Johnston, Mitchell (2020) Machine data collection and analysis for system fault rectification. [USQ Project]
Abstract
Machine data collection is an integral part of any fault rectification process within industry today. The majority of the fault data collection for root cause analysis within a system comes directly from the plant or equipment in question in the form of machine data. There is a large amount of system information and data held or observed by the technical staff responsible for rectifying the fault. This information is not usually used in the root cause analysis or used solely by the technical staff independent of any automated fault determination or analysis. There are few applications in existence that are designed to collect both data from machines and technical staff for analysis that can present the best-known method of fault rectification based on both data sources.
This research project determines the best methodology for simultaneous data collection from both plant/machine, and the technical staff rectifying the system faults. The research project discusses the best way to collect and store this data. Then shows how it can be thoroughly analysed and utilised in an IOT application to assist in the fault rectification process. This research project aimed to identify the best method to capture system data from PLCs, as well as event and system knowledge from maintenance technicians during the occurrence of a fault within an automated system. This research project also examined how this data can and should be stored and then utilised in the most effective way possible while considering the cost-benefit factors involved.
The project consists of four phases. Phase 1 identified a test site which was then used to implement a trial system and test the project outcomes. Identifying a suitable test site entailed sourcing a site with a large enough variety of equipment and maintenance staff. Once established, Phase 2 involved conducting a site review to identify existing infrastructure and fault rectification data. Phase 3 determined the best data collection, storage and analysis methodologies based on the site and literature reviews. Finally, Phase four was the implementation of the new methodology and trial system onsite for utilization by technical staff.
The projects key outcome was the implementation of a fault data collection and analysis system on the test site. The system successfully showed how a number of low-cost components and software packages which included NodeRED and SQLite, can be used to collect, store, sort and analyse fault data then present it back to an end-user for the purpose of system fault rectification. The impimentation of the system was successful and has provided proof of concept for the trial system. Positive feedback was provided from all parties involved which has paved the way for futher development of the current aplication on the test site and future implementation across additional sites, involving the development of IOT capabilities.
|
Statistics for this ePrint Item |
| Item Type: | USQ Project |
|---|---|
| Item Status: | Live Archive |
| Faculty/School / Institute/Centre: | Historic - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering (1 Jul 2013 - 31 Dec 2021) |
| Supervisors: | Hills, Catherine; Koh, Glendon |
| Qualification: | Bachelor of Engineering (Honours) (Electrical and Electronic) |
| Date Deposited: | 22 Jul 2021 01:21 |
| Last Modified: | 26 Jun 2023 04:08 |
| Uncontrolled Keywords: | machine data collection |
| URI: | https://sear.unisq.edu.au/id/eprint/42847 |
Actions (login required)
 |
Archive Repository Staff Only |
