Reliability‐based control algorithms for nonlinear hysteretic systems based on enhanced stochastic averaging of energy envelope

摘要: Summary Current reliability-based control techniques have been successfully applied to linear systems; however, incorporation of stochastic nonlinear behavior systems in such designs remains a challenge. This paper presents two algorithms that minimize failure probabilities hysteretic subjected excitations. The proposed methods include constrained (CRC) and unconstrained (URC) algorithms. Accurate probabilistic estimates system responses excitations are derived analytically using enhanced averaging energy envelope previously by the authors. Convolving these demand with capacity models yields reliability design process. CRC employs first-level second-level optimizations sequentially where optimization solves Hamilton–Jacobi–Bellman equation searches for optimal objective function parameters probability failure. In URC design, single minimizes directly searching gain. Application building on foundation has shown noticeable improvements performance under various appears be most method as it reduced slight damage 8.7%, compared 11.6% 19.2% case quadratic regulator, respectively. Under Kobe ground motion, normalized peak drift displacement respect regulator is 0.78 0.81 cases, respectively, at comparable force levels. Copyright © 2017 John Wiley & Sons, Ltd.