Marine Engine-Centered Data Analytics for Ship Performance Monitoring

Perera, Lokukaluge P.; Mo, Brage

doi:10.1115/1.4034923

Cited by 11 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The methods have shown to be valuable in demonstrating better collection and analyses that improves operations and reduces use of energy. Future development should implement solutions that avoids the necessity of manual inputs for data collection, include a wider scope of parameters to include weather and navigation information, and start to use methodologies within artificial intelligence to find/reduce sensor faults and increase/improve holistic analyses in the context of ship performance [13][14][15].…”

Section: Discussionmentioning

confidence: 99%

Automated System for Fleet Benchmarking and Assessment of Technical Condition

Dehli

Steinebach

et al. 2017

Volume 7A: Ocean Engineering

Self Cite

View full text Add to dashboard Cite

On-board Energy systems in vessels have to use fuel efficiently to maintain ship speed at lowest possible costs. This paper describes how to use ship operational data to improve and maintain efficiency of the vessel’s power production with respect to the condition and performance of equipment. The paper describes an overview of an automated Technical Operations Performance (TOP) Monitoring service. TOP monitors the performance of marine main and auxiliary diesel engines by use of the information collected onboard vessels at regular intervals. Performance data are stored in xml-reports sent as email attachments from ship to shore. This communication is reliable and cost efficient for merchant ships that are on-line only for shorter periods. Load, ISO and environmental corrections make results valid for benchmarking and trending. The service aggregates the hierarchical information obtained from different sources by transferring measurement readings into unified indicators, the Technical Condition Index (TCI) [1]. Experts manually check the automatically generated performance reports and add additional guidance on options to improve power production and machinery conditions analyzing the available data with respect to different targets, such as low engine degradation and high fuel efficiency. The performance reports then influence business processes indicating possible causes for loss of performance in equipment and possible erroneous instrumentation, and the need for maintenance actions. The obtained TCI values show the performance of individual units, or for a fleet/class of equipment and vessels.

show abstract

Section: Discussionmentioning

confidence: 99%

Automated System for Fleet Benchmarking and Assessment of Technical Condition

Dehli

Steinebach

et al. 2017

Volume 7A: Ocean Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…Gaussian mixture models (GMMs) with an expectation maximization (EM) algorithm [24] is proposed in this study to identify the respective data clusters in engine propeller combinator diagram. One should note that GMMs represent multivariate Gaussian distributions with the respective mean and covariance values.…”

Section: Gaussian Mixture Modelsmentioning

confidence: 99%

Ship performance and navigation information under high-dimensional digital models

Perera

2019

J Mar Sci Technol

Self Cite

View full text Add to dashboard Cite

Future vessels will be facilitated by modern internet of things (IoT) to collect various ship performance and navigation information. Such information is collected as large-scale data sets, so called Big Data and that should be utilized towards digitalization of the shipping industry. However, various data handling challenges are encountered by the shipping industry during the phase of digitalization, onboard as well as onshore. Data driven models, so called digital models, to support data handling frameworks of the shipping industry are proposed by this study. Such models can overcome the respective data handling challenges in shipping, where conventional mathematical models may fail to facilitate. These models can be derived from ship performance and navigation data sets by considering the high dimensional parameter space. Such high dimensional models consist of several data clusters and each data cluster may consist of a possible unique data structure. These data clusters often relate to sub-operational conditions of vessels and ship systems. The identification of the distribution of data clusters and the structure of each data cluster in relation to ship performance and navigation conditions have been done under this study for a selected vessel as the main contribution. Furthermore, the domain knowledge in shipping (i.e. vessel operational and navigation conditions) is also considered during this analysis to interpret the meaning in such digital models.Index Terms-Digitalization, Shipping industry, big data, digital models, data analytics, machine leaning and artificial intelligence.

show abstract

“…An alternative strategy to PMS is Conditioned-based Maintenance (CM), wherein the maintenance is carried out based on the condition of the machinery, which is detected by measuring several parameters during the vessel operation [ 4 ]. Fortunately, the recent development of IT technology has enabled real-time access to the machinery’s condition and energy efficiency using data collected from onboard sensors [ 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 ]. Such sensor-based monitoring can be used to detect abnormal behavior of the system that may indicate the degradation or fault of the system.…”

Section: Introductionmentioning

confidence: 99%

An Ensemble-Based Approach to Anomaly Detection in Marine Engine Sensor Streams for Efficient Condition Monitoring and Analysis

Kim¹,

Lee²,

Lee

2020

Sensors

View full text Add to dashboard Cite

This study proposes an unsupervised anomaly detection method using sensor streams from the marine engine to detect the anomalous system behavior, which may be a possible sign of system failure. Previous works on marine engine anomaly detection proposed a clustering-based or statistical control chart-based approach that is unstable according to the choice of hyperparameters, or cannot fit well to the high-dimensional dataset. As a remedy to this limitation, this study adopts an ensemble-based approach to anomaly detection. The idea is to train several anomaly detectors with varying hyperparameters in parallel and then combine its result in the anomaly detection phase. Because the anomaly is detected by the combination of different detectors, it is robust to the choice of hyperparameters without loss of accuracy. To demonstrate our methodology, an actual dataset obtained from a 200,000-ton cargo vessel from a Korean shipping company that uses two-stroke diesel engine is analyzed. As a result, anomalies were successfully detected from the high-dimensional and large-scale dataset. After detecting the anomaly, clustering analysis was conducted to the anomalous observation to examine anomaly patterns. By investigating each cluster’s feature distribution, several common patterns of abnormal behavior were successfully visualized. Although we analyzed the data from two-stroke diesel engine, our method can be applied to various types of marine engine.

show abstract

Marine Engine-Centered Data Analytics for Ship Performance Monitoring

Cited by 11 publications

References 9 publications

Automated System for Fleet Benchmarking and Assessment of Technical Condition

Automated System for Fleet Benchmarking and Assessment of Technical Condition

Ship performance and navigation information under high-dimensional digital models

An Ensemble-Based Approach to Anomaly Detection in Marine Engine Sensor Streams for Efficient Condition Monitoring and Analysis

Contact Info

Product

Resources

About