DESIGN AND CONTROL SESSION. A General Approach for DP Weathervane Control

Transcription

1 Return to Session Menu DYNAMIC POSITIONING CONFERENCE October 9-1, 212 DESIGN AND CONTROL SESSION A General Approach for DP Weathervane Control Michel R. Miyazaki, Eduardo A. Tannuri University of São Paulo

2 A General Approach for DP Weathervane Control Michel Rejani Miyazaki Numerical Offshore Tank (TPN) University of São Paulo São Paulo, SP, Brazil Eduardo Aoun Tannuri Numerical Offshore Tank (TPN) University of São Paulo São Paulo, SP, Brazil October 212 1

3 Introduction to Weathervane Weathervane is an instrument for showing the direction of the wind It is based on the principle that a body with a fore-end pivot aligns with the wind incident direction 2

4 Introduction to Weathervane This principle can be extended to a FPSO moored by a PASSIVE turret system at the bow This angle garantees minimum mooring forces 3

5 Introduction to Weathervane Curr Turret at bow Current=1.6m/s 16 from N ~6ton ~45ton Final Position 4

6 Introduction to Weathervane Turret at bow Curr Weathervane Basic Principle Stabilizing forces Mooring reaction forces Longit. Force Lateral Force Stabilizing Resulting Binary 5

7 Introduction to Weathervane Turret at bow Curr, Weathervane Basic Principle Stabilizing forces Longit. Force Final Stable Angle 6

8 Introduction to Weathervane However, er if the pivot point (Turret) is astern from the force application point, the equilibrium is not aligned with the environment Curr, 7

9 Introduction to Weathervane However, er if the pivot point (Turret) is astern from the force application Lateral M i point, the Force equilibrium is not aligned with the environment Curr, Mooring reaction forces Unstabilizing Resulting Binary Turret Point Force Application Point 8

10 Introduction to Weathervane Curr, Turret at mid-ship Current=1.6m/s 16 from N The non-weathervane angle leads to high loads in the mooring system ~45ton Final Position 9

11 Introduction to Weathervane The same principle p can be extended to a ACTIVE DP vessel Passive Turret Mooring with weathervane capacity Active DP Shuttle with weathervane capacity 1

12 Introduction to Weathervane Simulation of a environmental Direction variation Control Reference Point Close to the bow artificial Pivot Point Active DP Shuttle with weathervane capacity 11

13 Introduction to Weathervane This idea was introduced by J.A. Pinkster and U. Nienhuis (OTC, 1986) They proposed only one propeller at the bow, and to control a reference point close to the forward end of the vessel No heading control! Maybe dangerous Unstable behavior for reference control point close to the midship 12

14 Introduction to Weathervane No heading control! Maybe dangerous Green safe zone Environmental change From NE to S Weathervane control must be switched off by the DPO 13

15 Introduction to Weathervane Solution - Manual Weathervane control based on DPO Experience 16 Thrust (tonf) Before After Mean BowTunnel BowAzi SternAzi SternTunnel Main Prop. Actual Operation 29 (Brazilian Basin) 14

16 Introduction to Weathervane T.I. Fossen and J.P. Strand (Automatica, 21) They proposed the Wheather Optimal Heading Control (WOHC) ABB Patented Control of a forward bow point Circle reference point is changed to keep position of vessel reference point 15

17 Zero Power Control Weathervane This work presents a novel methodology for weathervane control. It is based on the Zero Power Control Technique Originally developed for Maglev Magnetic Levitation 16

18 Zero Power Control Traditional control: Monitoring the system output SP + - Controller Plant Y Zero Power Control: Monitoring the controller output and the system output K + + SP 1/s + - Controller Plant Y 17

19 Zero Power Control Weathervane How it works? Control Action is used to change the position, seeking the minimum consumption 18

20 Zero Power Control Weathervane Application to DP System: Sway Control Force is used to change the Heading Set-point Y SP + - Y Controller + + Y Dynamics Y -K/s Y Env Forces Ψ SP Ψ Controller + + Ψ Dynamics Ψ Ψ Env Forces 19

21 Zero Power Control Weathervane U=1m/s U=1m/s U=1m/s U=1m/s DP Total DP Total DP Total Force Y Force Y Force Y (-4ton) (-4ton) (-27ton) DP Total Force Y (-11ton) Current YForce (+4ton) Current YForce (+4ton) Current Y Force (+27ton) Current Y Force (+11ton) Start: Ψ =3 o Ψ SP = 3 o After t 1 : Ψ =3 o Ψ SP = 2 o After t 2 : Ψ =21 o Ψ SP = 1 o After t 3 : Ψ =11 o Ψ SP = o Speed of set-point change depends on K1 2

