DESIGN OF HYBRID MARINE CONTROL SYSTEMS FOR DYNAMIC ... T. Nguyen.pdf · design of hybrid marine control systems for dynamic positioning in extreme seas and in ice trong dong nguyen

1

3-Jun-13

DESIGN OF HYBRID MARINE CONTROL SYSTEMS FOR DYNAMIC POSITIONING IN EXTREME SEAS

AND IN ICE

Trong Dong Nguyen

3-Jun-13 2

• Introduction

• Control design for different environmental conditions

• Hybrid control from calm to extreme seas

• Hybrid DP in ice

• Switching control between station keeping and transit

Outline

3-Jun-13 3

Introduction – Dynamic positioning

• Keep floating structure in fixed position and

heading as well as to precisely maneuver

predetermined track exclusively by means of

vessel’s propulsion system

INTEGRATED THRUSTER

CONTROL SYSTEM

- DYNAMIC POSITIONING

- POSMOOR

- AUTOSAIL

- OPERATOR CONTROL

SYSTEM

INTEGRATED MONITORING &

CONTROL SYSTEM

- EXTENSION ALARM

- PROCESS CONTROL

POWER GENERATION

& DISTRIBUTION

PROCESS CONTROL

STATION

PROPULSION

WIND SENSORS

VRU

GYRO

BACK-UP

SYSTEMSAFETY SYSTEM

EMERGENCY SHUTDOWN

FIRE & GAS

ENERGY MANAGEMENT

SYSTEM

AZIPOD

INFORMATION MANAGEMENT

REMOTE DIAGNOSTIC

DRILLING DRIVE

SYSTEM

PLANT

NETWORK

CONTROL

NETWORK

FIELDBUS

NETWORK

3-Jun-13 4

Introduction – Thruster-assisted position mooring

• Thruster-assisted position mooring system:

– ATA/TA POSMOOR (DNV)

– PM

• Main objective of PM

– Control position

– Control line tension

• Turret, spread, single point, etc.

Single point mooring (SPM) Submerged turret loading (STL)

3-Jun-13 5

Introduction – Challenges

• Harsh environmental

conditions

– Strong current: Gulf of

Mexico, Northern

England, Southern

Norway, and South Africa

– Long and high wave:

North seas

– Strong wind with gust:

North Atlantic

– Ice: Arctic areas

3-Jun-13 6

• All-year operation

• Vessel Operational Conditions

– Control objective

– Speed

– Environmental condition

Trans

it

Low spe

ed m

aneu

verin

g

Mod

erat

e

10

(m/s)Speed

y 20

Calm

Sta

tion

keep

ing

2

6

4

High

Extre

me

Environment

z

Main function

x

Shuttle tanker,

station keeping

in calm seas

Shuttle tanker,

transit in high

seas

Ferry during transit

at high speed in

calm seas Ferry

maneuvering

at port

Introduction – Challenges

3-Jun-13 7

Introduction – Hybrid concept motivation

• Design of control system for changes of vessel

operational conditions (VOCs)

• One unique nonlinear controller

– complicated or maybe impossible

– difficult to satisfy many control objectives

• Hybrid controller (combine and switch among

different controllers)

– simpler solution

– satisfy different control objectives

3-Jun-13

Introduction – Hybrid concept motivation

• Flight control:

– Gain scheduling (McLean, 1990; Wang and Balakrishnan, 2002; Oosterom and Babuška, 2005)

– Switching in Nonminimum Phase Nonlinear Systems for VSTOL Aircraft (Oishi and Tomlin, 1999 and 2000)

• Vehicle control:

– Local network control (Hunt et al., 1997).

– Hybrid car: Toyota Prius

• Marine control system:

– Hybrid thruster controller (Smogeli et al., 2004).

• Supervisory control: (Hespanha, 2001; Hespanha et al.,

2003).

