The regenerative Tuned Mass Damper (TMD) can convert the vibration energy of the tall building into the electricity, by replacing the damping element with electromagnetic harvester. The energy harvesting circuit therein which can regulate the electricity and control the vibration will introduce some constraints when designing vibration controller. This paper designed the vibration controller based on Model Predictive Control (MPC). The control force constraints were taken into consideration before designing the controller. The building model with semi-active constraints due to the regenerative properties of the TMD is converted into a Mixed Logical Dynamical (MLD) system first. Then the optimal controller is designed by solving the Mixed Integer Quadratic Programming (MIQP) problem. The results were evaluated and compared to the ones using “clipped-optimal” controller with the same constraints. It is found that the MPC controller can provide the same or better vibration control Results depending on the predicted horizon. Besides, an explicit MPC is obtained to reduce the online computation effort.
A Numerically Robust Mixed-Integer Quadratic Programming Solver for Embedded Hybrid Model Predictive Control
