Teaching Computers to Think Like Engineers

The marine industry is increasingly adopting Condition-Based Maintenance (CBM) as cost-effective strategy for Reduced Total Ownership Cost, fostering the approach of performing maintenance only when objective evidence of need exists.  However, because of the special skills and time required to implement CBM, particularly as ship systems become more complex, future ship systems should employ artificial intelligence to make the equipment smart enough to assess its own health and alert control systems and crews of failures and  performance degradations.

This paper describes how our DEXTER system uses probabilistic neural networks for diagnostic and prognostic reasoning about machinery faults. DEXTER’s neural networks learn to associate patterns of alarm conditions (symptoms) with machinery faults for real-time pattern recognition within complex machinery plants, such as those found on modern ships and oil rigs. This allows DEXTER to monitor the heartbeat of a machinery plant just like a doctor monitors your blood pressure, the main benefit of both being early detection of health problems.

Click the link below to download the white paper …

Teaching Computers to Think Like Engineers

Posted in: DEXTER ™ on July 30th by macsea

No comments yet

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.