Identified from routine data
Dead time, process gain and the installed controller are all identified from closed-loop history the plant already logs. No producing unit has to sit through a step test.
Tune
Settle the loops that are quietly working against you.
Three level loops on this train cycle because their reset is far too fast for their gain. Tune shows what the history says each loop and its valve actually do, the gain and reset to enter instead, and the same hours replayed through both — before anything changes in the DCS.
3Level loops tuned
2Loops refused, with the reason
0Step tests required
12 daysOf one-minute history studied
MVPEvery figure above is counted in the minimum viable product: the amine train modelled for gas sweetening and post-combustion carbon capture.
What it does
Four things it does that a dashboard does not.
Three responses, not one opinion
Tight, Normal and Soft, each a fixed multiple of the loop dead time, so each carries a known robustness. The planner picks the trade; the arithmetic is not a preference.
Replayed before it is changed
Every candidate setting is replayed against the loop’s own measured disturbance. Level movement and valve travel — which drives packing wear — are shown side by side for the current and the new tuning.
It refuses when the data cannot support it
Two loops whose output never moves cannot be identified from history, so the tool says so instead of printing numbers. Every figure it does show carries its evidence beside it.
How it works
The pipeline, end to end.
- 1
Pick a loop
Each loop opens with its valve, its duty and the tags behind it.
- 2
See what the loop does today
The identified process and the installed gain and reset, with how much the level wanders and how hard the valve works.
- 3
Choose a response
Tight, Normal or Soft. The new gain and reset are computed live, with their robustness margins.
- 4
Compare the replay
The same hours of plant history, run through the current and the proposed setting.
- 5
Hand it to the engineer
The tool recommends. A person enters the change in the DCS.
Where it applies
Built for amine systems today. Not limited to them by design.
The platform is focused on acid gas removal — gas sweetening and post-combustion carbon capture — because that is where the model is calibrated and where the fault library is real. The method underneath is not amine-specific: a physics reference model, residuals against expected-healthy, fuzzy diagnosis over graded indicators, and costed consequences. Extending it to another unit means new datasheets and a new calibration, not a new platform.
The fastest way to judge this is to open it.
The platform runs on a real amine train with thirteen days of one-minute history behind it. Nothing on these pages is a mock-up.
Open the platform