Assess base-layer control performance and quantify the advanced process control opportunity on a candidate unit.
Oil, Gas & Energy
Advanced Process Control and Real-Time Optimisation in Refineries and Gas Plants
Builds the skills to design, commission and sustain multivariable process control and real-time optimisation on refinery and gas plant units, from step testing through to benefit tracking.
🗓️Duration5 training days
📚Content4 modules · 8 sessions
🎓On completionAccredited attendance certificate
About the programme
Course Overview
Base-layer PID control keeps a unit safe, but it rarely holds a refinery or gas plant at the true economic limit that feedstock and market conditions allow. This course is built around the full lifecycle of an advanced process control project: identifying which units justify the investment, designing the control structure around genuine economic and constraint priorities, and running the step tests needed to build accurate dynamic models. Participants configure and commission a multivariable model predictive controller, then extend the discussion into real-time optimisation, where a rigorous or hybrid steady-state model sets economic targets that the underlying controller tracks continuously. Applications cover distillation, fired heaters, compressor systems and gas plant fractionation trains, with attention to how real-time optimisation targets stay consistent with the refinery's overall linear programming plan. The closing module addresses what happens after commissioning: monitoring controller performance, diagnosing model decay, sustaining the economic benefit, and evaluating where digital twins and machine-learning soft sensors fit into the next generation of control architecture.
Expected Learning Outcomes
Design a control structure that sets clear constraint priorities and an economic objective for the controller to pursue.
Plan and execute step tests that identify accurate dynamic models for multivariable controller design.
Configure, commission and tune a multivariable model predictive controller through to safe closed-loop operation.
Build and integrate a real-time optimisation layer that keeps unit targets consistent with the refinery's economic plan.
Apply advanced process control and real-time optimisation to distillation, heater and gas plant fractionation applications.
Monitor controller performance after commissioning and sustain the economic benefit through model maintenance.
Who Should Attend
Process control engineers designing, commissioning or supporting advanced control applications.
Process engineers in refineries and gas plants evaluating units for advanced control investment.
Control system specialists responsible for multivariable controller tuning and performance monitoring.
Optimisation engineers integrating real-time optimisation with refinery planning and scheduling models.
Operations personnel who must understand controller behaviour, limits and override procedures.
Digitalisation leads assessing soft sensors and digital twins for future control system upgrades.
Course Modules
Select any module to see its sessions and points.
01Process Control Hierarchy and Advanced Control Opportunity
2 sessions · 8 points
Session 1From Regulatory Control to Multivariable Control
- Assess base-layer PID loop performance, including tuning and valve health, before committing a unit to an advanced process control project.
- Identify interacting control loops on a distillation column or furnace where single-loop control leaves the unit away from its true constraints.
- Quantify the control opportunity by comparing current operating variability against the distance to the nearest economic or safety constraint.
- Prioritise candidate units for an advanced process control programme based on constraint proximity, feedstock variability and expected margin uplift.
Session 2Defining the Control Objective and Economic Structure
- Translate refinery or gas plant economic drivers into a hierarchy of controlled, manipulated and disturbance variables for the control design.
- Define constraint priorities and give-up strategies so the controller pushes towards the correct limiting constraint under changing conditions.
- Build a control philosophy document that records the objective function, constraint set and expected operating envelope for later commissioning.
- Align control objectives with the unit's product specifications and safety instrumented system limits before any dynamic testing begins.
02Model Predictive Control Design, Testing and Commissioning
2 sessions · 8 points
Session 1Step Testing and Dynamic Model Identification
- Design a plant step test programme that excites manipulated variables sufficiently to identify accurate dynamic models without upsetting production.
- Identify step response models, including gain, dead time and time constant, for each manipulated-to-controlled variable pairing.
- Validate identified models against independent data and refine pairings where step responses show excessive noise or nonlinearity.
- Build inferential or soft sensor models to estimate quality variables, such as distillation cut points, that lack a fast online analyser.
Session 2Controller Configuration and Commissioning
- Configure a multivariable model predictive controller, such as DMCplus or Profit Controller, with move suppression and constraint weighting appropriate to the unit.
- Commission the controller in open-loop shadow mode before closing the loop, comparing predicted against actual process response.
- Tune constraint handling and economic optimisation weights so the controller drives the unit towards its true limiting constraint safely.
- Train operators on controller status displays, limit overrides and manual intervention procedures before full handover.
03Real-Time Optimisation and Integration with Planning
2 sessions · 8 points
Session 1Real-Time Optimisation Architecture and Modelling
- Build a steady-state rigorous or hybrid model of the unit for real-time optimisation that reconciles plant data through data validation and gross error detection.
- Integrate real-time optimisation set points with the underlying model predictive controller so economic targets update automatically as conditions change.
- Reconcile real-time optimisation results against the refinery linear programming plan to keep unit-level targets consistent with overall slate economics.
- Diagnose real-time optimisation model mismatch against plant data and trigger model recalibration when prediction error exceeds an agreed threshold.
Session 2Applying Real-Time Optimisation to Refinery and Gas Plant Units
- Apply real-time optimisation to a crude distillation unit to balance energy consumption against product yield and quality targets.
- Configure real-time optimisation and constraint control for gas plant fractionation trains to maximise recovery of natural gas liquids within compressor and column limits.
- Coordinate compressor anti-surge and load-sharing control with upstream advanced process control to avoid conflicting set point changes.
- Set escalation and fallback logic so the unit reverts safely to base-layer control if the optimisation layer becomes unavailable.
04Sustaining Performance and Digital Integration
2 sessions · 8 points
Session 1Controller Performance Monitoring and Benefit Sustainment
- Monitor controller uptime, constraint activity and variability reduction using key performance indicators tracked from the control system historian.
- Diagnose performance decay caused by model drift, sensor failure or process changes and schedule model maintenance accordingly.
- Recalculate sustained economic benefit after commissioning using before-and-after variability and constraint proximity comparisons.
- Build a benefit-tracking report that separates advanced process control contribution from other operational changes over the same period.
Session 2Digitalisation and Future Control Architecture
- Evaluate digital twin and first-principles simulation tools that support faster model building and controller retuning.
- Assess machine-learning-based soft sensors as a complement to first-principles inferential models where sufficient historical data exists.
- Plan a phased advanced process control and real-time optimisation roadmap that sequences units by value and control system readiness.
- Define change management and governance procedures so control model updates are tested and approved before deployment to live units.
What the participant receives
4 course modules
A structured syllabus
8 training sessions
across 5 days
32 detailed points
Applied, detailed content
Accredited attendance certificate
On completing the programme
Complete your registration
We will contact you within one business day to confirm.
Ready to start?
Reserve your seat and start building the skill.