A High Fidelity Integrated System Model for Marine Power Systems Jeroen Schuddebeurs Advanced Marine

1
A High Fidelity Integrated System Model for
Marine Power Systems Jeroen SchuddebeursAdva
nced Marine Electrical Propulsion Systems
(AMEPS)Consortium EPSRC funded research
projectUniversity of Strathclyde, UKUniversity
of Manchester, UKCranfield University,
UKIEEE Systems Conference, Montreal, April
2008
2
Presentation Overview

• Integrated Full Electrical Propulsion (IFEP)
• AMEPS (Advanced Marine Electrical Propulsion
  Systems) project
• Key challenges for modelling multi-domain systems
• Multiple time constants
• Validation
• Case study loss of propulsion load
• Summary

3
Integrated Full Electric PropulsionBackground
and advantages

• Common power supply for both the ships propulsion
  and hotel loads

Traditional
IFEP

• Some benefits
• Increased overall efficiency during part load
  operation
• Flexibility in machinery allocation
• Increased manoeuvrability
• Increased redundancy

PM
PM
Hotel load
PM Prime Mover
Propulsion
M Propulsion Motor
Hotel load
4
Integrated Full Electric Propulsion Typical IFEP
applications

• Commercial

• Military

5
Integrated Full Electric Propulsion
Disadvantages

• IFEP can be characterised by

• Power dense system
• Geographically small area with relatively large
  generation and load powers
• Could be up to 100 MW (e.g. QM2 Cunard Line)
• Relatively large electric machines
• Generator and motors
• Limited inertia and impedance
• Small cable lengths limited impedance

• Implications
• Disturbances can easily propagate across the
  entire system (crossing domain boundaries)

6
Integrated Full Electric PropulsionSystem level
approach
Generators and Electrical Power Distribution
System Electrical
Prime Movers Mechanical, Thermal
Electric Drives, and Propulsion
Motors Electrical, Mechanical
Propellers Mechanical
M
Characterising and understanding these
interactions require a systems level modelling
approach
7
Advanced Marine Electrical Propulsion
SystemsBackground AMEPS Consortium
Funded by the Engineering and Physical Sciences
Research Council (EPSRC)
Generators and Electrical Power Distribution
System
Electric Drives, Propulsion Motors and Propellers
Gas turbines
M
8
Advanced Marine Electrical Propulsion Systems
Type 45 based model

• Type 45 UK destroyer
• 2x WR21 GTA of 21 MW
• 2x W12V200 DG of 2 MW
• 2x AIM of 20MW
• 4160VAC MV voltage

9
Advanced Marine Electrical Propulsion Systems
Type 45 based model
Gas turbine

• Software packages
• Matlab/Simulink (SPS)
• Electrical network
• Electric drive
• Propeller
• Fortran
• Gas turbine

Passive MV filter
Fixed MV load
Electric drive
Fixed LV load
10
Key Challenges for modelling multi-domain systems

• There are a number of challenges for modelling
  complex
• multi - domain systems, some of these are listed
  below
• Multiple of time constants
• Causality
• Validation

11
Key Challenges for modelling multi-domain
systemsMultiple time constants

• Problem
• In a multi-domain model evaluated at a single
  rate, the slower time constant sub-models are
  forced to run at the rate of the fastest time
  constant system over-computation

• Solution
• Model abstraction
• Multi-rate simulation

12
Key Challenges for modelling multi-domain systems
Multiple time constants multi time steps
Generators and Electrical Power Distribution
System
Electric Drives, Propulsion Motors and Propellers
Gas turbines
M
13
Key Challenges for modelling multi-domain systems
Multiple time constants multi-rate simulation
Tfast
Tslow
t1
Tslow nTfast
tn
t1
tn
t1
14
Key Challenges for modelling multi-domain
systemsMultiple time constants multi-rate
simulation efficiency

• Electrical network and gas turbine model
• These simulations were performed on a Pentium 4

15
Key Challenges for modelling multi-domain
systemsMultiple time constants multi-rate
simulation error
latching effect
Error
Tfast
t
Tslow
What is the risk?

• Compensating actions
• Inverter control

16
Key Challenges for modelling multi-domain
systemsMultiple time constants multi-rate
simulation error
17
Key Challenges for modelling multi-domain systems
Validation

• Validation
• Need to know whether the model is correct for the
  purpose of the study

• Validation methods
• Face validation expert input
• Comparison of other models compare against
  other validated models
• Predictive validation compare against
  experimental or field data

18
Key Challenges for modelling multi-domain
systemsValidation
Face validation Overall system
Predictive validation Multi-phase motor (test
rig) Electric network (test rig)
Comparison of other models Propeller, electric
network and electric drive
19
Key Challenges for modelling multi-domain systems
Validation
Why is validation of a multi-domain system a
problem?

• Field data (predictive validation) not always
  available (confidentiality, future system, etc.)
• Multi-domain test rigs (predictive validation)
• Expensive
• Number of configurations are limited
• Subsystem test rigs (predictive validation)
• Multi-domain interactions not captured

Currently face validation is considered to be the
best method
20
Case study loss of propulsion load Introduction

• Using the AMEPS model
• Vessel speed 10m/s 19.4 knots
• Single-spool and two-spool gas turbine
• Fixed LV load 0.58MVA, pf0.86
• Fixed MV load2.3MVA, pf0.86
• At 2s, the propulsion load of 4MW is
  disconnected

21
Case study loss of propulsion load Introduction
Generators and Electrical Power Distribution
System
Electric Drives, Propulsion Motors and Propellers
Gas turbines
M
T
t
22
Case study loss of propulsion load Results
Steady state 95-105
Transient 90-110
Initialisation
Saturation
23
Summary

• IFEP offers potential benefits
• The very nature of IFEP requires a system level
  modelling approach
• There are a number of modelling challenges
• Multi-rate seems to improve computational
  efficiency

24
Thank you for your attentionAny Questions