Affiliation:
1. Stolt Tankers, Rotterdam, the Netherlands
2. Delft University of Technology, Department of Maritime and Transport Technology, Delft, the Netherlands
Abstract
Trim optimization improves the energy efficiency of ships, thus reducing operational costs and emissions; however, trim tables are only available for a limited number of ships. There is thus a desire to develop additional, more accurate trim tables without the need for expensive model testing. The objective of this research was to develop a method to decrease fuel consumption by trim optimization, by a dynamic shaft power estimation model based on available operational data. A method that uses noon report data and a grey-box modelling approach is proposed. The grey box model consists of a multi-layer feedforward neural network to estimate the required shaft power, using operational parameters and an initial estimate of the required shaft power. A case study is presented for a modern chemical tanker and sea trials have been conducted to validate the results. The method provides correct trim advice for full load conditions; however, the magnitude of the effect is smaller compared to sea trial results. The model is able to estimate the required power with an average accuracy of over 6% for a random subset of the noon report data. Due to challenges inherent to noon reports as a data source, the actual effect of trim and speed have a bigger magnitude than the extracted trend.
Subject
Mechanical Engineering,Ocean Engineering
Reference22 articles.
1. The impact of optimizing trim on reducing fuel consumption;Abouedafdl;Journal of shipping and Ocean Engineering,2016
2. L. Aldous, T. Smith and R. Bucknall, Noon report data uncertainty, in: Low Carbon Shipping Conference, London, 2013.
3. An artificial neural network based decision support system for energy efficient ship operations;Bal Beşikçi;Computers and Operations Research,2016
4. Vessels fuel consumption forecast and trim optimisation: A data analytics perspective;Coraddu;Ocean Engineering,2017
5. Da Silva et al., Artificial Neural Networks – a Practical Course, 1st edn, Springer International Publishing, 2017.
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献