Topology optimization of high-speed rail bridges considering passenger comfortGolecki, Thomas; Gomez, Fernando; Carrion, Juan; Spencer, Billie F.
2023 Structural and Multidisciplinary Optimization
doi: 10.1007/s00158-023-03666-x
Worldwide growth in high-speed rail (HSR) networks has brought a demand for improved structural performance of the bridges that make up a large portion of many of these HSR systems. While some improvement has been made via optimization either of the bridge or the passenger cars, the design criteria of passenger comfort has yet to be addressed in bridge optimization. The transient dynamics of vehicle–bridge interaction between a high-speed train and bridge make such structural optimization of these systems challenging. In this paper, we derive an approach for topology optimization of high-speed rail bridges including vehicle–bridge interaction that enables direct consideration of the passenger comfort in the objective function. Assuming constant contact between the vehicle’s wheels and the bridge, the two systems are combined into a single-state space system. The resulting system matrices are time dependent, as they are a function of the wheel contact locations which change as the vehicles move over the bridge. The equations of motion and the adjoint sensitivities are derived and solved numerically in the time domain. Several numerical examples are provided based on high-speed rail applications that minimize a multi-objective function comprised of bridge and vehicle responses, including passenger comfort. These examples generate topologies that improve passenger comfort at only a small cost to the bridge response and demonstrate the dependence of optimal topology on train speed and length. The proposed method offers the potential for improving high-speed rail passenger comfort through optimization of bridge topology by accounting for the vehicle–bridge interaction effects.
Resilience-based design optimization of engineering systems under degradation and different maintenance strategyWang, Zhonglai; Wen, Yang; Wang, Zhihua; Zhi, Pengpeng
2023 Structural and Multidisciplinary Optimization
doi: 10.1007/s00158-023-03671-0
Resilience-based design optimization is an effective way to design the high resilience into engineering systems during the design stage. Addressing the continuous degradation, discrete shocks, and different maintenance strategy, the resilience-based design optimization model and algorithms are proposed to ensure the operational safety of engineering systems in this paper. The unified framework of resilience under the continuous degradation, discrete shock, and different maintenance strategy is first proposed. The surrogate model to estimate the time-variant reliability under the continuous stochastic process and discrete Poisson process is then constructed. The mapping function between the performance improvement ratio and the maintenance strategy is defined to build the lifecycle cost models. The combination of Kriging surrogate model and particle swarm optimization is finally provided to conduct resilience-based design optimization. An engineering case of a harmonic reducer used for the lower extremity exoskeleton is employed to illustrate and testify the effectiveness of the proposed methods.
Multiscale concurrent topology optimization for thermoelastic structures under design-dependent varying temperature fieldGuo, Yanding; Wang, Yi; Wei, Dong; Chen, Lijie
2023 Structural and Multidisciplinary Optimization
doi: 10.1007/s00158-023-03649-y
In this paper, a new multiscale concurrent topology optimization method for thermoelastic structures considering the iterative variation of temperature field is proposed for the first time, which breaks the limitation that previous multiscale concurrent topology optimization studies being compliable merely to uniform temperature field. In this method, the iterative variations of macroscopic structural heat transfer, structural temperature, structural force transfer, structural displacement, design-dependent thermal stress load, microscopic effective thermal conductivity, effective elasticity and effective thermal expansion coefficient are all taken into consideration. In order to establish a compact hierarchical thermoelastic coupling equation on the above iterative factors, firstly, a thermoelastic coupling matrix with a distinct physical meaning is proposed to address the issues on accuracy of thermal stress loads and solution of adjoint sensitivity multipliers caused by design-dependent varying temperature field, and this matrix can be used as a new manner to solve homogenized effective thermal stress coefficient. Secondly, the compact coupling equation is derived using multiscale adjoint sensitivity analysis and its effectiveness is illustrated by comparative cases. Finally, the generality and stability of proposed method are illustrated through diverse scenarios involving compliance optimization, multimaterial concurrent design, maximum displacement control, multicellular structure design, asymmetric boundary conditions and three-dimensional structures. It is obvious that this pioneering approach has a broad potential in advanced integrated structures and materials design of thermoelastic structures.
Robust optimal design and trajectory planning of an aircraft with morphing airfoil sectionsRudnick-Cohen, Eliot S.; Reich, Gregory W.; Pankonien, Alexander M.; Beran, Philip S.
2023 Structural and Multidisciplinary Optimization
doi: 10.1007/s00158-023-03664-z
Aircraft performance is heavily dependent on how an aircraft’s design interacts with the flight trajectories and missions that it flies. An optimal aircraft must, thus, optimize both the physical design of the aircraft alongside its flight trajectories. However, an aircraft design optimized for either a single trajectory or a small number of trajectories is likely to be overspecialized, and may perform poorly when flying other trajectories. This paper presents an approach for designing an airfoil for a morphing aircraft wing while optimizing its worst-case flight time to states defined within a continuous range of different flight states. Sampling-based motion planning techniques are used to plan optimal flight trajectories for each design evaluated during design optimization. Several improvements to methods for using sampling-based motion planning as a subproblem within a design optimization problem are presented. An approach for defining a suitable continuous set of flight states is proposed to ensure that only reachable states are used when determining the optimal aircraft design. A robust optimization approach is used in order to solve the resulting optimization problem. Results are presented comparing the design found using the proposed robust optimization approach against a multi-objective design and motion planning approach previously used in the literature. The results show that the designs found using the robust optimization approach have better worst-case flight times than those found using the multi-objective approach. The results also indicate that the robust optimization approach is able to find a design which avoids overspecialization, unlike those found by the multi-objective approach.
Gradient-based adaptive sampling framework and application in the laser-driven ion accelerationWang, Binglin; Sha, Rong; Yan, Liang; Yu, Tongpu; Duan, Xiaojun
2023 Structural and Multidisciplinary Optimization
doi: 10.1007/s00158-023-03669-8
Physical model optimisation has frequently been complemented by experimental design in scientific research. However, it can be time consuming to perform real-world experiments and difficult to find affordable experimental designs. Bayesian optimisation based on the Gaussian process model has attracted extensive attention in the field of experimental design because it can build a good surrogate model and generate a sequential design simultaneously. However, it can create problems if researchers have a weak understanding of the system’s overall trend. This study introduces gradient information and proposes a new framework for constructing surrogate models: GRAdient-enhanced SEquential SUrrogate MOdelling (GRASE-SUMO). First-order gradient information is utilised as a guidance for selecting sampling space, and second-order gradient information is then adopted as an objective function in Bayesian optimisation. GRASE-SUMO is designed to mimic system changes and allows general system trends to be easily identified without a high level of prior knowledge. Experiments were conducted to verify the accuracy and stability of GRASE-SUMO, which works especially well in dealing with plate-shaped or valley-shaped response surfaces. When applied to laser-proton acceleration, GRASE-SUMO succeeded in rectifying and expanding the suitable conditions for optimal acceleration using only 30 samples, while the conventional sampling method requires about 102-3\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$^{2-3}$$\end{document} samples with only three variables.