Predictions-on-chip: model-based training and automated deployment of machine learning models at runtime

Pilarski, Sebastian; Staniszewski, Martin; Bryan, Matthew; Villeneuve, Frederic; Varró, Dániel

doi:10.1007/s10270-020-00856-9

Cited by 7 publications

(2 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Some applications, such as critical infrastructure or manufacturing processes, require real-time fault diagnosis. Achieving low-latency processing while maintaining high accuracy is a challenge [196], [197].…”

Section: Classification Of Intelligent Fault Diagnosis (Ifd) Methodsmentioning

confidence: 99%

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management & Fault Diagnosis in Resource-Constrained Environments

Khan,

Akram,

Khattak

et al. 2024

Preprint

View full text Add to dashboard Cite

Industry 4.0 represents the fourth industrial revolution, which is characterized by the incorporation of digital technologies, the Internet of Things (IoT), artificial intelligence, big data, and other advanced technologies into industrial processes.Industrial Machinery Health Management (IMHM) is a crucial element, based on the Industrial Internet of Things (IIoT), which focuses on monitoring the health and condition of industrial machinery. The academic community has focused on various aspects of IMHM, such as prognostic maintenance, condition monitoring, estimation of remaining useful life (RUL), intelligent fault diagnosis (IFD), and architectures based on edge computing. Each of these categories holds its own significance in the context of industrial processes. In this survey, we specifically examine the research on RUL prediction, edge-based architectures, and intelligent fault diagnosis, with a primary focus on the domain of intelligent fault diagnosis. The importance of IFD methods in ensuring the smooth execution of industrial processes has become increasingly evident. However, most methods are formulated under the assumption of complete, balanced, and abundant data, which often does not align with real-world engineering scenarios. The difficulties linked to these classifications of IMHM have received noteworthy attention from the research community, leading to a substantial number of published papers on the topic. While there are existing comprehensive reviews that address major challenges and limitations in this field, there is still a gap in thoroughly investigating research perspectives across RUL prediction, edge-based architectures, and complete intelligent fault diagnosis processes. To fill this gap,we undertake a comprehensive survey that reviews and discusses research achievements in this domain, specifically focusing on IFD. Initially, we classify the existing IFD methods into three distinct perspectives: the method of processing data, which aims to optimize inputs for the intelligent fault diagnosis model and mitigate limitations in the training sample set; the method of constructing the model, which involves designing the structure and features of the model to enhance its resilience to challenges; and the method of optimizing training, which focuses on refining the training process for intelligent fault diagnosis models and emphasizes the importance of ideal data in the training process. Subsequently, the survey covers techniques related to RUL prediction and edge-cloud architectures for resource-constrained environments. Finally, this survey consolidates the outlook on relevant issues in IMHM, explores potential solutions, and offers practical recommendations for further consideration.

show abstract

Section: Classification Of Intelligent Fault Diagnosis (Ifd) Methodsmentioning

confidence: 99%

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management & Fault Diagnosis in Resource-Constrained Environments

Khan,

Akram,

Khattak

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…As a result, engineering design information relates to all of these domains (McKay et al, 1996) along with information on the mappings themselves and rationale for decisions made (e.g., see Design Rationale editor (Bracewell, et al, 2009). Some authors report work using design analysis and optimisation approaches to generate training data (Pilarski et al, 2021;Sharpe et al, 2019;Tallman et al, 2019) but, again, this is limited to shape models and associated analysis parameters and objective functions that do not reflect the full richness of the design requirements being addressed. In addition, the problems addressed tend to be focussed on specific design problems for which it is possible to specify specific design parameters and goals.…”

Section: Literature Reviewmentioning

confidence: 99%

Design Descriptions in the Development of Machine Learning Based Design Tools

et al. 2023

View full text Add to dashboard Cite

Applications of machine learning technologies are becoming ubiquitous in many sectors and their impacts, both positive and negative, are widely reported. As a result, there is substantial interest from the engineering community to integrate machine learning technologies into design workflows with a view to improving the performance of the product development process. In essence, machine learning technologies are thought to have the potential to underpin future generations of data-enabled engineering design system that will deliver radical improvements to product development and so organisational performance. In this paper we report learning from experiments where we applied machine learning to two shape-based design challenges: in a given collection of designed shapes, clustering (i) visually similar shapes and (ii) shapes that are likely to be manufactured using the same primary process. Both challenges were identified with our industry partners and are embodied in a design case study. We report early results and conclude with issues for design descriptions that need to be addressed if the full potential of machine learning is to be realised in engineering design.

show abstract

Guideline for Deployment of Machine Learning Models for Predictive Quality in Production

et al. 2022

View full text Add to dashboard Cite

Predictions-on-chip: model-based training and automated deployment of machine learning models at runtime

Cited by 7 publications

References 32 publications

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management & Fault Diagnosis in Resource-Constrained Environments

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management & Fault Diagnosis in Resource-Constrained Environments

Design Descriptions in the Development of Machine Learning Based Design Tools

Guideline for Deployment of Machine Learning Models for Predictive Quality in Production

Contact Info

Product

Resources

About

Predictions-on-chip: model-based training and automated deployment of machine learning models at runtime

Cited by 7 publications

References 32 publications

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management &amp; Fault Diagnosis in Resource-Constrained Environments

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management &amp; Fault Diagnosis in Resource-Constrained Environments

Design Descriptions in the Development of Machine Learning Based Design Tools

Guideline for Deployment of Machine Learning Models for Predictive Quality in Production

Contact Info

Product

Resources

About

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management & Fault Diagnosis in Resource-Constrained Environments

Deep Learning based Approaches for Intelligent Industrial Machinery Health Management & Fault Diagnosis in Resource-Constrained Environments