Estimating Ship Engine Power for EFDC+ Propeller Wash Simulations
Developing a propeller wash model requires specifying the ship and propeller properties as model input parameters. In our propeller wash blog series, we have shared details of the EFDC+ propeller wash modeling features. This post introduces guidelines for estimating ship engine power specifications, which can be used as the model inputs for EFDC+ propeller wash simulations when the required information is missing or unavailable.
Since ship owners in competitive markets may be reluctant to disclose detailed information about their ships, it is sometimes difficult for modelers to identify the detailed ship characteristics and operational information required for model simulations. Public and commercial databases of Automatic Identification System (AIS) provide only basic ship information such as Maritime Mobile Service Identity (MMSI) number, vessel type, length, beam, gross tonnage, and time track records (location, speed, and heading direction). To overcome the limitation in data availability, DSI devised several approaches to estimating the essential model parameters of ship and propeller properties using public and commercial databases, including PIANC (2015), NYSDOT (2005), MAN (2018), and TugboatInformation.com (2021).
Installed Engine Power
The installed engine power of the main propulsion system is the most important ship characteristic for the propeller wash simulation; it can be estimated based on ship type and dimensions (e.g., length, beam, and deadweight tonnage). This is accomplished by computing the deadweight tonnage using the identified vessel type and length and then estimating the installed engine power using the computed deadweight tonnage.
Figure 1 shows data from PIANC (2015) on the overall length and deadweight tonnage of various types of ships. The deadweight represents the ship’s loading capacity, including bunkers, fuel, lube oils, and other supplies necessary for the ship’s operation. With these datasets, the relationship between ship length L (m) and deadweight tonnage DWT (ton) is identified using the power function regression presented as Equation (1) for all vessels and for each vessel type separately and the coefficient values in Table 1:
DWT=a×L^b (1)
Table 1. Equation (1) regression coefficients for ship length and deadweight tonnage relationship.
| Vessel Type | a | b |
| All Vessels | 0.0021 | 3.10 |
| Tankers | 0.0003 | 3.52 |
| Carriers | 0.0002 | 3.50 |
| Container Ships | 0.0335 | 2.57 |
| General Cargo Ships | 0.0028 | 3.09 |
| Roll-on/roll-off Cargo Ships | 0.0160 | 2.65 |
| Passenger Ships | 0.0037 | 2.97 |
| Fishing Boats | 0.0018 | 3.15 |
PIANC (2015) also includes samples that relate the installed engine power (hp) and deadweight tonnage at the design draft, as shown in Figure 2. A regression analysis defines the installed engine power P as a function of deadweight tonnage DWT for these samples. In this case, the relationship behaves differently when the deadweight tonnage is greater than 50,000 tons as compared to a deadweight tonnage of less than 50,000 tons, so that the different regression functions can be applied to each condition, as follows:
P = \begin{cases} & \, \, \, \,\,\, \, \, \, \, \,0.0326 \times DWT^{1.3167}\, \,\,\, \, \,\, \, \, \, \, \, \, \, \, \,\mathrm{for} \, \, \, \, \, \, DWT < 50\,000 \, \, \mathrm{tons}\\ &64 696\times \mathrm{ln}\left ( DWT \right ) – 650 483\, \, \, \, \, \, \,\mathrm{for}\, \, \, \, \, \, DWT \geqslant 50\,000 \, \, \mathrm{tons} \end{cases} (2)
As a result of this analysis, modelers seeking to estimate missing ship engine power data can use Equation (1) to calculate the deadweight tonnage using the identified vessel type and length. They can then employ Equation (2) to estimate the installed engine power using the previously calculated deadweight tonnage.
While this method is useful for larger vessels, the PIANC (2015) data does not include sufficient information for ships with a length of less than 50 m (see Figure 1). Typically, deadweight tonnage information is not available for such small vessels, so the approach suggested above might not apply to small tugboats, which usually have high engine powers to push or pull a barge. Tugboats may represent a large portion of ship traffic in harbor or bay areas. To fill the data gap, we collected 110 samples of ship dimensions and engine power for tugboats smaller than 50 m from TugboatInformation.com (2021). In Figure 3, the collected tugboat dataset indicates a positive correlation between installed engine power P (hp) and ship beam B (m), which can be expressed as the exponential function regression Equation (3). Therefore, modelers may employ this approach when estimating the installed engine power for such small-sized tugboats.
P = 828.31\times \mathrm{exp} (0.1587\times B) (3)
Applied Power
The propeller wash simulation also requires specifying an applied power for each ship track point, which represents how the ship engine is operated at a given time and location. In practice, the applied power of a moving ship is determined solely by the ship operators’ discretion, based on weather conditions and standard procedures for embarking, disembarking, and sailing. For propeller wash model inputs, modelers can specify the applied power as a fraction of the installed engine power following the general engine operation guidelines by PIANC (2015) regarding vessel types and sailing conditions, as described below.
For regular sailing, ships typically use about 75% of the installed engine power when they operate at the design service speed under normal load and weather conditions. For berthing and departure maneuvering, PIANC (2015) recommends the applied power as about 5% to 15% of the installed engine power for large ships (e.g., tankers, carriers, container ships, and passenger cruises). A value of 5% might be applied for sheltered berth locations with bed protection and no currents (e.g., berthing in harbor basins). A value of 15% is recommended for exposed berth locations with riverine or tidal currents and bed protection (e.g., quay walls along the river). For smaller ships, the applied power for berthing and departure may be 30% to 40% of the installed engine power. Additionally, tugboats can operate at 75% to 100% of the installed power when maneuvering with pushing or pulling a barge.
Further Recommendation
The guidelines presented above are for specifying the installed engine power and applied power factor based on ship dimensions and sailing conditions. In the EFDC+ Propeller Wash White Paper, several different approaches are described for estimating the other essential model inputs, including ship draft, distance between the propeller and AIS antenna, and propeller diameter. These estimation guidelines will be beneficial to modelers when the information needed for propeller wash model inputs is not readily available.
Talk to the experts
References
MAN (2018). Basic Principles of Ship Propulsion. MAN Energy Solutions, Copenhagen, Denmark.
NYSDOT (2005). “Newtown Creek navigation analysis: Kosciuszko Bridge project.” Technical report, New York State Department of Transportation, New York, NY, USA.
PIANC (2015). “Guidelines for protecting berthing structures from scour caused by ships.” Marcom Report 180, Permanent International Association of Navigation Congresses Secrétariat Général, Bruxelles, Belgium.
TugboatInformation.com (2021). Tugboat Information, URL: https://www.tugboatinformation.com. [Online; accessed: 03.01.2021].