8

3-Jun-13 9

Marine Control System Structure

(Sørensen, 2005)

Ship 1: Operational management

Local optimization (min-hour)

Plant control

Actuator control

R

eal-

Tim

e N

etw

ork

Off

ice N

etw

ork

Business enterprise/ Fleet management

Ship 2: Ship 3:

Control layers

High level (0.1-5 s)

Low level (0.001-1 s)

Real-Time Control

Office Systems

Fau

lt-T

ole

ran

t C

on

tro

l • Guidance systems

• Optimal setpoint chasing

• Plant diagnostics and condition monitoring

• DP control

• Power management

• Ballast control

• Supply chain management

• Thruster control

• Motor and drive control

Control design

3-Jun-13 10

DP: Control design for different environmental conditions

Total motion =

Measured

position Observer

Estimated

states

0 10 20 30 40 50 -4

-2

0

2

4

Po

sitio

n (

m) Measurement

LF motion

Controller Commanded

force and

moment

Wave frequency motion + Low frequency motion

WF LF

Normal env.

• Compensates LF motion

• Feedforward (Wind)

• PID controller – P: Position

– D: Velocity

– I: drift motion

Harsh env.

• Compensates WF + LF motion

• Feedforward (Wind)

• PID + ABF controller – PID

– Acceleration feedback

Filter WF

or not?

3-Jun-13 11

Mod

erat

e

2

Sta

tion

keep

ing

Calm

Low spe

ed m

aneu

verin

g

10

Speed(m/s)

y 20

6

4

Trans

it

Main function

x

Environment

Extre

me

High

z

DP: Hybrid control from calm to extreme seas

Compensate LF

motion

Output-PID controller

Compensate WF+LF

motion

Output-PID and AFB

controller

3-Jun-13 12

• Hybrid control system – continuous state multi-

controllers, and

– discrete state logics

Controller 4

Controller 3

ep

: switching signal

y Marine

vessel

Environment

u

Model 1

Model 2

Model 3

Supervisor

Model set

y

y

Controller set

Switch

Controller 1

Controller 2

u

u

Model 4

Operator

Supervisory

control

Decision

Supervisor

DP: Hybrid control from calm to extreme seas

3-Jun-13 13

• Models/Estimators

– Input: y, u, w0p

– Output: yp

Controller 4

Controller 3

ep

: switching signal

y Marine

vessel

Environment

u

Model 1

Model 2

Model 3

Supervisor

Model set

y

y

Controller set

Switch

Controller 1

Controller 2

u

u

Model 4

Operator

Supervisory

control

Decision

DP: Switching control from calm to extreme seas

Model i

y

u

w0p

y yp

ep

3-Jun-13 14

• Controllers


Controller 4

Controller 3

ep

: switching signal

y Marine

vessel

Environment

u

Model 1

Model 2

Model 3

Supervisor

Model set

y

y

Controller set

Switch

Controller 1

Controller 2

u

u

Model 4

Operator

Supervisory

control

Decision

compensate low

frequency (LF) motion

compensate LF motion

compensating both

WF and LF motions

Smooth transition of

moderate and extreme

controllers

Calm

M

odera

te

Hig

h

Extr

em

e

Output-PID

Output-PID

Output-PID+AFB

3-Jun-13 15

• Supervisory control:

– automatically switch to

controller designed for that VOC

– scale-independent hysteresis

switching logic: slow down

switching to prevent chattering


yes

start

: arg min

:p

(1 ) ,ph p

no

Do not

switch

Do

switch

Scale-independent hysteresis

switching logic, Hespanha (2001)

