Recursive function identification and optimization for ships
2019 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE credits
Student thesis
Abstract [en]
Due to the fact that the fuel consumption of a ship is large, shipowners have a great interest in different kinds of fuel-saving solutions. For any fuel efficiency device, a mathematical function is required, that with high precision is able to give the fuel consumption.
The goal of this thesis is to identify the parameters of this function using the recursive least squares (RLS). These parameters will then be used for making predictions of the fuel consumption for a voyage. The implementation is made in Matlab. We will also investigate the quality of predicting fuel consumption using a neural network, which is implemented in Python.
The results show that RLS is an effective method for making predictions, whereas the neural network could use some further development before any conclusions can be made.
Place, publisher, year, edition, pages
2019. , p. 40
Keywords [en]
Least squares, recursive least squares, recursive identification methods, machine learning, neural network
National Category
Mathematics
Identifiers
URN: urn:nbn:se:mdh:diva-43647OAI: oai:DiVA.org:mdh-43647DiVA, id: diva2:1321235
External cooperation
Q-tagg R&D AB
Subject / course
Mathematics/Applied Mathematics
Supervisors
Examiners
2019-06-112019-06-072019-06-11Bibliographically approved