Investigating an SVM-driven, one-class approach to estimating ship systems condition

Lazakis, Iraklis; Gkerekos, Christos; Theotokatos, Gerasimos

doi:10.1080/17445302.2018.1500189

Cited by 52 publications

(27 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Further elaboration of used machine learning methods, experiments and results are available in [21] and [22]. Usually, those are neural networks (ANN) [23], support vector machines (SVM) [22,24] and random forests (RF) [21]. In ANN, whose prediction results are similar to SVM and RF, there is a demanding pre-processing procedure, as well as finding optimal parameters for building a successful model.…”

Section: Uc Model Structurementioning

confidence: 99%

A framework for the application of shipboard energy efficiency monitoring, operational data prediction and reporting

2021

View full text Add to dashboard Cite

In this study, a framework for the application of shipboard energy efficiency monitoring, operational data prediction and reporting based on the ship’s measurement data and meteorological and oceanographic data by the geographic position and time of navigation is presented. General system theory in synergy with machine learning (ML) is used to construct the framework. The general system theory is utilized for identification and transition of components of the proposed framework of energy efficiency monitoring and prediction. A systematic investigation of the internal and external environment is conducted, and the definition of information flow between the individual components provided. Then, the external opportunities and threats that the system faces were opposed to internal strengths and weaknesses to formulate strategies in which weaknesses and threats of the system are offset by existing strengths and probabilities. After assessing the results of the strengths, weaknesses, opportunities and threats (SWOT) and threats, opportunities, weaknesses and strengths (TOWS) analysis, it can be concluded that the proposed framework is feasible and widely applicable in the maritime industry. The novelty is that the proposed framework is using on-board data processing and is integrated into the existing ship monitoring, decision-making and reporting system, thus satisfying the prerequisites for simple application.

show abstract

Section: Uc Model Structurementioning

confidence: 99%

A framework for the application of shipboard energy efficiency monitoring, operational data prediction and reporting

2021

View full text Add to dashboard Cite

show abstract

“…Condition-based maintenance (CBM) is a maintenance strategy based on the monitoring of assets' conditions, implemented to reduce the number of failures associated with machinery (Lazakis et al 2016;Raptodimos and Lazakis 2018;Lazakis et al 2019;Raptodimos and Lazakis 2020). Owing to its capacity to increase safety and reduce risk, Industrial Internet of Things (IIoT) is applied to install a large number of sensors alongside the most critical components of the ship and around their environment to assess not only the conditions of the assets but also the operational environment.…”

Section: Introductionmentioning

confidence: 99%

A novel framework for imputing large gaps of missing values from time series sensor data of marine machinery systems

Velasco-Gallego

Lazakis

2021

Ships and Offshore Structures

Self Cite

View full text Add to dashboard Cite

Condition-based maintenance is a maintenance strategy that implements Industrial Internet of Things to monitor the assets' condition. Despite its undeniable benefits, several challenges are encountered, such as the incompleteness of sensor data. Hence, while data imputation is an important practise, there is a lack of analysis and formalisation of data imputation in the maritime industry. Accordingly, a novel framework is introduced by implementing the first-order Markov chain in tandem with a multivariate imputation approach based on a comparative methodology of 16 machine learning and time series forecasting models. To highlight its performance efficiency, a comparative study is conducted between the proposed framework and the MICE approach by the implementation of a case study on a total of 4 parameters, obtained from sensors installed on the marine machinery systems of a cargo vessel. The results demonstrated an improvement of 21-77%, indicating its performance efficiency as a data imputation technique.

show abstract

“…Deriving a model that can accurately predict vessel performance under varying ship operational profiles and environmental conditions can assist in the identification of optimal operating profiles. Additionally, the existence of such a model could help recognise deviating performance patterns that could imply the vessel's systems and/or subsystems degradation (Cipollini et al, 2018;Raptodimos and Lazakis, 2018;Lazakis et al, 2018Lazakis et al, , 2019.…”

Section: Introductionmentioning

confidence: 99%

Machine learning models for predicting ship main engine Fuel Oil Consumption: A comparative study

2019

Self Cite

View full text Add to dashboard Cite

As Fuel Oil Consumption (FOC) constitutes over 25% of a vessel's overall operating cost, its accurate forecasting, and the reliable prediction of the relevant ship operating expenditures can majorly impact the ship operation sustainability and profitability. This study presents a comparison of data-driven, multiple regression algorithms for predicting ship main engine FOC considering two different shipboard data acquisition strategies, noon-reports and Automated Data Logging & Monitoring (ADLM) systems. For this, various multiple regression algorithms including Support Vector Machines (SVMs), Random Forest Regressors (RFRs), Extra Trees Regressors (ETRs), Artificial Neural Networks (ANNs), and ensemble methods are employed. The effectiveness of the tested algorithms is investigated based on a number of key performance indicators, such as the mean and median average error and the coefficient of determination (R 2). ETR and RFR models were found to perform best in both cases, whilst the existence of an ADLM system increased accuracy by 7% and reduced the the required period for data collection by up to 90%. The derived models can accurately predict the FOC of vessels sailing under different load conditions, weather conditions, speed, sailing distance, and drafts.

show abstract

Investigating an SVM-driven, one-class approach to estimating ship systems condition

Cited by 52 publications

References 33 publications

A framework for the application of shipboard energy efficiency monitoring, operational data prediction and reporting

A framework for the application of shipboard energy efficiency monitoring, operational data prediction and reporting

A novel framework for imputing large gaps of missing values from time series sensor data of marine machinery systems

Machine learning models for predicting ship main engine Fuel Oil Consumption: A comparative study

Contact Info

Product

Resources

About