Controller 4

Controller 3

ep

: switching signal

y Marine

vessel

Environment

u

Model 1

Model 2

Model 3

Supervisor

Model set

y

y

Controller set

Switch

Controller 1

Controller 2

u

u

Model 4

Operator

Supervisory

control

Decision

3-Jun-13 16

DP: Switching control

• Experimental

results

Calm Varying from calm to extreme Extreme

3-Jun-13 17

Experiment – Set 2

Estimated WPF and switching

signal,

3-Jun-13 18

• Switching control stabilized and improved

performance of DP vessel in harsh environments

• Switching control expanded the weather

operation window

DP: Switching control from calm to extreme seas – Summary

3-Jun-13 19

DP in ice

• Motivation

– Oil demand

– Oil reservation in Arctic area

– Global warming

3-Jun-13 20

DP in ice

• Environmental condition

– Waterdepth = 1200m,

– Ice thickness = 1-3m, 9/10 icecover

– Ice drift speed up to 0.154 m/s

• DP drilling vessel is Vidar Viking

• Experience instability of auto DP mode

• Manual station keeping based on captain’s

experience

Sovetskiy

Soyuz

Oden

Vidar Viking

Deepwater drilling in heavy sea ice

IODP Expedition 302, 2004

3-Jun-13 21

DP in ice

• Challenges for control design

– Good enough model for ice loads slush ice,

broken ice,

pack ice,

level ice,

ridges, and

Iceberg

• Other challenges: remoteness, coldness, darkness.

Control

theory Structure

Ice and

hydro-

dynamic

load

model

Intersection of the different disciplines

3-Jun-13 22

DP in level ice

©2007 by Nguyen

hydrodynamic pressure

shearwave crushing

normal

friction

fracture


flexural

shearwave crushing

fracture

fracture


ventilation

wave

begins

possible contact


shearwave crushing

normal

friction

fracture

possible contact

normal

friction

normal

friction


sliding and submersion

wave

Initial contact

Crushing and

Bending failure

Rotation of

Impact between

plate deflection

broken ice

rotating ice fragment

fragment

and vessel's hull

normal

friction


impact

wave

Sliding & submersion

and new contact

normal

friction

normalfriction

V Vice's velocity vessel's velocity

ice vessel

crushing

bending failure

nextbending failure

unbroken ice sheet

new contact

unbroken ice sheet

broken icefragment

1

2

3

4

5

6

Top view

of icebreaking

3-Jun-13 23

DP in level ice - Modelling

Ice Data: hi, f, c, V,

ice edge before

contact,

Dynamic Structure

Data: Mass, Damping,

Stiffness and Effects

of Control properties,

contact surface.

Contact

Time History of

Dynamic Ice Load and

Vessel’s Response

Vessel

dynamics

Literature Review

Field Investigation

Laboratory Tests

Icebreaker

Drill rig

FPSO

Solve for each

time step

Fi(x, y, tj),

Few(x, y, tj)

Vv-i(tj+1),

Fw(x, y, tj+1)

Ice force vector

Ice edge geometry

and location

Relative velocity

Water line location

Crushing

Bending

failure

Clear away

Contact

Crushing

ice crush bending submersion velocityτ τ τ τ τ

3-Jun-13

DP in level ice – Modelling

F

Advance of vessel in ice0

1

2

3

4

5

6

1: Local failure

2: Global failure

3: Maximum

4: Start of rotation

5: 2nd force peak

6: End of rotation

crushτ

F

0

Simulated ice load

Measured ice load

3-Jun-13


• Time domain simulation

• Discretization of ice edge

and vessel hull

-30 -20 -10 0 10 20 30-40

-30

-20

-10

0

10

20

30

40

50

-10 -8 -6 -4 -2 0 2 4 6 8 1035

40

45

50

3-Jun-13


• Validation

– MESH 6: Comparison between mean simulated

ice load and empirical ice resistance for different

ice thicknesses and ice velocities.

