Case Study – Force off Position

1 Overview

A DP Class 3 vessel was conducting pipelay operations in the high North Sea area.  Weather conditions were deteriorating quickly and in accordance with the ASOG, it was decided to suspend operational activities and remove all connection with the seabed and recover equipment.

Problems with the mission equipment meant that there were delays in recovering equipment and laying down the pipe for abandonment.  The vessel was being hit by ever increasing large waves.  At half past midnight, the vessel was hit by a series of large waves creating a 36m excursion from position.  This change in position caused the DP system to reject all DP reference systems as they passed the prediction error rejection limit.  This “limit” allowed for a 5m deviance between the estimated (model) position of the DP system and the position derived from the reference systems.  All DP positions were rejected from DP Control.

The DPO on the bridge observed a total loss of reference systems as well as, critical alarms for breach of the red watch circle, the violent nature of the vessel motion the decision was made to abandon the pipe to the seabed.

2 Events leading up to the incident

Several hours earlier, the weather reports received on the vessel bridge forecast advised adverse weather inclusive of wind speeds exceeding 50 knots which required the vessel to go in to “yellow DP alert mode” according to the ASOG.

At this time the DP system was set up with three position reference systems online:

• x2 DGNSS’s which includes inertial aided navigation and, 1 additional DGNSS which was in monitoring mode only.
• 1 HPR operating in Long Base Line (LBL) mode with three transponders on the seabed.

A red watch circle was set to 35m (ASOG states max 36m), and a yellow watch circle set to 15m.  The “Reduced GPS weight” setting was not chosen, and all reference systems appeared to have equal weighting.

HPR2 was using gyro 3 as heading reference and MRU3 as pitch and roll reference.  The DP system was using MRU1 as pitch and roll reference.

The power system was operating in two split modes with all six generators and all six thrusters online.  The environment was such that thruster high force and switchboard power limits alarms were being activated and some being issued by the thruster drives themselves rather than the PMS.

Acceptance limits for prediction error was set to “normal”, which is 5m.  This mode is recommended by the OEM for “rough” weather conditions.  This was also the recommendation stated within the vessels DP Operations Manual.

The vessel was positioned to take the current straight on the bow the wind coming from approximately 20 degrees port bow.  The water depth at the location was 105m.

The weather peak was forecasted around midnight, and the vessel should, according to their time plan, be disconnected and on standby around 22.00hrs.  This left them over 6hrs to be in a safe condition.  This was deemed sufficient time.

Work continued until they experienced problems with the A&R Winch laydown system which caused a delay in the operation.

New assessments of the latest weather forecasts were analysed.  However, no significant change was anticipated.

All of the problems experienced delayed the operation substantially and major work was required in order to get the A&R winch in service.

In the time immediately preceding the event, there were a number of alarms relating to high utilization of the thrusters.  The environmental forces were increasing and in line with the received forecasts which suggested the consequence analysis alarm was triggered.

The DPO described a series of large waves just ahead of the incident.  The investigation analysed the raw data from the GNSS logs and with the help of OEMs found that this wave, or series of waves, made the vessel move around 30m, causing the vessel to exceed the prediction error rejection limit.

2 What can be concluded?

• Time to Terminate – Difficulties in terminating the operation caused the vessel to still be connected for too long into the weather event.
• Environment – The environment was above the vessels DP Position Keeping capabilities.  The vessel was already utilizing a lot of its power and propulsion capabilities to counteract the existing environmental forces.
• Sudden high amplitude\long period waves – This caused the excursion.  The DP systems then had to combat additional forces presented by the sudden increase in wave action.
• Prediction error setting – Had the setting for prediction error acceptance limit been set to “wide”, the deviation between DP model and the average position calculated based on the PRSs may not have caused the system to reject the reference systems.  However, the force off was so great and fast that even set in ‘wide’ the same results could have been realised.  The forces affecting the vessel created a position excursion that was so wide and quick that it exceeded the set limit between predicted position and actual “average position” which is calculated based on input from all PRS.
• A train wave resulted in the vessel being forced off position with a high speed and resulting in activation of the prediction error rejection of the PRS.  The DPO should be familiar with the recovery action, which is OEM and DP control system generation specific.

3 Root Cause

• At the time of the incident, the vessel was being operated beyond its DP Position Keeping Capabilities.  An experienced DPO knows that in case of wave trains, the vessel can be forced off and activate the prediction error and should know the recovery procedure.
• Inadequate procedure for description of settings of rejection limit.  The recommendation from the vendor is not specific enough when describing weather criteria for settings.  In this case the “normal” setting did not allow enough deviation to avoid losing reference systems input to the DP system.  Even though the weather at the time could be described as rough.
• Inadequate plan for securing the pipe before weather peak.  The vessel had planned for what they believed to be enough time to lay down and be disconnected before the weather peak.  In this specific case, the plan was foiled due to unforeseen problems.  This was difficult to plan for.  Correct maintenance and regular function test of mission equipment would have reduced probability of such misfunction.

4 Guidance that would be relevant

The following IMCA Guidance would be relevant to this case study:

• IMCA M220 – Guidance on operational planning
• IMCA M252 – Guidance on position reference systems and sensors for DP Operations
• IMCA M242 – Guidance on satellite-based positioning systems for offshore applications