VIBRATION SENSING FOR ENGINE CONDITION MONITORING AND PREDICTIVE MAINTENANCE
Masinde Mtesigwa, Masinde (2020)
Masinde Mtesigwa, Masinde
Åbo Akademi
2020
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi-fe2020051838056
https://urn.fi/URN:NBN:fi-fe2020051838056
Tiivistelmä
Edge computing and Edge analytics which can also be called fog computing send onlythe critical data to the cloud after filtering, aggregation and analysis at the edge. Theidea is to bring computing, storage and analytics closer to the source of the data. This isbecause of the problem which IoT cannot solve. The edge computing helps Industry 4.0 toexpand to the areas where IoT cannot reach and also solves the high costs of bandwidth.The latency problem due to the architecture gives birth to edge. Cloud is the ultimatedestination where all the edge devices will send the filtered, aggregated data for analysisand building models.Edge computing is one of the emerging paradigms which have become a feasiblesupplementary solution under the cloud architecture by increasing computing and storageresources available at the network edge. The edge solves the problems of latency ofsending to the cloud and the performance burden under IoT. [1]. Edge computing has thepotential to deliver low-latency, bandwidth-efficient and resilient services to IoT devicesvia adopting platforms that provide intermediate layers of computation, networking andstorage on the edge.The envisioned edge-driven IoT environment consists of three layers: IoT devices,edge layer, and cloud. The main layer, edge, plays the crucial role of bridging and in-terfacing the central cloud with IoT. The edge element in this thesis is Adlink FanlessEmbedded Computer which provides computing capability for data analysis, storage, andnetworking resources to the applications deployed across IoT devices and cloud. EdgeComputing is a highly virtualised platform which has computing, storage, and network-ing services between end devices and traditional Cloud Computing Data Centres. Today, an increasing part of added value for new technical solutions comes from dig-italisation and advanced automation. Through the use of big data and cloud analytics,machines can be made more reliable, more energy-efficient and the operation can be opti-mised. In this thesis, the target is to transfer the capabilities of big data and cloud process-ing to the edge, enabling real-time safety-critical operation, regardless of communicationavailability and at the same time minimising data transfer costs. The developed solutionand methods are utilised for the needs of machine diagnostic and analytics.