Savi LoopIQ • Advanced control practical • advanced-smith-deadtime

Smith predictor for dead-time-dominant process

Exercise-specific learner workbook. Complete each section while using the matching native Savi LoopIQ Lab.

1 • Learn2 • Practice3 • Explain4 • Repeat

Learning objective

Use an internal process model to remove known dead time from the controller-facing prediction.

Exercise context

Recognize a process whose dead time fundamentally limits conventional feedback performance. Verify Smith compensation against the same controller gain without the predictor and understand model-mismatch risk.

Practice procedure

  1. Predict the system response and record your assumptions before running the exercise.
  2. Open this exact case in its corresponding native Savi LoopIQ simulator or graded Lab.
  3. Observe the trend and record evidence for model direction, lag, delay and uncertainty.
  4. Explain why your action is supported or why the safe response is to stop and review.
  5. Compare with the in-app feedback, then retry from memory after 2–3 days and 1–2 weeks.

Explain the engineering decision

  1. Identify the controller structure and cause-effect direction.
  2. What baseline comparison should remain fixed?
  3. Which mismatch or failure mode can reverse the benefit?
  4. What safety-gate evidence would you keep?

Facilitator / reviewer guidance

Compare against the in-app deterministic practical assessment and retain its safety and uncertainty conditions.

Evidence and repetition record

Attempt 1 date: ______________

Confidence (0–5): ______

Observed result: __________________
Attempt 2 date: ______________

Confidence (0–5): ______

What changed? ___________________
Engineering safety boundary. All exercises are educational and use synthetic examples. A worksheet, simulation score or training credential cannot command industrial equipment or authorize a live process change. Follow site management-of-change and qualified-engineer review.