DeltaV InSight automatically learns users’ processes with embedded learning algorithms running at the controller level and develops process models based on day-to-day operations. These models allow users to identify operational benchmarks, diagnose problems and calculate optimum loop tuning across the entire control system.
Upon process changes like an operator setpoint change or a sequential logic-induced change which causes the process to move to a new operating state, the software learns the dynamics of the process from this change and provides recommendations on new tuning calculations. I discussed his capability in detail in an earlier post with DeltaV Advanced Control product manager, John Caldwell.
Over the years, I’ve become a fan of blogger and new Microsoft employee, Jon Udell and his use of screencasts. These short screen-captured videos really save thousands of words and help quickly demonstrate something he is discussing.
The screencast begins with a one-slide overview of DeltaV InSight followed by a demonstration of the software. I hope it conveys in its 3:22 second run-time a sample of this process of recognizing, learning, recommending, and implementing the change. There is also a product data sheet and whitepaper now available presenting additional capabilities.
John adds that the development team worked closely with several process manufacturers in developing and testing this functionality. We developed a video from one of the manufacturers, Lubrizol, last fall to document some of the initial results they saw.