Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.583951
Title: Compact information technology enabled systems for intelligent process monitoring
Author: Ahsan, Qaisar
Awarding Body: Cardiff University
Current Institution: Cardiff University
Date of Award: 2006
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
The use of computers in industrial process applications is ever-increasing. Initially used to provide help to the machine operator, their application has evolved through automatic process control to monitoring of process health and performance. The latter, together with the quality control of the end product directly affect plant economics and ultimately the financial viability of the company. The research reported in this thesis is a contribution towards providing a cost-effective method of calculating a measure of the current health of a process and predicting any maintenance issues that may arise in the near future. Embedded systems are utilised and the monitoring system is designed to work automatically with a minimal input from the operator. This eliminates the need for peripherals such as keyboards, mice, and monitors thus reducing the overall system price and footprint. User interfaces are provided via the Internet and mobile phones giving remote access to multiple users. Single chip microcontrollers are at the heart of the embedded system rather than microprocessors, thereby reducing the relative system cost and size at the expense of localised processing power. The microcontrollers are distributed in a hierarchical network to attain the required processing power whilst minimising data storage and communications and to improve signal-to-noise ratios. The Controller Area Network (CAN) bus was selected, and used for the inter-microcontroller communications, for its robust performance in noisy environments. In the developed system architecture, each microcontroller node acquires one of the required process sensor signals and applies initial signal processing. A novel sweeping filter technique is developed to perform frequency analysis using the microcontrollers. The processed data from all nodes are then combined using situation-based criteria to reach conclusions often not evident from single sensor data. The Internet-based system is provided with the capability to upload any monitoring software or updates. Plug & play capability of the monitoring nodes is also provided so that the system can be seamlessly adapted to new or changed applications. The design and development of the system are detailed along with its deployment on various applications. Fault detection, isolation, and prediction were achieved on batch and continuous processes. A machine tool application proved the frequency analysis and network traffic reduction capabilities. On-line monitoring of an industrial valve was also performed.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.583951  DOI: Not available
Share: