Performance monitoring and simulation refine ship design and reduce fuel costs by optimising propulsion
Naval architects consider operational data from performance monitoring systems and use voyage simulations when designing a new generation of ships. Foreship used data analysis to improve its design for a client, changing the engineroom layout to increase flexibility and energy efficiency.
It uses hindcast weather and reference ship performance databases to support realistic voyage simulations for new ship designs. Measured operational data is used to tune the analytical models being used at the ship design stage, says Foreship senior specialist for ship dynamics Matthew Patey.
The standard procedure for optimising propulsion power is to evaluate the hull performance using reference data from previous designs, use computational fluid dynamics (CFD) to optimise the hull form and validate it with self-propulsion tests in a model basin.
“These are tried and tested ship design methods, but they leave room for uncertainty,” says Mr Patey. “CFD procedures may not capture the flow in sufficient detail and model test procedures suffer from scale effects.” The methods establish whether one hull design is better than another, but there is uncertainty over the extent of the improvement.
“Ultimately, the designer estimates the required propulsion power based on standard model test procedures at constant speed in calm water then adds a 15% sea margin to account for the weather, making an educated guess on wind resistance,” Mr Patey continues.
“It is up to the hydrodynamicist to balance the propulsion power estimate with commercial pressures to optimise the hull shape for minimum engine power.”
A ship’s propulsion solution, engine arrangement and related systems should consider many more factors, including marine growth on the hull, hotel load requirements, planned and unplanned engine maintenance, energy saving devices (waste heat recovery and air lubrication systems) and the logistics of shipping operations.
“Assessing electrical power demand for a diesel electric-powered ship is therefore no simple task,” says Mr Patey.
Foreship offered the opportunity to use full-scale operational data when it approached an owner with a new way of calculating fuel consumption as part of a newbuilding project.
“The subsequent work supported the decision to make a significant change in the ship’s powering solution,” says Mr Patey.
The performance data available from the owner’s ships was used to calibrate fuel consumption and propulsion power models. This was by “measuring specific voyages using their reference vessel and then applying correction factors to the new design” he explains, by accounting for contributions made by energy efficiency devices.
“We were primarily interested in the required propulsion power and fuel consumption,” Mr Patey explains.
Foreship used six months of operational data to measure actual fuel consumption and develop a daily average power profile of the service load for the reference ship in operation.
This was compared with 36 individual port-to-port voyage simulations. “Deriving correction factors started with simulating a section of the measured voyages, during which the speed was more or less constant and using hindcast weather data over a defined timespan,” says Mr Patey.
“The specific fuel oil consumption curves were corrected by a factor provided by the owner based on its own experience and marine growth rates applied based on the owner’s own data.”
Propulsion power was calibrated by repeated adjustment of the baseline speed-power curve to maximise the correlation between the measured average propulsion power on each voyage and the simulated propulsion power.
“This approach achieved a 98% correlation between predicted and measured total fuel consumption and 96% between predicted and measured average propulsion power, which was close enough,” explains Mr Patey.
The tweaked operational speed-power curve was then compared to the design speed-power curve for the reference vessel to develop a set of correction factors to apply to the new vessel.
“The final step in making the scenario realistic involved incorporating the owner’s statistics on engine availability,” says Mr Patey. Planned maintenance was reflected by removing one engine from consideration on a regular basis for a single port-to-port voyage.
Unplanned maintenance was simulated using a random number generator and giving the probability of failure to remove an engine for a port-to-port voyage.
The reference vessel in operation and the new design were then run through a series of voyage simulations covering four round-trip itineraries in four different sea areas. Simulations were run over a 25-year period for which hindcast weather data was available. The information generated from simulations included:
One of the main conclusions from the process was that the proposed engines for the new vessel would feature insufficient total installed power, if the current drydocking and hull cleaning schedule were used.
“The simulations resulted in several late arrivals on one of the itineraries, but not for the reference vessel,” says Mr Patey, adding this was in circumstances where high levels of marine growth were factored in or when one or more engines were out of service and marine growth was moderate or high.
“This was despite the installed engine power being well in excess of the required power according to the design’s speed-power curve, the expected service load and the traditional 15% sea margin,” he says.
In addition, the engines proposed offered less potential to optimise fuel consumption because all the engines were the same in the new ship.
“The reference vessel featured two engine sizes allowing more flexibility to adapt to variations in the power demand encountered,” says Mr Patey. “Further simulations of the new vessel with half of the engines replaced with larger engines showed improved fuel efficiency, lower EEOI values and fewer late arrivals.”
This analysis led to Foreship increasing the installed power and changing the size and configuration of the engines used. “Most significantly, it demonstrated the importance in the design stage of considering operational data from performance monitoring systems and using voyage simulations to reflect real future operations of the vessel,” Mr Patey concludes.
EEOI is the total carbon emissions in a time period per unit of revenue tonne-miles