Intelligent monitoring of large scale belt conveyor systems
Report 2006.TL.7056, Transport Engineering and Logistics.
Because of the ever growing complexity and size of large scale belt conveyor
systems, intelligent condition monitoring of these systems is increasingly
interesting in order to keep the operational costs as low as possible and
reduce system downtime. The existing techniques and possible techniques for
the future for the monitoring of belt conveyor systems available in
literature have been surveyed. The purpose of this survey is to investigate
techniques for intelligent condition monitoring of a large scale belt
conveyor system, therefore these questions are important:
The last step is to describe how an intelligent condition monitoring system
can be designed.
- Which techniques are mentioned in literature for the condition monitoring
of the subsystems of a large scale belt conveyor system and how can these
techniques be used for intelligent monitoring?
- What is the problem with traditional monitoring?
- Why should intelligent condition monitoring be implemented?
The steps that have to be taken to implement a condition monitoring system
are; establish need, select maintenance strategy, economic justification of
condition monitoring, investigate how equipment fails, select condition
monitoring technique, install monitoring system, data collection and data
analysis. All recent techniques for condition monitoring of large scale belt
conveyor monitoring are surveyed as well as possible techniques for the
future. For all subsystems of a large scale belt conveyor system several
condition monitoring techniques are described.
Finally the implementation of an intelligent condition monitoring system is
described. Because the traditional condition monitoring techniques can not be
interpreted by a computer often new techniques need to be used or developed.
The use of 'smart sensors' can provide the system with local signal
conditioning and pre-processing before transmitting the data to a condition
monitoring system. The possible intelligent decision making techniques: neural
networks, case-based reasoning, expert systems and fuzzy logic are described to
provide the system with learning and decision making abilities. Because of the
complexity of the system the use of condition monitoring agents for the
determination of the condition of a subsystem can be used to keep the amount of
data to be processed acceptable.
Reports on Transport Engineering and Logistics (in Dutch)
, TU Delft