The Problem

Control systems should execute the strategy laid out in the control narrative. Flawlessly. In reality, errors creep in. The control narrative may contain errors or may not have been written in the way the process designers had intended. In addition, the engineer responsible for coding the control systems may have interpreted the control narrative incorrectly or may have made errors in coding the system.

The Consequence

Obvious errors are easily spotted. However, there may be errors present that appear under certain plant conditions, or when the control system enters a specific mode of operation. These errors may cause trips that lead to production losses, and by their nature, may be hidden from detection.

Errors of this nature undermine confidence in the control system, operating staff may resign themselves to the fact that “that’s just the way it is.”

The Cause

As described above, there are several root causes to consider:

x Control narrative errors

x Control narratives not reflective of the process design intent

x Incorrect interpretation of the control narrative

x Errors made in coding the control system

The Solution

It’s a major challenge for control-system engineers to demonstrate errors on control narratives, as they are not necessarily process experts. That said there are techniques that can tease out control-narrative errors and coding errors.

✓An internal factory acceptance test (IFAT) is conducted for each system. Performing functional tests on each aspect of the control strategy, in line with the control narrative. Each input or output is physically exercised to verify correct wiring and configuration.

✓ An extended FAT (EFAT) is conducted after the FAT to demonstrate to the client that the system is functioning as specified. At this point, engineers and plant operators attend the EFAT to bring their experiences to bear. Various use cases are tested to demonstrate how the system responds under simulated plant

conditions.

✓ In extreme circumstances, developed thermodynamic simulators to simulate plant operations can be connected to the control system instead of simple tie-back or looped simulation. Any errors in logic (either coding errors or errors in the control narrative) become apparent when attempting to control the plant through the HMI.

Rigorous testing and coding discipline ensure errors are kept to a minimum.