditions, which will support the
  design of PMSM fault detection and diagnosis.

24
Fault Detection and Diagnosis

• Fault Identification
• VTB and Simulink result is identical

25
Simulation Results

• Fault Identification
• VTB Graph Simulink Graph

26
Publications

• Li Liu, David A. Cartes, Adaptive Parameter
  Identification of Nonlinear Permanent Magnet
  Synchronous Motors, submitted to Automatica for
  publication.
• Li Liu, David A. Cartes, "New Permanent Magnet
  Synchronous Motor Modeling Approach for Fault
  Diagnosis Purposes", submitted to ASNE Electric
  Machinery Technology Symposium 2006.
• Li Liu, David A. Cartes, A Particle Swarm
  Optimization Approach for Automatic Diagnosis of
  PMSM Stator Fault, submitted to American Control
  Conference, June 2006.
• Li Liu, Kevin P. Logan, David A. Cartes,  Fault
  Detection, Diagnostics, and Prognostics Software
  Agent Solutions, IEEE Electric Ship Technologies
  Symposium (ESTS 2005).
• Li Liu, David A. Cartes, On-line Identification
  and Robust Fault Diagnosis for Nonlinear PMSM
  Drives, American Control Conference, June 2005.
• Li Liu, David A. Cartes, Robust Fault Diagnosis
  for Electric Ship Propulsion Motors, ASNE
  Intelligent Ship Symposium VI, June 2005.
• Li Liu, David A. Cartes, Robust Fault Diagnosis
  with Parameter Adaptation Applied to Propulsion
  PMSM Motor Drives, ASNE Advanced Naval
  Propulsion Symposium, Nov. 2004.