A Novel Data Collection Framework for Telemetry and Anomaly Detection in Industrial IoT Systems
21 Apr 2020-pp 245-251
TL;DR: An industrial IoT architectural framework that allows data offloading between the cloud and the edge is proposed and an anomaly detection algorithm that exploits deep learning techniques to assess the working conditions of the plant is designed.
Abstract: The advent of IoTs has catalyzed the development of a variety of cyber-physical systems in which hundreds of sensor-actuator enabled devices (including industrial IoTs) cooperatively interact with the physical and human worlds. However, due to the large volume and heterogeneity of data generated by such systems and the stringent time requirements of industrial applications, the design of efficient frameworks to store, monitor and analyze the IoT data is quite challenging. This paper proposes an industrial IoT architectural framework that allows data offloading between the cloud and the edge. Specifically, we use this framework for telemetry of a set of heterogeneous sensors attached to a scale replica of an industrial assembly plant. We also design an anomaly detection algorithm that exploits deep learning techniques to assess the working conditions of the plant. Experimental results show that the proposed anomaly detector is able to detect 99% of the anomalies occurred in the industrial system demonstrating the feasibility of our approach.
Citations
More filters
TL;DR: It is pointed out that predictive maintenance is a hot topic in the context of Industry 4.0 but with several challenges to be better investigated in the area of machine learning and the application of reasoning.
189 citations
TL;DR: This paper proposes a comprehensive literature review of recent outlier detection techniques used in the IoTs context and provides the fundamentals of outlier Detection while discussing the different sources of an outlier, the existing approaches, how to evaluate anoutlier detection technique, and the challenges facing designing such techniques.
Abstract: The Internet of Things (IoT) is a fact today where a high number of nodes are used for various applications. From small home networks to large-scale networks, the aim is the same: transmitting data from the sensors to the base station. However, these data are susceptible to different factors that may affect the collected data efficiency or the network functioning, and therefore the desired quality of service (QoS). In this context, one of the main issues requiring more research and adapted solutions is the outlier detection problem. The challenge is to detect outliers and classify them as either errors to be ignored, or important events requiring actions to prevent further service degradation. In this paper, we propose a comprehensive literature review of recent outlier detection techniques used in the IoTs context. First, we provide the fundamentals of outlier detection while discussing the different sources of an outlier, the existing approaches, how we can evaluate an outlier detection technique, and the challenges facing designing such techniques. Second, comparison and discussion of the most recent outlier detection techniques are presented and classified into seven main categories, which are: statistical-based, clustering-based, nearest neighbour-based, classification-based, artificial intelligent-based, spectral decomposition-based, and hybrid-based. For each category, available techniques are discussed, while highlighting the advantages and disadvantages of each of them. The related works for each of them are presented. Finally, a comparative study for these techniques is provided.
23 citations
TL;DR: In this paper, the authors proposed a Distributed Anomaly Detection (DAD) system to discover zero-day attacks in edge networks using Gaussian Mixture-based Correntropy, a novel ensemble one-class statistical learning model.
14 citations
01 Jan 2022
TL;DR: In this paper , the authors present a Digital Twin Pipeline Framework of the COGNITWIN project that supports Hybrid and Cognitive Digital Twins, through four Big Data and AI pipeline steps adapted for Digital Twins.
Abstract: Abstract This chapter presents a Digital Twin Pipeline Framework of the COGNITWIN project that supports Hybrid and Cognitive Digital Twins, through four Big Data and AI pipeline steps adapted for Digital Twins. The pipeline steps are Data Acquisition, Data Representation, AI/Machine learning, and Visualisation and Control. Big Data and AI Technology selections of the Digital Twin system are related to the different technology areas in the BDV Reference Model. A Hybrid Digital Twin is defined as a combination of a data-driven Digital Twin with First-order Physical models. The chapter illustrates the use of a Hybrid Digital Twin approach by describing an application example of Spiral Welded Steel Industrial Machinery maintenance, with a focus on the Digital Twin support for Predictive Maintenance. A further extension is in progress to support Cognitive Digital Twins includes support for learning, understanding, and planning, including the use of domain and human knowledge. By using digital, hybrid, and cognitive twins, the project’s presented pilot aims to reduce energy consumption and average duration of machine downtimes. Data-driven artificial intelligence methods and predictive analytics models that are deployed in the Digital Twin pipeline have been detailed with a focus on decreasing the machinery’s unplanned downtime. We conclude that the presented pipeline can be used for similar cases in the process industry.
8 citations
References
More filters
TL;DR: In this article, the authors present a cloud centric vision for worldwide implementation of Internet of Things (IoT) and present a Cloud implementation using Aneka, which is based on interaction of private and public Clouds, and conclude their IoT vision by expanding on the need for convergence of WSN, the Internet and distributed computing directed at technological research community.
