Mid Sweden University

With the rapid development of artificial intelligence and industrial Internet of Things (IIoT) technologies, intelligent predictive maintenance (IPdM) has received considerable attention from researchers and practitioners. To efficiently predict impending failures and mitigate unexpected downtime, while satisfying the instant maintenance demands of industrial facilities is very important for improving the production efficiency. In this article, a self-attention based "Perception and Prediction" framework, called DeepHealth, is proposed for the instant IPdM. Specifically, the framework is composed of two submodels (i.e., DH-1 and DH-2), which are respectively utilized to perform the health perception and sequence prediction. By operating the framework, the proposed models can predict the health conditions via predicting the future signal samples, thereby completing the instant IPdM. Considering the potential temporal correlation in time series, we deploy an enhanced attention mechanism to capture global dependencies from the vibration signals, and leverage the long- and short-term sequence prediction of sensor signals to support instant maintenance decision-making. On this basis, we conduct a destructive experiment based on the IIoT-enabled rotating machinery and construct a balanced industrial dataset for model evaluations. Extensive experiment results show that the proposed solution achieves good prediction accuracy for instant IPdM on the automatic washing equipment and Case Western Reserve University datasets.

Industrial wireless sensor networks (IWSNs) have the potential to contribute significantly to a variety of wireless sensing endpoints, such as cable replacement, mobility, flexibility and cost reduction. However, the harsh and varied industrial environment entails addressing severe challenges, such as dust, heat, electromagnetic interference (EMI) and radio frequency interference from the other heterogeneous networks. One of the important challenges is the link burstiness in industrial wireless environments, which has not been heavily researched. In this paper, we propose a new timeslot scheduling algorithm to address this problem. The algorithm allows transmissions on the same link to be separated at least by a minimum timeslot distance. Moreover, to provide a synergistic complement to the timeslot scheduling, we also propose a timeslot reuse scheme, which facilitates improved efficiency and reliability in the bandwidth utilization and retransmission and decreases the average delay of the packet arrival. We evaluate the proposed algorithm and scheme by a real implementation, targeting a specific industrial application. The experimental results generally indicate that the proposed media access control (MAC) scheme greatly improves the reliability and decreases the packet arrival delay. Compared with the existing popular methods, the execution of our scheme indicates a minimal link burstiness influence and possesses more efficient bandwidth utilization.

Industrial wireless sensor networks (IWSNs) have mainly been used to monitor applications, but recently an interest in control and safety applications has emerged. Functional safety and communication in open transmission systems have been laid down in the IEC 61784-3-3 standard. The standard is based on a cyclic polling mechanism, which consumes a considerable amount of bandwidth; since existing IWSNs are very resource-constrained, this becomes a major challenge. To overcome this problem, this paper proposes a novel framework that uses an event-triggered failsafe mechanism based on synchronous wired polling and wireless time-slotted time division multiple access. We analytically derive the minimum and maximum bound for the most important metric for safety-critical applications, safety function response time (SFRT). A new metric, normal state interrupt time (NSIT), is proposed in this paper. Furthermore, we also implement the proposed framework by using the WirelessHART standard. The results are compared to the classical time-triggered approach used in the IEC 61784-3-3 standard. The obtained results show that the proposed framework can reduce the bandwidth usage by 90% and support safety-critical applications that require a SFRT less or equal to 150 ms.

Industrial wireless sensor networks (IWSNs) have the potential to contribute significantly in areas such as cable replacement, mobility, flexibility, and cost reduction. Nevertheless, the industrial environment that the IWSNs operate in is very challenging because of dust, heat, water, electromagnetic interference, and interference from other wireless devices, which make it difficult for current IWSNs to guarantee reliable real-time communication. In this paper, we present a novel method based on the segmented slot assignment, fast slot competition, and free node concept that will improve the reliability and real-time communication significantly so that more advanced applications can be enabled. The main purpose of the algorithms is to improve the retransmission efficiency for time-division-multiple-access-based multihop IWSNs by using limited shared slot resources more efficiently. More importantly, the proposed algorithms support efficient slot rescheduling caused by link or node failure. We evaluate the proposed methods by using simulations and a real implementation targeting monitoring of welder machines. Our obtained results show that the proposed method outperforms the first published and most widely used IWSN standard called WirelessHART.

The invention concerns a method for providing wireless communication between a source node and a destination node in a wireless network wherein said wireless network uses a time division multiple access (TDMA) protocol. One or more dedicated TDMA transaction timeslots (60-71) are arranged in a superframe. The superframe comprises at least one dedicated transaction slot forming a hybrid protocol with two or more shared transaction slots 50 per superframe. The method is of particular advantage when practised with a WirelessHART protocol wireless sensor network and when used for monitoring and control equipment and processes in an industrial installation. A computer program, and a wireless node and a wireless network using the inventive method are also disclosed.

