Analysis and modeling of sensor data for ship motion prediction

Li, Guoyuan; Zhang, Houxiang; Kawan, Bikram; Wang, Hao; Osen, Ottar L.; Styve, Arne

doi:10.1109/oceansap.2016.7485648

Cited by 18 publications

(21 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The detail process is discussed in our previous article (Li et al, 2016). The third step is to perform sensitivity analysis to find out the contribution of each attribute for the outputs as discussed in (Li et al, 2016). For our case study, we found that surge velocity, sway velocity, yaw velocity, roll velocity, yaw velocity, position x, position y, heading, and roll has a significant relation for the prediction of the pitch.…”

Section: Experiments and Resultsmentioning

confidence: 93%

“…We performed three data cleaning methods i) removing noise using low pass filter ii) resampling is performed to make both module (M1 and M2 shown in Table I) in same frequency iii) phase correction is performed to make the continuity of raw data. The detail process is discussed in our previous article (Li et al, 2016). The third step is to perform sensitivity analysis to find out the contribution of each attribute for the outputs as discussed in (Li et al, 2016).…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…The heading degree changes from 360 to 0 after making one complete cycle. To remove this kind of discontinuity an algorithm is used to correct the such parameter in order to make the data consistent, (Li et al, 2016).…”

Section: Data Cleaningmentioning

confidence: 99%

See 2 more Smart Citations

Proceedings of the 58th Conference on Simulation and Modelling (SIMS 58) Reykjavik, Iceland, September 25th – 27th, 2017

2017

Linköping Electronic Conference Proceedings

View full text Add to dashboard Cite

Welcome to the 58th Conference on Simulation Modelling and Optimization in Reykjavik, Iceland. Reykjavik is the northernmost and one of the smallest capital cities in the world. Although small for a capital, it is the country's largest and most populated city and with the urban area population of around 200000, it is the home of the vast majority of Iceland's people. The city is pioneer in the use of geothermal energy to provide low-cost heating and electricity. Iceland's energy consumption per capita is among the highest in the world, and the proportion of this consumption provided by renewable energy sources exceeds most other countries. The electricity is produced almost entirely from renewable energy sources, hydroelectric (70%) and geothermal (30%) and simulation, modelling and optimization are the key elements to utilize the renewable energy more efficiently.The SIMS 58 will cover broad aspects of simulation, modelling and optimization and be a technical forum for participants to further their knowledge by being exposed to diverse topics, and exchange opinions and ideas both from industry and academia in a variety of topics related to simulation, modelling and optimization. The ambition is to bring the field of applied modelling and simulation techniques of energy systems to the general public as a contributing technology for efficient and sustainable energy systems. This year, the conference program contains two keynote speakers, 53 research papers and presentations organized into 14 sessions and an excursion to a geothermal park.The focus of the conference is on simulation, modelling and optimization in a variety of applied contexts including topics such as geothermal, hydro and bio energy, oil production, automotive industry and methodology and technologies for simulation, modelling and optimization.A key component of the SIMS conference is the opportunity to socialize and make new connections. The conference offers several possibilities such as a conference dinner at Bryggjan Brugghús (Bistro and Brewery), an excursion to the Reykjanes GEOPARK and last but not least some coffee and lunch breaks.SIMS stands for the Scandinavia Simulation Society with members from five Nordic countries, Denmark, Finland, Norway, Sweden and Iceland. The goal of SIMS is to develop further the science and practice of modelling and simulation in all application areas and to be a forum for information interchange between professionals and non-professionals in the Nordic countries.Thanks to all participants, authors, keynote speakers and presenters for their contribution to this conference. I also want to acknowledge the support I have received from the conference board, program committee and the SIMS board. The support from the National energy authority, Iceland is gratefully acknowledged.Welcome, I hope you will enjoy the conference and the interactions with your colleagues and other presenter here.Magnus Excursion:The conference excursion will be on Wednesday, September 27 th , at 10:00. Resource Park in the Reykjanes pen...

show abstract

Section: Experiments and Resultsmentioning

confidence: 93%

Section: Experiments and Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Proceedings of the 58th Conference on Simulation and Modelling (SIMS 58) Reykjavik, Iceland, September 25th – 27th, 2017

