Multi-Objective Gradient-Based Optimization of Double Control Laws for Active Mass Dampers Considering Multi-Hazard Life-Cycle Cost

  • Kleingesinds, Shalom (Ariel University)

Please login to view abstract download link

The use of performance-based design for tall buildings has attracted increasing attention in recent years, including the consideration of multiple hazard load effects and their integration with protective systems design. A double control law strategy has recently been proposed for Active Mass Dampers (AMDs) in tall buildings subjected to multi-hazard loading. In this strategy, distinct control laws are employed for wind and seismic actions and designed using an LQR optimal control formulation with probabilistic constraints. This approach has been shown to outperform analogous single control laws, achieving reductions of 13.4% to 27.7% in total AMD cost for a prescribed performance target. To extend AMD design beyond single-target optimization, this work presents a multi-objective framework, in which both the initial cost of the AMD and the multi-hazard life-cycle cost (LCC) associated with long-term hazard-induced restoration are minimized simultaneously. Due to the high computational cost of the resulting optimization problem, a gradient-based optimization algorithm is developed. Analytical sensitivity analyses are derived and validated against equivalent zero-order solutions. This formulation is embedded within a multi-stage multi-objective optimization procedure, whose final stage adopts a population-based aggregative strategy to efficiently identify Pareto-optimal solutions. Nine design cases are investigated, each characterized by different constraints on the maximum control force and stroke, leading to nine Pareto fronts of optimized solutions. The results reveal clear trade-offs between force-limited and stroke-limited designs and demonstrate the effectiveness of the proposed strategy in reducing multi-hazard structural responses. The proposed framework enables systematic exploration of the AMD design space and provides a systematic and computationally efficient basis for cost–benefit assessment in performance-based multi-hazard design of tall buildings.