Abstract
Estimating shaft power of a crew boat is very important to be analysed because it has high-speed operational characteristics along with limited routes. To understand the phenomena, 3 sister crew-boats with operational distance about 40-60 nautical miles every day are investigated. The daily operational time is 8 hours and the configurations are: 4.04% full speed, 13.63% economical speed, 1.81% slow speed, 7.65% snatching, 1.25% manoeuvring, 5.29% idle, and the remaining time is in standby condition.The crew boats are fitted with a monitoring system namely SHIMOS®, in which data is sent to a server in the centre office every 2 minutes. The data consists of time capture, boat position (latitude and longitude), speed, left and right RPM engine, left and right flow-meter data engine, and average of fuel consumption data in everyday operation. Three years of data has been collected for the vessel.The present study proposed characteristics of crew-boat shaft power, which affected by external factors using Artificial Neural Network (ANN) back propagation method and optimisation in 4 hidden layers and 40 neurons with relative error 6.2%. The results demonstrates good agreement with previous popular method that using statistical models.
Highlights
The use of energy from fossil fuels has increased excessively in the past 50 years and continues to increase considering the development of the human population to date in accordance with data on world population prospects 2019 reaching 7.7 billion [1]
The crew boats are fitted with a monitoring system namely SHIMOS®, in which data is sent to a server in the centre office every 2 minutes
The data consists of time capture, boat position, speed, left and right RPM engine, left and right flow-meter data engine, and average of fuel consumption data in everyday operation
Summary
The use of energy from fossil fuels has increased excessively in the past 50 years and continues to increase considering the development of the human population to date in accordance with data on world population prospects 2019 reaching 7.7 billion [1]. This method based on the backpropagation learning algorithm that is applied with n- hidden layer, n-weighted and activation function with bipolar sigmoid function-tanh hyperbolic function. The system is trained using supervised learning methods, where errors between system outputs and expected outputs are known to be presented to the system and used to modify the internal state of the weight, neurons, and hidden layers. This process will repeat itself according to iteration by updating the weight, neurons and hidden layers based on the conjugate gradient until the expected error criteria is obtained. The testing process carried out directly on the network and there is no back propagation and iteration process
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IOP Conference Series: Earth and Environmental Science
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