3-Jun-13


• Compare with the empirical formulation

(Lindqvist, 1989)

3-Jun-13

DP in level ice – Controller design

• Control plant model for open water and ice

• Observer design for open water and ice

Without ice-load

measurements With ice-load

measurements

3-Jun-13


• Controller design

3-Jun-13


• Controller design

3-Jun-13

DP in level ice – Simulation results

Open water

controller

Ice controller without ice-

load measurements

Ice controller with ice-load

measurements

Vessel’s position

Open water DP in Ice Open water

3-Jun-13 34

Switching control between station keeping and transit

Onshore tank

Shuttle

tanker

1. Transit: autopilot

2. Approaching: low speed maneuvering

3. Connecting: station keeping using DP

4. Connected:

Tower: SPM using DP Buoy: STL using POSMOOR ATA/TA

Tower or buoy

Offshore field

0. Low speed

maneuvering

from port

• Motivation – Different controllers for different

tasks and different speed regimes

– Time consuming if manually switch

• Objective: – Develop switching control for

integrated station keeping control and transit

– Illustrate feasibility of switching control system via shuttle tanker example

4

Sta

tion

keep

ing

10

20

Speed(m/s)

y

6

Low spe

ed m

aneu

verin

g

Trans

it

Calm

2

Extre

me

High

Environment

z

x

Main function

Mod

erat

e

3-Jun-13 35


Model 1 - Transit

Model 2 - Smooth transformation

Stration keeping

Model 3 - DP

Model 4 - Smooth switch

Model 5 - PM

from transit to station keeping

between DP - PM

Model set

Supervisor

Decided by

Decided by

switching

operator

logic

Observer 1 Controller 1

Controller 2Observer 2


Controller set Switch

Marine vessel

Environment

: switching signal

u

y ep1 p1

y

u

u

y

y

p5eyp5

3-Jun-13 36


– Switching between

DP and PM (STL) Model 1 - Transit

Model 2 - Smooth transformation

Stration keeping

Model 3 - DP

Model 4 - Smooth switch

Model 5 - PM

from transit to station keeping

between DP - PM

Model set

Supervisor

Decided by

Decided by

switching

operator

logic

Observer 1 Controller 1



Controller set Switch

Marine vessel

Environment

: switching signal

u

y ep1 p1

y

u

u

y

y

p5eyp5

Submerged turret loading (STL)

3-Jun-13 37


3-Jun-13

End of Presentation

• List of reference: – Nguyen H. D., Nguyen T. D., Quek S. T. and Sørensen A. J. (2011). Position-moored

drilling vessel in level ice by control of riser end angles. Journal of Cold Regions

Science and Technology, 66(2-3), pp 65–74.

– Nguyen T. D. and Sørensen A. J. (2009). Switching Control for Thruster-Assisted

Position Mooring System. Control Engineering Practice, 17(9), pp. 985–994.

– Nguyen T. D. and Sørensen A. J. (2009). Setpoint chasing for thruster-assisted

position mooring. IEEE Journal of Oceanic Engineering, 34(4), pp 548–558.

– Nguyen T. D., Sørensen A. J. and Quek S. T. (2008). Multi-Operational Hybrid

Controller Structure for Station Keeping and Transit Operations of Marine Vessels.

IEEE Transactions on Control Systems Technology, 16(3), pp. 491–498.

– Nguyen T. D., Sørensen A. J. and Quek S. T. (2007). Design of High Level Hybrid

Controller for Dynamic Positioning from Calm to Extreme Sea Conditions. Automatica,

43(5), pp. 768–785.

– Hassani, V., Sørensen A. J., Pascoal A. M. and Nguyen T. D. (2012). Multiple Model

Adaptive Dynamic Positioning. In Proc. MCMC’2012, 9th IFAC Conference on

Manoeuvring and Control of Marine Craft, Arenzano, Italy, 19-21 September.

– Nguyen, T. D., Sørbø A. H. and Sørensen A. J. (2009). Modelling and Control for

Dynamic Positioned Vessels in Level Ice. In Proceedings of 8th Conference on

Manoeuvring and Control of Marine Craft (MCMC’2009), pp. 229-236, September 16-

18, Guarujá, Brazil.

38

©2007 by Nguyen

Documents

DESIGN OF HYBRID MARINE CONTROL SYSTEMS FOR DYNAMIC ... T. Nguyen.pdf · design of hybrid marine control systems for dynamic positioning in extreme seas and in ice trong dong nguyen