A laboratory-based verification rig for inverse models of ice-induced propeller moments on ship propulsion shafts

Date
2023-03
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH SUMMARY: The rise in maritime traffic in icy polar regions, create hazardous conditions for ships that may require costly and time-consuming in-port repairs. The drive towards environmentally friendly, energy-efficient ships pressure ship builders to conform to stringent design guidelines, set by classification societies. Improving efficiency and production cost is often coupled with a reduction in over-designed components. As a result, there is a critical need for openly-available, accurate, measurement data to support theoretical computations. Due to cost, installation challenges, and harsh operating conditions, there are few openly-available shaft and propeller torque measurements during ice interaction. As such, available propulsion shaft response data, along with inverse models, are used to determine propeller torsional loads. It remains to verify and validate these models in a controlled environment. Common verification methods consist of subjecting the inverse models to standardised load cases which compare the resulting inverse to the input load cases by the classification society. The use of experimental data, however, quantifies actual ice-loading responses, as opposed to ideal load cases, and prevents inverse crime when implemented in inverse models. For this purpose, a laboratory rig was developed, manufactured and commissioned that is capable of exerting a measured external torque to the end of a shaft whilst performing conventional shaft measurements. A disc brake was implemented in the place of a propeller as load device. The laboratory rig is able to simulate important phenomena observed in ship shaft measurements during propeller-ice contact. A continuous, modal superposition model was adapted to the dynamics of the rig, and the inverse brake torque was calculated from shaft measurements. It was found that the first torque peak, of the inverse solution, is within 14% of the measured torque. Subsequent torque peaks, such as those from multi-impact ice milling cases, however, are not inverted successfully, as the free-response of the shaft contaminates the torque solution. Consequently, the nature of ice-loading is not adequately described by the continuous model. Recommendations are made to improve the laboratory rig and the continuous model investigated.
AFRIKAANSE OPSOMMING: Die toename in maritieme verkeer in poolstreke, skep gevaarlike toestande vir skepe, wat duur en tydrowende herstelwerk in die hawe kan vereis. Die strewe na omgewingsvriendelike, energiedoeltreffende skepe druk skeepsbouers om te voldoen aan streng ontwerpriglyne wat deur klassifikasieliggame gestel word. Verbetering van doeltreffendheid en produksiekoste gaan dikwels gepaard met ’n vermindering in oorontwerpte komponente. As gevolg hiervan, is daar ’n kritieke behoefte aan openlik beskikbare, akkurate, metingsdata om teoretiese berekeninge te ondersteun. As gevolg van koste, installasieprobleme, en strawwe bedryfstoestande, is daar min openlik beskikbare, dryfas- en skroefmetings, tydens ysinteraksie. Sodanig word beskikbare dryfas-reaksiedata, saam met inverse modelle, gebruik om skroefladings te bepaal. Daar is steeds ’n behoefte om hierdie modelle te verifieer en te valideer in ’n gekontroleerde omgewing. Algemene verifikasiemetodes bestaan uit die onderwerping van die inverse modelle aan gestandaardiseerde lasgevalle, wat die resulterende inverse vergelyk met die inset lasgevalle wat deur die klassifikasieliggaam bepaal is. Die gebruik van eksperimentele data kwantifiseer egter werklike yslaai-reaksies, in teenstelling met ideale lasgevalle, en voorkom wiskundige wanpraktyke wanneer dit in inverse modelle geimplementeer word. Vir hierdie doel is ’n laboratoriumopstelling ontwikkel, vervaardig, en in gebruik geneem wat in staat is om ’n gemete eksterne wringkrag aan die einde van ’n dryfas uit te oefen, terwyl konvensionele dryfasmetings uitgevoer word. ’n Skyfrem is in die plek van ’n skroef as lastoestel geimplementeer. Die laboratoriumopstelling is in staat om belangrike verskynsels te simuleer wat in skeepdryfasmetings tydens skroef-ys-kontak waargeneem is. ’n Kontinue, modale superposisie-model, is aangepas tot die dinamika van die opstelling, en die inverse wringkrag is uit dryfasmetings bereken. Daar is gevind dat die eerste inverse wringkragpiek binne 14% van die gemete wringkrag is. Daaropvolgende wringkragpieke, soos in die geval van multi-impak yslaste op die propeller, is egter nie suksesvol bepaal nie, aangesien die vrye vibrasie van die dryfas die inverse wringkragoplossing besoedel. Gevolglik word die aard van yslading nie voldoende deur die model beskryf nie. Aanbevelings word gemaak om die laboratoriumtopstelling te verbeter sowel as die kontinue model wat ondersoek is.
Description
Thesis (MEng)--Stellenbosch University, 2023.
Keywords
Citation