Visualization and diagnostic data for the Condition-Based Maintenance system
MetadataShow full item record
Condition-Based Maintenance (CBM) system is getting quite popular in the industry and defense fields these days. A CBM system can efficiently predict the potential failures of the system and send the alarms. As a result, the lifetime of the system can be prolonged and the breakdown time can be decreased dramatically. The intelligent CBM system even can make the decision without any human intervention in the future. This master thesis concentrates on the diagnosis part and the visualization part of the CBM system for a fan. The diagnosis is the core part of a CBM, which requires an efficient and optimum method for the different situations in reality. This method directly impacts the performance of the CBM. The visualization is an interface by which the personnel and the software system exchange information. The quality of this interface also can impact the feeling of customers and its availability. In this master thesis, I developed a intelligent diagnosis solution for the CBM system for the fan. The system has the ability to learn from the experience and the known data, rather than just judges the values with the certain thresholds which is widely used as a tradition system state monitor method. I also implement the prototype. This master thesis deals with a partial solution of an efficient CBM system for the fan in reality. The diagnosis solution can predict the potential failure in advance. The graphic user interface is designed to be easily understandable and neat. The personnel can read all related information from the interface at the first place. However, there is still much work to do. The method in this thesis and its prototype implementation can give the readers an overview and knowledge of a CBM system.
Masteroppgave i informasjons- og kommunikasjonsteknologi 2009 – Universitetet i Agder, Grimstad