Knut Erik Knutsen
Knut Erik Knutsen has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Erik Vanem
DNV, Høvik, Norway
4 shared publications - Houxiang Zhang
Department of Ocean Operations and Civil Engineering, Norwegian University of Science and Technology, Larsgaardsvegen 2, 6009 Aalesund, Norway
3 shared publications - Qin Liang
DNV Veracity
3 shared publications - Vilmar Æsøy
Norwegian University of Science and Technology
2 shared publications - Andrej Kuznetsov
Department of Physics and Centre for Materials Science and Nanotechnology, University of Oslo, Oslo, Norway
1 shared publication - Bengt Svensson
Department of Biochemistry, Molecular Biology, and Biophysics, University of Minnesota, 6-155 Jackson Hall, 321 Church Street S.E., Minneapolis, MN 55455, USA
1 shared publication
Latest Publications
- 13 November 2024
- Published by ASME International in Journal of Offshore Mechanics and Arctic Engineering
- Vol. 147 (3) , 1-38
A Hybrid Approach Integrating Physics-Based Models and Expert-Augmented Neural Networks for Ship Fuel Consumption Prediction
The International Maritime Organization’s recent approval of the 2023 strategy on reduction of greenhouse gas emissions amplifies the pressure on stakeholders to achieve net-zero emissions in shipping by 2050. Considering the anticipated predominance of traditional single-fuel engines into the next decade, due to their high efficiency and economic benefits, the implementation of operational measures stands as the foremost effective method for mitigating emissions and reducing fuel consumption. Accurate fuel consumption prediction is crucial for informed decision-making and operational efficiency. This paper introduces an innovative hybrid model, combining an advanced physics-based model with an expert-augmented neural network, offering superior fuel consumption predictions. Expert knowledge is integrated into the neural network model to enhance its learning capabilities. Performance is validated against DNV Navigator Insight and publicly available fuel consumption reporting data, demonstrating superiority over purely data-driven and physics-based models. This hybrid approach bridges accuracy and scalability for sustainable maritime operations.
- 13 November 2024
- Published by ASME International in Journal of Offshore Mechanics and Arctic Engineering
- Vol. 147 (3) , 1-38
A Hybrid Approach Integrating Physics-Based Models and Expert-Augmented Neural Networks for Ship Fuel Consumption Prediction
The International Maritime Organization’s recent approval of the 2023 strategy on reduction of greenhouse gas emissions amplifies the pressure on stakeholders to achieve net-zero emissions in shipping by 2050. Considering the anticipated predominance of traditional single-fuel engines into the next decade, due to their high efficiency and economic benefits, the implementation of operational measures stands as the foremost effective method for mitigating emissions and reducing fuel consumption. Accurate fuel consumption prediction is crucial for informed decision-making and operational efficiency. This paper introduces an innovative hybrid model, combining an advanced physics-based model with an expert-augmented neural network, offering superior fuel consumption predictions. Expert knowledge is integrated into the neural network model to enhance its learning capabilities. Performance is validated against DNV Navigator Insight and publicly available fuel consumption reporting data, demonstrating superiority over purely data-driven and physics-based models. This hybrid approach bridges accuracy and scalability for sustainable maritime operations.
- Open Access
- 8 October 2024
- Published by PHM Society in International Journal of Prognostics and Health Management
- Vol. 15 (3)
Anomaly Detection in Time Series Data
This paper presents a novel unsupervised approach for anomaly detection in marine diesel engines using a Transformer Neural Network based autoencoder (TAE) and residual analysis with Sequential Probability Ratio Test (SPRT) and Sum of Squares of Normalized Residuals (SSNR). This approach adeptly captures temporal dependencies within normal time-series data without the necessity for labeled failure data. To assess the performance of the proposed methodology, a dataset containing faulty data is generated under the same operational profile as the normal training data. The model undergoes training using normal data, after which the faulty data is reconstructed utilizing the trained model. Subsequently, SPRT and SSNR are used to analyze the residuals from the observed and reconstructed faulty data. Significant deviations surpassing a predefined threshold are identified as anomalous behavior. Additionally, this study explores various architectures of Transformer neural networks and other types of neural networks to conduct a comprehensive comparative analysis of the performance of the proposed approach. Insights and recommendations derived from the performance analysis are also presented, offering valuable information for potential users to leverage. The experimental results demonstrate that the proposed approach can accurately and efficiently detect anomalies in marine diesel engines. Therefore, this approach can be considered as a promising solution for early anomaly detection, leading to timely maintenance and repair, and preventing costly downtime. Also, avoid accidents with probability severe consequences.
