Multi-Step Model Predictive Control for Power Electronics and Electrical Drives | Research Capability & Technology Portfolio

Model predictive control has emerged as a promising alternative control technique for power electronic applications which provides rapid dynamics and responses to demand changes. It can handle multiple variables and system constraints and achieve a fast, dynamic response.

Competitive advantage

The use of a computationally efficient optimizer, namely sphere decoding algorithm (SDA)
Experience in the practical application of n-multistep MPC using the ‘Sphere Decoding Algorithm’ in power electronics systems and electrical drives

Impact

Demonstrable improvement in electrical machine and drive performance
Ability to predict future horizon response and control capabilities
Reduced harmonic distortion and higher efficiency in machine drives and converter systems

Successful applications

These advanced control techniques have been applied, and shown to be effective in:
Conventional induction machine drives
Three-level inverter systems including neutral-point clamped and flying capacitor technologies

Capabilities and facilities

Rapid prototyping of multistep predictive control techniques
Multiple testing platforms including machine types and inverter topologies