9,593 citations
TL;DR: The fields of application for IoT technologies are as numerous as they are diverse, as IoT solutions are increasingly extending to virtually all areas of everyday.
Abstract: It has been next to impossible in the past months not to come across the term ‘‘Internet of Things’’ (IoT) one way or another. Especially the past year has seen a tremendous surge of interest in the Internet of Things. Consortia have been formed to define frameworks and standards for the IoT. Companies have started to introduce numerous IoTbased products and services. And a number of IoT-related acquisitions have been making the headlines, including, e.g., the prominent takeover of Nest by Google for $3.2 billion and the subsequent acquisitions of Dropcam by Nest and of SmartThings by Samsung. Politicians as well as practitioners increasingly acknowledge the Internet of Things as a real business opportunity, and estimates currently suggest that the IoT could grow into a market worth $7.1 trillion by 2020 (IDC 2014). While the term Internet of Things is now more and more broadly used, there is no common definition or understanding today of what the IoT actually encompasses. The origins of the term date back more than 15 years and have been attributed to the work of the Auto-ID Labs at the Massachusetts Institute of Technology (MIT) on networked radio-frequency identification (RFID) infrastructures (Atzori et al. 2010; Mattern and Floerkemeier 2010). Since then, visions for the Internet of Things have been further developed and extended beyond the scope of RFID technologies. The International Telecommunication Union (ITU) for instance now defines the Internet of Things as ‘‘a global infrastructure for the Information Society, enabling advanced services by interconnecting (physical and virtual) things based on, existing and evolving, interoperable information and communication technologies’’ (ITU 2012). At the same time, a multitude of alternative definitions has been proposed. Some of these definitions exhibit an emphasis on the things which become connected in the IoT. Other definitions focus on Internet-related aspects of the IoT, such as Internet protocols and network technology. And a third type centers on semantic challenges in the IoT relating to, e.g., the storage, search and organization of large volumes of information (Atzori et al. 2010). The fields of application for IoT technologies are as numerous as they are diverse, as IoT solutions are increasingly extending to virtually all areas of everyday. The most prominent areas of application include, e.g., the smart industry, where the development of intelligent production systems and connected production sites is often discussed under the heading of Industry 4.0. In the smart home or building area, intelligent thermostats and security systems are receiving a lot of attention, while smart energy applications focus on smart electricity, gas and water meters. Smart transport solutions include, e.g., vehicle fleet tracking and mobile ticketing, while in the smart health area, topics such as patients’ surveillance and chronic disease management are being addressed. And in the context of Accepted after one revision by Prof. Dr. Sinz.
3,499 citations
"A Novel Data Collection Framework f..." refers background in this paper
...Such objects act as bridges to the physical world, catering to various service needs such as decision making [1], [2]....
[...]
TL;DR: This paper provides a comprehensive review of associated topics such as intelligent manufacturing, Internet of Things (IoT)-enabled manufacturing, and cloud manufacturing and describes worldwide movements in intelligent manufacturing.
1,602 citations
"A Novel Data Collection Framework f..." refers background in this paper
...In particular, machine learning helps build intelligence in the industrial systems, allowing them to make autonomous decisions [6]....
[...]
TL;DR: The concepts of IoT, Industrial IoT, and Industry 4.0 are clarified and the challenges associated with the need of energy efficiency, real-time performance, coexistence, interoperability, and security and privacy are focused on.
Abstract: Internet of Things (IoT) is an emerging domain that promises ubiquitous connection to the Internet, turning common objects into connected devices. The IoT paradigm is changing the way people interact with things around them. It paves the way for creating pervasively connected infrastructures to support innovative services and promises better flexibility and efficiency. Such advantages are attractive not only for consumer applications, but also for the industrial domain. Over the last few years, we have been witnessing the IoT paradigm making its way into the industry marketplace with purposely designed solutions. In this paper, we clarify the concepts of IoT, Industrial IoT, and Industry 4.0. We highlight the opportunities brought in by this paradigm shift as well as the challenges for its realization. In particular, we focus on the challenges associated with the need of energy efficiency, real-time performance, coexistence, interoperability, and security and privacy. We also provide a systematic overview of the state-of-the-art research efforts and potential research directions to solve Industrial IoT challenges.
1,402 citations
"A Novel Data Collection Framework f..." refers background in this paper
...See [4], [8] for underlying challenges and future directions of the problems discussed in the previous section....
[...]
TL;DR: A unified architectural model and a new taxonomy are presented, by comparing a large number of solutions to support the requirements of IoT applications that could not be met by today’s solutions.
184 citations
"A Novel Data Collection Framework f..." refers background in this paper
...To realize this paradigm, the edge and cloud computing are key enabling technologies [5]....
[...]