- 8 October 2024
- Published by PHM Society in International Journal of Prognostics and Health Management
- Vol. 15 (3)
Open AccessAnomaly Detection in Time Series Data
This paper presents a novel unsupervised approach for anomaly detection in marine diesel engines using a Transformer Neural Network based autoencoder (TAE) and residual analysis with Sequential Probability Ratio Test (SPRT) and Sum of Squares of Normalized Residuals (SSNR). This approach adeptly captures temporal dependencies within normal time-series data without the necessity for labeled failure data. To assess the performance of the proposed methodology, a dataset containing faulty data is generated under the same operational profile as the normal training data. The model undergoes training using normal data, after which the faulty data is reconstructed utilizing the trained model. Subsequently, SPRT and SSNR are used to analyze the residuals from the observed and reconstructed faulty data. Significant deviations surpassing a predefined threshold are identified as anomalous behavior. Additionally, this study explores various architectures of Transformer neural networks and other types of neural networks to conduct a comprehensive comparative analysis of the performance of the proposed approach. Insights and recommendations derived from the performance analysis are also presented, offering valuable information for potential users to leverage. The experimental results demonstrate that the proposed approach can accurately and efficiently detect anomalies in marine diesel engines. Therefore, this approach can be considered as a promising solution for early anomaly detection, leading to timely maintenance and repair, and preventing costly downtime. Also, avoid accidents with probability severe consequences.
- Open Access
- 4 September 2023
- Published by PHM Society in PHM Society Asia-Pacific Conference
- Vol. 4 (1)
Unsupervised Anomaly Detection in Marine Diesel Engines using Transformer Neural Networks and Residual Analysis
This paper presents a novel unsupervised approach for detecting anomalies in marine diesel engines using a Transformer Neural Network based autoencoder (TAE) and residual analysis with Sequential Probability Ratio Test (SPRT) and Sum of Squares of Normalized Residuals (SSNR). The proposed method can capture temporal dependencies in normal timeseries data without the need for labeled failure data. To assess the effectiveness of the proposed approach, a dataset of faulty data is generated under the same operational profile as the normal training data. The model is trained using normal data, and the faulty data is reconstructed using the trained model. SPRT and SSNR are then used to analyze the residuals from the observed and reconstructed faulty data, with significant deviations exceeding a predefined threshold being identified as anomalous behavior. The experimental results demonstrate that the proposed approach can accurately and efficiently detect anomalies in marine diesel engines. Therefore, this approach can be considered as a promising solution for early anomaly detection, leading to timely maintenance and repair, and preventing costly downtime.
- 4 September 2023
- Published by PHM Society in PHM Society Asia-Pacific Conference
- Vol. 4 (1)
Open AccessUnsupervised Anomaly Detection in Marine Diesel Engines using Transformer Neural Networks and Residual Analysis
This paper presents a novel unsupervised approach for detecting anomalies in marine diesel engines using a Transformer Neural Network based autoencoder (TAE) and residual analysis with Sequential Probability Ratio Test (SPRT) and Sum of Squares of Normalized Residuals (SSNR). The proposed method can capture temporal dependencies in normal timeseries data without the need for labeled failure data. To assess the effectiveness of the proposed approach, a dataset of faulty data is generated under the same operational profile as the normal training data. The model is trained using normal data, and the faulty data is reconstructed using the trained model. SPRT and SSNR are then used to analyze the residuals from the observed and reconstructed faulty data, with significant deviations exceeding a predefined threshold being identified as anomalous behavior. The experimental results demonstrate that the proposed approach can accurately and efficiently detect anomalies in marine diesel engines. Therefore, this approach can be considered as a promising solution for early anomaly detection, leading to timely maintenance and repair, and preventing costly downtime.