22 Zero Power Control Current Wind x Stable 2 equilibriums y Bow incidence is stalble Stern Incidence is unstable Unstable 21

23 Advantages of ZPC Weathervane control Can control any reference point Heading is controlled, so limits can be imposed All horizontal plane controllers maintain the same control structure and same tuning 22

24 Numerical Results 23

25 Numerical Results Property value Total length (m) Length entre perpendiculars L PP (m) Beam B (m) 3.48 Draft T (m) 5.18 Mass (ton) 1792 Moment of Inertia I Z (ton.m 2 ) DP Barge with 6 azimuth thursters Numerical Offshore Tank TPN Simulator Time Domain Simulator 24

26 Condition 1 - Aligned Final position.7m/s 9m/s Hs=2m Tp=8s.7m/s 9m/s Hs=2m Tp=8s Final position Initial position Initial position FY [kn] time [s] FY [kn] time [s] Midship control Side-point control 25

27 Simulation: Brazilian Typical Environment #1.7m/s 9m/s 5 Final position Pos X [m] Hs=2m Tp=8s Pos Y [m] time [s] Initial position Yaw SP[º] 4 2 Yaw Set-point [º] Yaw angle [º] FY [kn] time [s] time [s] 26

28 Experimental Results 27

29 Vessel Properties Main Propeller Fmax=+.617N Fmin=-.276M X=-132mm Stern Fmax/min=+/-.93N X=-757mm 1:125 Reduced Scale DP Tanker Bow Thruster Fmax/min=+/-.125N X=966mm Property Real value Model Vl Value Total Length 277m 2.22m Length between Perpendiculars [L PP ] 262m 2.1m Beam [B] 46m.37m Draft [T] 8m.64m Mass [M] 8,617 ton 41.3 kg Inertia [I z ] 4.85 x 1 7 ton m kg m 2 28

30 Laboratory Facilities Wind Generation System: Adjustable Angle Adjustable Speed Wave Generation System: Adjustable Frequency Adjustable Height 29

31 Variable wind 3

32 Comparison: Midship control Simulation Experiment Y (m) 1 5 Final position Y (m) 1 5 Final position Wind 22m/s -5-5 Pos X [m] Pos Y Initial position X (m) Simulation Experiment Simulation Experiment time [s] g Angle [deg] Heading Initial position X (m) Simulation Experiment time [s] y Force [N] Sway x 1 5 Simulation forces Measured Experiment Forces time [s] 31

33 Comparison: Port side control Simulation Experiment Y (m) 5-5 Final position Y (m) 5-5 Final position Wind 16m/s Initial position X (m) Initial position X (m) Pos X [m] Pos Y 1 5 Simulation Experiment Simulation Experiment time [s] Heading Angle [deg] x Simulation Experiment Simulation Experiment time [s] time [s] y Force [N] Sway

34 Comparison: Stern control Simulation Experiment Y (m) -5 Y (m) -5 Wind X (m) X (m) Pos Y Pos X [m] x Simulation Experiment Simulation Experiment g Angle [deg] Heading Simulation Experiment y Force [N] Sway Simulation Experiment time [s] time [s] time [s] 33

35 Experiments Summary 1 Final position 5 15 Ahead point control 1 Y (m) -5 Wind direction Initial position 5-5 Initial position Final position -1 Midship control -1 Wind direction X (m) X (m) 15 1 Final position 1 5 Final position Y (m) Wind direction Initial position Lateral point control -5-1 Wind direction Initial position Stern point control X (m) X (m) 34

36 Midship control 35

37 Bow control 36

38 Port side control 37

39 Stern control 38

40 Experiments Summary 3.5 x x Midship control Ahead point control ( T 2 ) Time(s) Time(s) 3.5 x x 16 ( T 2 ) Lateral point control 3 Stern point control Time(s) Time(s) 39

41 Analysis For the stern control point, the large heading oscillation is responsible for thepoweramplification.thefigureshowsthattheaveragedpswayforce reduces after the ZPC-W is turned on. So, the weathervane action is working properly, and the oscillation problems may be mitigated by a fine tuning of the control gains. DP Force (kn) 1 5 Sway Total Yaw Angle ( o ) Time(s) 4

42 CONCLUSIONS Satisfactory results so far, decreasing DP utilization (Except stern control) Successfully controlled the vessel bow, midship and port side and variable environment condition More studies need to be conducted, but the results of this paper shows that this controller is commercially viable, with great advantages over simpler weathervane controllers, specially for offshore applications 41

43 FUTURE WORK More simulations Deeper stability analysis Real time simulation Implementation TPN-USP Real Time Simulator 42

44 Acknowledgments: THANKS 43