2017

Linköping Electronic Conference Proceedings

View full text Add to dashboard Cite

show abstract

“…It is therefore of great significance to make efforts to develop more general prediction model to investigate the effects of different strategies for short/long term ship motion prediction. The research of this paper builds upon the work by Li et al (2016), but focuses on modeling and analysis of NN construction and learning strategies. The main contribution of this study lies in a complete procedure from raw data analysis to NN modeling to short/long term prediction of different aspects of ship motion.…”

Section: Introductionmentioning

confidence: 99%

Neural-network-based modelling and analysis for time series prediction of ship motion

Kawan

Wang

et al. 2017

Ship Technology Research

Self Cite

View full text Add to dashboard Cite

Abstract:This paper presents a data-driven model for time series prediction of ship motion. Prediction based on past time series of data is a powerful function in modern ship support systems. For a large amount of ship sensor data, neural network (NN) is considered as a proper tool in modeling the prediction system. Efforts are made to compact the NN structure through sensitive analysis, in which the importance of each input to the output is quantized and lower ranked inputs are eliminated. Further analysis about the impact of three different learning strategies, i.e., offline, online and hybrid learning on the NN is conducted. The hybrid learning combining the advantages of both the offline learning and the online learning exhibits superior prediction performance. According to the long term prediction ability of recurrent NN, multi-stepahead prediction under the hybrid learning strategy is realized in a multi-stage prediction form. Experiments are carried out using collected ship sensor data on a vessel. The results show the feasibility of generating a data-driven model through modeling and analysis of the NN for ship motion prediction.Additional Information: Question ResponseIs your article an invited contribution for a special issue? If yes, please select the title of the special issue from the list below.Please select an option ONLY if you have received an invitation to submit. ABSTRACT This paper presents a data-driven model for time series prediction of ship motion. Prediction based on past time series of data is a powerful function in modern ship support systems. For a large amount of ship sensor data, neural network (NN) is considered as a proper tool in modeling the prediction system. Efforts are made to compact the NN structure through sensitive analysis, in which the importance of each input to the output is quantized and lower ranked inputs are eliminated. Further analysis about the impact of three different learning strategies, i.e., offline, online and hybrid learning on the NN is conducted. The hybrid learning combining the advantages of both the offline learning and the online learning exhibits superior prediction performance. According to the long term prediction ability of recurrent NN, multi-step-ahead prediction under the hybrid learning strategy is realized in a multi-stage prediction form. Experiments are carried out using collected ship sensor data on a vessel. The results show the feasibility of generating a data-driven model through modeling and analysis of the NN for ship motion prediction. Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation

show abstract

“…To date, a variety of novel intelligent approximation-based techniques and algorithms like fuzzy logic, Kalman filtering, Bayesian network, regression analysis and ANN have been applied to create predictive models [3][4][5][6]. Those methods have their own pros and cons at specific aspects.…”

Section: Introductionmentioning

confidence: 99%

Simplifying Neural Network Based Model for Ship Motion Prediction: A Comparative Study of Sensitivity Analysis

Cheng

Chen

Diao

et al. 2017

Volume 1: Offshore Technology

Self Cite

View full text Add to dashboard Cite

This paper presents a comparative study of sensitivity analysis (SA) and simplification on artificial neural network (ANN) based model used for ship motion prediction. Considering traditional structural complexity of ANN usually results in slow convergence, SA, as an efficient tool for correlation analysis, can help to reconstruct the ANN model for ship motion prediction. An ANN-Garson method and an ANN-EFAST method are proposed, both of which utilize the ANN for modeling but select the input parameters in a local and a global fashion, respectively. Through the benchmark tests, ANN-EFAST exhibits superior performance in both linear and nonlinear systems. Further test on ANN-EFAST via a case study of ship heading prediction shows its cost-effective and timely in compacting the ANN based prediction model.

show abstract

Analysis and modeling of sensor data for ship motion prediction

Cited by 18 publications

References 11 publications

Proceedings of the 58th Conference on Simulation and Modelling (SIMS 58) Reykjavik, Iceland, September 25th – 27th, 2017

Proceedings of the 58th Conference on Simulation and Modelling (SIMS 58) Reykjavik, Iceland, September 25th – 27th, 2017

Neural-network-based modelling and analysis for time series prediction of ship motion

Simplifying Neural Network Based Model for Ship Motion Prediction: A Comparative Study of Sensitivity Analysis

Contact Info

Product

Resources

About