Overview

Model Predictive Control (MPC) is the most successful advanced multivariable control technique used in applications across industrial sectors, particularly where high performance is required. It is particularly valuable for complex systems that are difficult to control due to difficult dynamics, interactions, multivariable nature, noise, disturbances, transport delays, constraints, or nonlinearities.

The MPC algorithm reliably handles multiple input/output variables and constraints and introduces prediction capabilities to enhance the accuracy and performance of industrial processes or electromechanical control systems over a future horizon. While the full mathematical background is not necessary, an intuitive understanding of the MPC design stages and computations is crucial.

This training course covers basic ideas, design stages, and implementation steps to avoid commissioning or operational difficulties. It provides an intuitive introduction to MPC, detailing model choice, cost-function definition, optimisation methods, and the form of optimal control solutions.

The course targets engineers interested in applying MPC and need to understand the design procedures and enhancements for faster and more effective practical implementation.

Course Structure

Early presentations will motivate the use of MPC and introduce necessary physical system modelling methods or system identification techniques. Most predictive control laws are obtained by optimising a criterion representative of the required control, whether for energy minimisation or a general criterion where the cost-function weightings act rather like PID tuning parameters.

Optimisation methods for both constrained and unconstrained MPC problems will be covered, including implementation of hard constraints in the optimisation algorithms.

After presenting basic modelling, optimisation, and optimal control algorithms, the focus will shift to designing systems to meet industrial requirements and specifications, considering system, noise, disturbance models, and the form and parameterisation of the cost-function.

The design stage will address choosing the cost-function and weighting terms. MPC controllers can be assessed via simulation, but real-world implementation may face issues like inaccuracies, uncertainties, and robustness. Various enhancements to basic MPC can mitigate these problems, including introducing integral action, allowing for different transport delays, and ways for dealing with unknown disturbances.

Artificial intelligence (AI) and machine learning will significantly impact on MPC's future use and implementation. Adaptive AI-based predictive controls and data-driven methods, such as neural networks or support vector machines, can enhance MPC algorithms. AI predictors can also provide longer-range prediction horizon. The final presentation is a look to the future of AI-enhanced MPC to indicate the developments that companies should be aware of to maintain a competitive edge.

Hands-on Simulink exercises throughout the course will ensure understanding of modelling and optimisation methods. A servosystem will be used for all examples, from initial modelling to final implementation. Different application options can be offered through tailored versions of the course.

Instructors

Professor Michael Grimble

Professor Mike Grimble understands the needs of industry well, having worked for Ciba Geigy and Associated Electrical Industries (later GEC at Rugby). At Imperial College's Industrial Automation Group, he focused on modelling for the control of cold rolling mills. He later established industrial groups at Sheffield Hallam University and the University of Strathclyde where he continues as a Research Professor. His expertise lies in designing high-performance, robust control systems for various industrial applications. His industrial background enriches the industrial training courses with valuable motivation and insights.

Dr Pawel Majecki

Dr. Pawel Majecki conducted his research at the Industrial Control Centre at Strathclyde University before joining Industrial Systems and Control Ltd (ISC), which promotes technology transfer. He has worked with major international companies, applying advanced control methods, including predictive and optimal controls. Dr. Majecki has also led control training courses in the UK, Norway, Italy, Spain, and the USA. His extensive experience with MATLAB/Simulink enhances the hands-on training, helping delegates gain deeper insights into these tools and their application in design methods.

Delegates will find both instructors delighted to answer questions and discuss industrial problems during coffee breaks, lunch breaks, and at the end of the day.

Additional Information

Event Venue

Scottish Engineering, 105 W George St, Glasgow G2 1QL

Accommodation

Glasgow City Centre offers a wide range of accommodation, you can find our recommendations here.

Course Summary and Agenda (Download PDF)

Timings may change slightly.

Prices and Discounts

• Registration before 1st May 2025: £1,428 + VAT per delegate
• Registration on/after 1st May 2025: £1,608 + VAT per delegate

For two or more places from the same organisation, each additional place is 10% off the single place fee.

Registration

Please complete the Online Registration Form.

Can't Find the Course You Need?

Try our bespoke course service.

Tell us your training needs.