Spectral Co-Design of Ship Rudder Geometry and Heading Autopilot for Energy Efficiency and Manoeuvring Safety
Transactions on Environment, Energy and Earth Sciences · 13.11.2026 · 7 · 158-181
DOI: 10.62051/bgsje749 ↗Kurz gefasst
Fachartikel in der Zeitschrift Transactions on Environment, Energy and Earth Sciences, online erschienen am 13.11.2026. Verfasst von Fuyong Zhou. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Rudders determine the manoeuvring safety of a ship and permanently consume propulsive power through profile drag, weather helm and autopilot action. Because rudder geometry and course-keeping controller are designed separately and the energy cost of steering requires long stochastic simulations, their joint trade-off between energy efficiency and safety has remained unquantified. This paper proposes a spectral co-design method that optimises rudder area, span and section thickness together with the parameters of a PID heading autopilot. A baseline-anchored, stall-aware rudder model preserves the captive-model calibration of the MMG standard method while extrapolating to new geometries with section data from a physics-informed airfoil model. A frequency-domain evaluator combines steady-sailing equilibria in wind and wave drift with closed-loop second moments under gusts, slowly varying drift and first-order wave loads, and obtains the mean steering resistance from a second-order expansion of the surge balance. The constrained three-objective problem—rudder-related power, utilisation of the IMO manoeuvring criteria and heading deviation, subject to adverse-weather course keeping, steering-gear rate, cavitation and stability—is solved for the KVLCC2 tanker over a climate derived from 194,475 NOAA buoy records. The evaluator reproduces nonlinear time-domain statistics with R2 = 0.976–0.996 at an 1100-fold speed-up. Against the original rudder with a conventionally tuned autopilot, the knee co-design reduces the rudder-related power by 15.0% while improving both safety objectives. In every run, co-design outperforms sequential design and geometry optimisation with a fixed autopilot (hypervolume +2.0% and +2.8%), and nonlinear time-domain simulation over the full climate confirms the saving of the knee design (15.4%). A quantile-based robust formulation raises the Monte Carlo reliability of the Pareto designs from 0.77 to 0.91 under hydrodynamic-coefficient uncertainty.
Zitieren
APA
Zhou, F. (2026). Spectral Co-Design of Ship Rudder Geometry and Heading Autopilot for Energy Efficiency and Manoeuvring Safety. Transactions on Environment, Energy and Earth Sciences, 7, 158-181. https://doi.org/10.62051/bgsje749
Vancouver
Zhou F. Spectral Co-Design of Ship Rudder Geometry and Heading Autopilot for Energy Efficiency and Manoeuvring Safety. Transactions on Environment, Energy and Earth Sciences. 2026;7:158-181. doi:10.62051/bgsje749
BibTeX
@article{Zhou2026,
author = {Zhou, Fuyong},
title = {Spectral Co-Design of Ship Rudder Geometry and Heading Autopilot for Energy Efficiency and Manoeuvring Safety},
journal = {Transactions on Environment, Energy and Earth Sciences},
year = {2026},
volume = {7},
pages = {158--181},
doi = {10.62051/bgsje749}
}
EUVIMED