A5102007825- Probabilistic and Robust Engineering Design
- Structural Health Monitoring Techniques
- Industrial Technology and Control Systems
- Simulation Techniques and Applications
- Risk and Portfolio Optimization
- Simulation and Modeling Applications
- Ultrasonics and Acoustic Wave Propagation
- Spreadsheets and End-User Computing
- Advanced Database Systems and Queries
- Non-Destructive Testing Techniques
- Advanced Statistical Process Monitoring
- Manufacturing Process and Optimization
- Advanced Queuing Theory Analysis
- Water resources management and optimization
- Cancer-related molecular mechanisms research
- Iterative Methods for Nonlinear Equations
- Statistical Methods and Inference
- Advanced Sensor and Control Systems
- Matrix Theory and Algorithms
- Educational Technology and Assessment
- MicroRNA in disease regulation
- Markov Chains and Monte Carlo Methods
- Metallurgy and Material Forming
- Concrete Corrosion and Durability
- Advanced Algorithms and Applications
Hainan University
2023-2025
Nantong University
2023-2024
Shanghai Tunnel Engineering Rail Transit Design & Research Institute
2023
China Academy of Launch Vehicle Technology
2014-2023
Shanghai Jiao Tong University
2023
Amazon (United States)
2022
Georgia Institute of Technology
2007-2021
Dalian University of Technology
2009-2019
Simon Fraser University
2018
North China Electric Power University
2017
The performances of engineering structures are degrading in service and maintenance interventions can enhance the structural reliability make sure that effectively functional. Engineering often exhibit obvious time cumulative effects due to multisource time-varying uncertainties including property degradation materials, changeable environment conditions, dynamic loading processes, which makes assessment design much harder. In this paper, a new nonprobabilistic reliability-oriented optimal...
A large class of stochastic programs involve optimizing an expectation taken with respect to underlying distribution that is unknown in practice. One popular approach addressing the distributional uncertainty, known as distributionally robust optimization, hedge against worst case over uncertainty set candidate distributions. However, it has been observed inappropriate construction can sometimes result overconservative solutions. To explore middle ground between optimistically ignoring and...
Temperature and load variations significantly affect ultrasonic guided wave propagation therefore lead to the increasing of diagnostic uncertainty wave-based structural health monitoring system, which may cause false reports positives limit practical application method. Most existing researches focus on compensation for a single factor. Therefore it's an important challenge that how compensate waves under influence multiple environmental factors. This study firstly analyzes coupled effect...
In many applications, input data are collected frequently to update the simulation model of system, whereas is run compare different designs/strategies identify best one with a high confidence. “Data-Driven Ranking and Selection Under Input Uncertainty,” Wu, Wang, Zhou consider such simulation-based ranking selection (R&S) problem, in which distribution estimated updated arriving batches over time. Unlike most existing works R&S that conduct under fixed distribution, this data-driven...
An impact monitoring method based on time series analysis and a sparse sensor network is proposed to identify the location of event that occurs aircraft complex structures estimate energy. Generally, high-density required for high localization accuracy event, but with trade-off deployment costs. A piezoelectric together coarse fine two-step designed balance cost. The stress wave signals caused by are measured regarded as series. problem converted into classification series, strategy...
With the increasing design dimensionality, it is more difficult to solve multidisciplinary optimization (MDO) problems. Many MDO decomposition strategies have been developed reduce dimensionality. Those consider problem as a black-box function. However, practitioners usually certain knowledge of their problem. In this paper, method leveraging causal graph and qualitative analysis dimensionality by systematically modeling incorporating about into optimization. Causal created show input–output...
A widely acknowledged challenge in ranking and selection is how to allocate the simulation budget such that probability of correction (PCS) maximized. However, there yet another challenge: when input distributions are estimated using finite real-world data, output subject uncertainty we may fail identify best system even infinite budget. We propose a new formulation captures tradeoff between collecting data running simulations. To solve formulation, develop an algorithm for two-stage...
Multifunctional sensor network has become a research focus in the field of structural health monitoring. To improve reliability and stability diagnosis results, it is necessary to fuse heterogeneous signals under interference external load damage. In this paper, piezoelectric-fiber hybrid integrated monitor crack growth around hole aviation aluminum plate. The effect change on piezoelectric transducers (PZTs) optical fiber sensors analyzed. damage result obtained by ultrasonic guided wave...
In stochastic simulation, input uncertainty (IU) is caused by the error in estimating distributions using finite real-world data. When it comes to simulation-based Ranking and Selection (R&S), ignoring IU can lead failure of many existing procedures. this paper, we study a new version fixed confidence R&S problem, where sequential data be acquired reduce over time. To solve first propose moving average estimator for online estimation with Then, procedure designed extending Sequential...
In this article, we aim to solve Bayesian Risk Optimization (BRO), which is a recently proposed framework that formulates simulation optimization under input uncertainty. order efficiently the BRO problem, derive nested stochastic gradient estimators and propose corresponding approximation algorithms. We show our are asymptotically unbiased consistent, algorithms converge asymptotically. demonstrate empirical performance of on two-sided market model. Our independent interest in extending...