L'invention concerne un procédé permettant d'assurer une communication sans fil entre un nœud source et un nœud de destination dans un réseau sans fil, ledit réseau sans fil utilisant un protocole d'accès multiple par répartition temporelle (TDMA). Un ou plusieurs créneaux temporels de transaction TDMA dédiés (60-71) sont agencés dans une super-trame. La super-trame comprend au moins un créneau temporel de transaction dédié formant un protocole hybride comprenant au moins deux créneaux temporels de transaction partagés (50) par super-trame. Le procédé offre un avantage particulier lorsqu'il est mis en pratique avec un réseau de capteurs sans fil à protocole WirelessHART et lorsqu'il est utilisé pour surveiller et commander des équipements et des processus dans une installation industrielle. L'invention porte également sur un programme informatique, et un nœud sans fil et un réseau sans fil utilisant le procédé de l'invention.

Despite lots of research efforts in the area of Industrial Wireless Sensor Networks (IWSNs), there is a lack of really practical IWSN implementations, deployments and in-field experiments. This demo presents the design and implementation of an IWSN for welder machine systems based on the first IWSN standard WirelessHART. The goal of this work is to find the problems and challenges from IWSN standard to implementation, and motivate other designers to explore more IWSN applications.

A growing trend in the automation industry is to use wireless technologies to reduce cable costs and deployment time, unlock stranded information in deployed devices and enable wireless control applications. The WirelessHART protocol is the first open and interoperable industrial wireless sensor network standard, which proposes a combination of time division multiple access (TDMA) and channel hopping as a medium access control (MAC) protocol. TDMA is a schedule-based protocol, which is suitable for predictable cyclic traffic. However, for burst acyclic traffic, TDMA degenerates into an inefficient slotted ALOHA. In this article, we propose a complementary mechanism to TDMA, CCA-Embedded TDMA, which can be applied to improve the transmission efficiency and system stability of TDMA-based systems such as the WirelessHART protocol. A Markov model is proposed to evaluate the efficiency of system throughput and expected delay of CCA-Embedded TDMA. Furthermore, we show by analytical and simulation results that our proposed CCA-Embedded TDMA scheme can increase the throughput by 100% and reduce the expected delay by 75% for a common scenario. We also implement CCA-Embedded TDMA into an embedded system as a part of the WirelessHART MAC protocol. The experimental results verify the efficiency of CCA-Embedded TDMA and its backward compatibility with the WirelessHART protocol.

Industrial wireless applications have stringent requirements of safety, data update rate, time delay, reliability, failure tolerance, and security. In this paper, a concept of cellbased mesh networks is proposed, and the performance is evaluated by computer simulations. The results show that the average number of hops from sensor nodes to the backbone routers and the average route cost are significantly reduced with a comparison to a conventional mesh topology in multipath fading environments. It is also shown that the cell-based mesh networks can tolerate failures of links, notes and backbone routers. All those imply that the cell-based mesh networks are suitable for industrial wireless sensor networks with stringent requirements of data update rate, time delay, and reliability and failure tolerance.

TDMA has been resumed for industrial wireless sensor network (IWSN) like WirelessHART and ISA 100.11a recently, because TDMA is suitable for periodic traffic which is predictable to pre-assign. However, for random traffic TDMA degenerates to inefficient slotted ALOHA. In our previous work, we propose a new medium access control (MAC) scheme called discrete-time CSMA (DT-CSMA) that can be applied to improve the efficiency of TDMA-based system that transmits periodic and random traffic. In this paper, we give a model to evaluate system throughput for DT-CSMA. An analysis of typical scenario shows DT-CSMA improves the throughput by 100%. We also build a model to evaluate synchronization influence on DT-CSMA, which indicates that the influence is limited.

As more and more wireless devices use the 2.4 GHz radio spectrum, the coexistence of 2.4 GHz wireless devices operating in one place has become a hot topic. With low transmit power, the widely deployed IEEE 802.15.4-based networks are easily interfered with by other 2.4 GHz wireless networks, such as IEEE 802.11. IEEE 802.15.4-based wireless networks have paid great attention to the coexistence between themselves and with other non-IEEE 802.15.4 wireless networks. This problem has been further promoted by two new industry wireless standards, WirelessHART and ISA100, to meet special industry requirements. This paper surveys the studies on the coexistence between IEEE 802.11 and IEEE 802.15.4-based networks following the general analysis method of “question-analysis-solution.” Based on the survey study, we discuss about some open research issues and developments in this field.