TY - GEN
T1 - Reliability-based co-design for state-constrained stochastic dynamical systems
AU - Cui, Tonghui
AU - Allison, James
AU - Wang, Pingfeng
N1 - Publisher Copyright:
© 2020, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.
PY - 2020
Y1 - 2020
N2 - In this article, we propose a novel framework for reliability-based co-design of stochastic dynamical systems. Specifically, it solves a combined reliability-based design optimization problem of the plant design with parametric uncertainties and the state-constrained stochastic optimal control. In the past, co-design strategies have been applied primarily in a deterministic manner, often employing open-loop optimal control techniques in generating solutions and system design insights. Here we extend the scope of co-design methods to a specific class of stochastic dynamical systems by providing a stationary solution to its Hamiltonian-Jacobi-Bellman equation, based on logarithmic transformation and linearization. For reliability assessment, Monte Carlo Simulation, Performance Measurement Approach, and Sequential Optimization and Reliability Assessment methods are evaluated with respect to accuracy and computational expense. The proposed method is demonstrated and assessed using both an analytical example and a co-design problem involving an actively controlled spring-mass-damper system.
AB - In this article, we propose a novel framework for reliability-based co-design of stochastic dynamical systems. Specifically, it solves a combined reliability-based design optimization problem of the plant design with parametric uncertainties and the state-constrained stochastic optimal control. In the past, co-design strategies have been applied primarily in a deterministic manner, often employing open-loop optimal control techniques in generating solutions and system design insights. Here we extend the scope of co-design methods to a specific class of stochastic dynamical systems by providing a stationary solution to its Hamiltonian-Jacobi-Bellman equation, based on logarithmic transformation and linearization. For reliability assessment, Monte Carlo Simulation, Performance Measurement Approach, and Sequential Optimization and Reliability Assessment methods are evaluated with respect to accuracy and computational expense. The proposed method is demonstrated and assessed using both an analytical example and a co-design problem involving an actively controlled spring-mass-damper system.
UR - http://www.scopus.com/inward/record.url?scp=85092388345&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092388345&partnerID=8YFLogxK
U2 - 10.2514/6.2020-0413
DO - 10.2514/6.2020-0413
M3 - Conference contribution
AN - SCOPUS:85092388345
SN - 9781624105951
T3 - AIAA Scitech 2020 Forum
BT - AIAA Scitech 2020 Forum
PB - American Institute of Aeronautics and Astronautics Inc, AIAA
T2 - AIAA Scitech Forum, 2020
Y2 - 6 January 2020 through 10 January 2020
ER -