- Open Access
- 15 March 2023
- Published by Elsevier in Ocean Engineering
- Vol. 275
Integrated machine learning and probabilistic degradation approach for vessel electric motor prognostics
- 15 March 2023
- Published by Elsevier in Ocean Engineering
- Vol. 275
Open AccessIntegrated machine learning and probabilistic degradation approach for vessel electric motor prognostics
- Open Access
- 28 October 2022
- Published by PHM Society in Annual Conference of the PHM Society
- Vol. 14 (1)
Federated Learning on Tamperproof Data for PHM on Marine Vessels Using a Docker Based Infrastructure
Prognostics and health management (PHM) on systems such as vehicles and marine vessels are sometimes held back by complexities relating to data ownership and intellectual property rights. This is particularly true when multiple Original Equipment Manufacturers (OEMs) deliver components or sub-systems to a customer while having an interest in monitoring and maintenance of said components or sub-systems. Further, the collection of PHM data from a fleet which may be non-uniform and spread across the globe with varying degrees of connectivity can be challenging from a bandwidth and cybersecurity point of view. Federated learning may address some of these challenges and open up new opportunities for how to approach PHM on a global and non-uniform fleet of components or systems. In this article, we will discuss how a Docker based infrastructure for data collection, storage and analysis in combination with a methodology for tamperproofing PHM data can be a powerful substrate for bringing trust and transparency to federated learning implementations of PHM algorithms. We also demonstrate a basic blockchain enabled federated learning experiment using the proposed Docker based infrastructure in combination with centralized data centers.
- 28 October 2022
- Published by PHM Society in Annual Conference of the PHM Society
- Vol. 14 (1)
Open AccessFederated Learning on Tamperproof Data for PHM on Marine Vessels Using a Docker Based Infrastructure
Prognostics and health management (PHM) on systems such as vehicles and marine vessels are sometimes held back by complexities relating to data ownership and intellectual property rights. This is particularly true when multiple Original Equipment Manufacturers (OEMs) deliver components or sub-systems to a customer while having an interest in monitoring and maintenance of said components or sub-systems. Further, the collection of PHM data from a fleet which may be non-uniform and spread across the globe with varying degrees of connectivity can be challenging from a bandwidth and cybersecurity point of view. Federated learning may address some of these challenges and open up new opportunities for how to approach PHM on a global and non-uniform fleet of components or systems. In this article, we will discuss how a Docker based infrastructure for data collection, storage and analysis in combination with a methodology for tamperproofing PHM data can be a powerful substrate for bringing trust and transparency to federated learning implementations of PHM algorithms. We also demonstrate a basic blockchain enabled federated learning experiment using the proposed Docker based infrastructure in combination with centralized data centers.
- Open Access
- 8 January 2013
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 113 (2) , 023702
Diffusion and configuration of Li in ZnO
Diffusion of Li into ZnO from an “infinite” surface source under oxygen-rich conditions is studied using secondary ion mass spectrometry. The Li concentration-versus-depth profiles exhibit a distinct and sharp drop, which evolves in position with temperature and time. The sharp drop is associated with an efficient conversion from highly mobile Li-interstitials (Lii) to practically immobile Li-substitutionals (LiZn) via a kick-out mechanism. The characteristic concentration level at which Li drops provides a measure of the active donor concentration in the samples at the processing temperature, and gives evidence of residual impurities being responsible for the commonly observed “native” n-type conductivity. These donors are suggested to arise from different impurities, with Al and Si as the prevailing ones in hydrothermal and melt grown material. Further, evidence of electric field effects on Li diffusion profiles is obtained, and they are considered as a main reason for the slow diffusivity obtained in this work (using O-rich conditions) relative to those previously reported in the literature (obtained under Zn-rich conditions).
- 8 January 2013
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 113 (2) , 023702
Open AccessDiffusion and configuration of Li in ZnO
Diffusion of Li into ZnO from an “infinite” surface source under oxygen-rich conditions is studied using secondary ion mass spectrometry. The Li concentration-versus-depth profiles exhibit a distinct and sharp drop, which evolves in position with temperature and time. The sharp drop is associated with an efficient conversion from highly mobile Li-interstitials (Lii) to practically immobile Li-substitutionals (LiZn) via a kick-out mechanism. The characteristic concentration level at which Li drops provides a measure of the active donor concentration in the samples at the processing temperature, and gives evidence of residual impurities being responsible for the commonly observed “native” n-type conductivity. These donors are suggested to arise from different impurities, with Al and Si as the prevailing ones in hydrothermal and melt grown material. Further, evidence of electric field effects on Li diffusion profiles is obtained, and they are considered as a main reason for the slow diffusivity obtained in this work (using O-rich conditions) relative to those previously reported in the literature (obtained under Zn-rich conditions).