Decentralised Predictive Control with Energy Dissipation Bounds for Wireless Structural Control Applications

Decentralized control architectures for multi-storey buildings using wireless sensors and distributed embed- ding systems have been attracting more research recently. The designs for this application are however limited by the com- putational capacity embedded in each subsystem, the communication bandwidth and range of wireless channels, as well as the temporal constraints for real-time seismic protections. This paper presents a novel decentralized predictive control (DPC) strategy which can meet these design constraints. The control design takes into account the guaranteed closed-loop stability for the large-scale structure which is established from the dissipative systems theory. The proposed DPC strategy features two offline control laws to be switched-over online to minimize the computational and communication demands. By emu- lating the energy dissipation rate with a stabilizing constraint for the DPC optimization, the seismic protection performance of structural networked controls can be improved. The approach is designated with energy-dissipation DPC for structures using semiactive dampers. The approximated polynomial models of magneto-rheological (MR) dampers are adopted for linear systems in decentralized control designs. A numerical example is provided to illustrate the analytical developments.