Indirect load case estimation for propeller-ice moments from shaft line torque measurements

In order to estimate the forces exerted on ship propellers during ice navigation, the rotational dynamics of the propulsion system need to be accurately modelled. The blade measurements of ice loads on the propellers of ships during ice navigation is challenged by the harsh operating environment. Shaft line measurements are therefore performed inboard, and the required propeller loads are subsequently estimated through the use of a dynamic model and the solution of an inverse problem. The inverse problem is mathematically ill-posed and requires the determination of the ice-induced load on the propeller blade from shaft line measurements. The present study investigated full-scale torsional responses on the shaft line of a polar supply and research vessel during navigation through sea ice on a 68-day voyage between Cape Town and Antarctica. The vessel spent almost 11 days in ice with observed concentrations above 90% and a maximum thickness of 3 m. The aim was to evaluate the extreme ice-induced moments on the shaft line and thereby determine how sparsely published operational loadings compare to the design loads of an ice-going vessel. Ice-induced moments on the propeller were obtained from operational measurements through three previously published approaches to solving the ill-conditioned inverse problem. The regularization methods used included truncated Singular Value Decomposition, truncated Generalized Singular Value Decomposition and Tikhonov regularization. The maximum ice-induced external moment was found to be 941.5 kNm, which was just within the maximum allowed ice-induced torque on the propeller. The duration of ice impacts on the propeller ranged from 25 to 228 ms. A secondary peak was evident in torsional responses obtained from propeller-ice impacts which is thought to be a shear stress wave that propagates and reflects back in the shaft line. From the inversely determined ice-induced loads, the number of impacts, the duration, the shape and the damping of water on the propeller was identifiable. The results obtained were physically reasonable, indicating that the current methods are suitable for obtaining ice-induced loading on the propeller from shaft line measurements.