Model predictive control for LiDAR
Assessing the new design capabilities of LiDAR assisted model predictive control (MPC) applied to wind turbines.
Objective
To develop new design capabilities for the modern control strategy of model predictive control (MPC) applied to wind turbines.
Results
- Field test data obtained at NREL (National Renewable Energy Laboratory) and analysed to confirm performance improvements using simple LiDAR-assisted feed-forward control.
- Design process and controller structure defined for MPC.
- MPC implemented in controller software and tested using Bladed. The advantages over classical control were evaluated.
- Publications at industry conferences.
Value created
- A new approach to wind turbine control design was developed using MPC theory.
- This new service has vast potential for more robust controller design.
- MPC approaches include system constraints, disturbances, coupling and nonlinearity when determining the optimum control actions.
Quote from DNV project expert
“Today we can measure the environment affecting our systems better than ever before. Model predictive control allows us to couple this new information with our state-of-the-art system modelling capabilities to provide the next level of turbine closed loop performance for our clients.”
Avishek Kumar, Control Algorithm Designer, DNV - Energy
