The Smith predictor control algorithm goes back to the 1950s. Its purpose and history are described in Wikipedia.
The Smith predictor (invented by O. J. M. Smith in 1957) is a type of predictive controller designed to control systems with a significant feedback time delay.
But is it the best approach to handle feedback time delay?
A fascinating ControlGlobal.com article, Face-off: Smith Predictor vs. deadtime compensated PID, highlights the possibility of a better approach. Written by Peter Morgan and Greg McMillan, the article provides a:
…performance comparison of the Smith Predictor and deadtime compensated PID.
The article’s authors challenge:
…the traditional view that the Smith Predictor is the pre-eminent choice for deadtime compensation. It establishes, by analysis, the best theoretical integral absolute error (IAE) for a Smith Predictor exactly modeling the process. Then, it compares the performance of the Smith Predictor and deadtime-compensated controller employing external reset feedback (ERFB).
Process automation hall of fame legend F.Greg Shinsky, in a ControlGlobal.com article, The power of external-reset feedback, noted:
Preventing [integral] windup requires reconfiguration of the controller, and external-reset feedback is the most satisfactory method of accomplishing it.
Peter and Greg compare the performance between the Smith predictor and deadtime-compensated controller employing ERFB for several cases, including:
- Unmeasured load disturbance (first order plus deadtime process approximation)
- Unmeasured load disturbance (second order plus deadtime process approximation))
- Setpoint change (first order plus deadtime process approximation)
- Setpoint change (second order plus deadtime process approximation)
You’ll want to read the article to see how the performance compares in each case, but I’ll highlight the final conclusion from their testing.
The deadtime-compensated PID controller can outperform the Smith Predictor for setpoint changes and unmeasured load changes. This strongly suggests that implementing the PID controller using integral action through positive feedback should see wider use. This would allow a simple and effective means of deadtime compensation when deadtime is adjusted to match process deadtime to avoid ERFB deadtime that’s smaller or larger than the actual deadtime.
Not all PID controllers have this deadtime-compensated with external reset feedback capability, but if you’re a DeltaV distributed control system user, you do (Dynamic Reset Limit). It may be time to use this capability more for your lag-dominant processes.
