Process control lab / challenge 01

Separator Control Challenge

Educational simulatorNot for field operation or equipment sizing

Three-phase separator / operator briefing

Can you keep a three-phase separator stable?

Tune two PID control loops, survive realistic well-testing disturbances, and keep the process inside its safe operating envelope.

SEP-101 / NORMAL OPERATION / IDLE VISUAL

SEP-101 horizontal separator with gas, oil, and water. True total level 50.0 percent; measured total level 50.0 percent. True interface 20.0 percent; measured interface 20.0 percent. Total inlet 50.0 cubic metres per hour; oil outlet 35.0 cubic metres per hour; water outlet 15.0 cubic metres per hour. Actual oil valve 50.0 percent; actual water valve 50.0 percent. Safety: Normal operating envelope.

SEP-101 · engineering fallback

Separator process cutaway

Normal operating envelope
GAS SPACEOIL INVENTORYWATER INVENTORYTOTAL INLET · 50.0 m³/hOIL OUTLET · 35.0 m³/hWATER OUTLET · 15.0 m³/hTOTAL TRUE 50.0%MEASURED 50.0%INTERFACE TRUE 20.0%MEASURED 20.0%SEP-101 · THREE-PHASE INVENTORY
Total true
50.0%
Total measured
50.0%
Total SP
50.0%
Interface true
20.0%
Interface measured
20.0%
Interface SP
20.0%
Total inlet
50.0 m³/h
Oil outlet
35.0 m³/h
Water outlet
15.0 m³/h
Oil command
50.0%
Oil actual
50.0%
Water command
50.0%
Water actual
50.0%
Active safety
Normal operating envelope

2D engineering fallback · process values and safety logic remain fully active.

  1. 01
    Tune

    Set the oil and water PID gains before the disturbance arrives.

  2. 02
    Survive

    Read the coupled inventories and hold the safety envelope.

  3. 03
    Score

    Finish a deterministic attempt and inspect the control response.

Project note / authorship

About this project

Separator Control Challenge is an independent educational project by Marcos Esteban Soto, combining well testing and industrial process control with ongoing work in machine learning and intelligent industrial systems.

Built with AI-assisted engineering using GPT-5.6 and Codex. Process assumptions, physical simplifications, control behavior, and educational content were defined and reviewed using engineering judgment.