dc.contributor.author | Neocleous, Costas K. | en |
dc.contributor.author | Schizas, Christos N. | en |
dc.creator | Neocleous, Costas K. | en |
dc.creator | Schizas, Christos N. | en |
dc.date.accessioned | 2019-11-13T10:41:26Z | |
dc.date.available | 2019-11-13T10:41:26Z | |
dc.date.issued | 2002 | |
dc.identifier.uri | http://gnosis.library.ucy.ac.cy/handle/7/54594 | |
dc.description.abstract | Cavitation in marine propellers can be a serious problem that may result in severe deterioration in performance. This is particularly important in heavily loaded propellers, commonly encountered in small craft. Efforts have been made to generate polynomials that fit experimental data on propeller performance and hence to facilitate the propeller selection procedures (Blount and Hubble, 1981). These polynomial fits are not accurate in capturing the performance of propellers, and also do not account for cavitating conditions. In the present work, neural networks have been developed that predict the performance of marine propellers in all tested conditions, including cavitation. The USN-series of experimental data (Denny et al, 1989) were applied on different neural network architectures and learning parameters, aiming at establishing a near optimum setup. The results of the networks are superior to those of the polynomial fit, and give an acceptable accuracy even in the cavitating conditions, thus enabling a naval architect/engineer to improve on the propeller selection process. | en |
dc.source | Proceedings of the International Joint Conference on Neural Networks | en |
dc.source | 2002 International Joint Conference on Neural Networks (IJCNN '02) | en |
dc.source.uri | https://www.scopus.com/inward/record.uri?eid=2-s2.0-0036079756&partnerID=40&md5=8a5cd58e1135e90fd8897523cd4b195e | |
dc.subject | Neural networks | en |
dc.subject | Parameter estimation | en |
dc.subject | Polynomials | en |
dc.subject | Torque | en |
dc.subject | Learning systems | en |
dc.subject | Ship propellers | en |
dc.subject | Cavitation corrosion | en |
dc.subject | Expanded area ratio | en |
dc.subject | Marine propellers | en |
dc.title | Artificial neural networks in estimating marine propeller cavitation | en |
dc.type | info:eu-repo/semantics/conferenceObject | |
dc.description.volume | 2 | |
dc.description.startingpage | 1848 | |
dc.description.endingpage | 1852 | |
dc.author.faculty | 002 Σχολή Θετικών και Εφαρμοσμένων Επιστημών / Faculty of Pure and Applied Sciences | |
dc.author.department | Τμήμα Πληροφορικής / Department of Computer Science | |
dc.type.uhtype | Conference Object | en |
dc.description.notes | <p>Sponsors: IEEE, NNS, INNS | en |
dc.description.notes | Conference code: 59176 | en |
dc.description.notes | Cited By :4</p> | en |
dc.contributor.orcid | Schizas, Christos N. [0000-0001-6548-4980] | |
dc.gnosis.orcid | 0000-0001-6548-4980 | |