Multi-surrogate-based global optimization using a score-based infill criterion

Dong, Huachao; Sun, Siqing; Song, Baowei; Wang, Peng

doi:10.1007/s00158-018-2079-z

Cited by 36 publications

(9 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Selecting infill points requires information about model outputs and is generally more complex than space filling methods, such as maximin LHS, that only consider model inputs. There are a number of recent output-oriented adaptive design methods that can iteratively develop informative infill points as an optimization algorithm proceeds (Dong et al, 2019;Liu et al, 2018;Mo et al, 2017). In fact, many of these techniques choose new samples using a score function that involves a local exploitation term and a global exploration term (Liu et al, 2018), a concept that is central to the operation of the PSO algorithm itself.…”

Section: Adaptive Resampling Via Swarm Intelligencementioning

confidence: 99%

Reduced‐Dimensional Gaussian Process Machine Learning for Groundwater Allocation Planning Using Swarm Theory

Siade

Cui

Karelse³

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

Groundwater management and allocation planning involves a rigorous assessment of the performance of operational decisions such as extraction/injection rates on community and environmental objectives. Maximizing performance through numerical optimization can be essential for high‐value resources and is often computationally infeasible due to long simulation model run times combined with nonconvex objectives and constraints. In order to mitigate these drawbacks, surrogate models can be used in place of complex models during the optimization process. There exist a number of machine learning techniques that can be used to develop a data‐driven surrogate model. However, the curse of dimensionality, common to groundwater management, limits the use of these techniques due to the necessity for large training data sets. Even though it is now possible to handle large data sets, the generation of these data sets themselves remains computationally prohibitive as they require numerous simulations to produce accurate surrogates. In this study, we integrate a dimensionality reduction method using truncated singular value decomposition to reduce the number of decision variables, thereby reducing the size of the training data set needed. Correspondingly, we demonstrate a simple technique for acquiring an approximate minimax Latin Hypercube design from within the subspace. We also implement a novel technique for adaptive resampling through particle swarm optimization in order to maintain accuracy of the surrogate model throughout the optimization process. The resulting accurate surrogate model for the Perth regional aquifer system of Western Australia runs in a matter of seconds. Adopting this approach can produce timely solutions, making formal optimization tractable for practitioners.

show abstract

Section: Adaptive Resampling Via Swarm Intelligencementioning

confidence: 99%

Reduced‐Dimensional Gaussian Process Machine Learning for Groundwater Allocation Planning Using Swarm Theory

Siade

Cui

Karelse³

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…Unlike the direct and bi‐univocal relation between the value of the ECMS equivalent factor and the value of SOC reached at the cycle final time, 39 the relation between λ C and the final SOC is a computation‐intensive, black‐box problem. Recently, advanced meta‐model based global optimization algorithm has been found as an effective tool to solve this type of problems 36,41,42 . The multi‐start space reduction (MSSR) global optimization algorithm 43 replaces the shooting method to find the optimal cost equivalent factor minimizing the SOC difference value between the initial SOC and final SOC :

\min : ΔSOC = (||, {SOC}_{final} - {SOC}_{init}),

…”

Section: Optimal Energy Management Strategy Solutionmentioning

confidence: 99%

Optimal energy management strategy of fuel‐cell battery hybrid electric mining truck to achieve minimum lifecycle operation costs

Feng

Dong

2020

Int J Energy Res

View full text Add to dashboard Cite

Summary Proton exchange membrane fuel cell (PEMFC) electric vehicle is an effective solution for improving fuel efficiency and onboard emissions, taking advantage of the high energy density and short refuelling time. However, the higher cost and short life of the PEMFC system and battery in an electric vehicle prohibit the fuel cell electric vehicle (FCEV) from becoming the mainstream transportation solution. The fuel efficiency‐oriented energy management strategy (EMS) cannot guarantee the improvement of total operating costs. This paper proposes an EMS to minimize the overall operation costs of FCEVs, including the cost of hydrogen fuel, as well as the cost associated with the degradations of the PEMFC system and battery energy storage system (ESS). Based on the PEMFC and battery performance degradation models, their remaining useful life (RUL) models are introduced. The control parameters of the EMS are then optimized using a meta‐model based global optimization algorithm. This study presents a new optimal control method for a large mining truck operating on a real closed‐road operation cycle, using the combined energy efficiency and performance degradation cost measures of the PEMFC system and lithium‐ion battery ESS. Simulation results showed that the proposed EMS could improve the total operating costs and the life of the FCEV.

show abstract

“…Moreover, when pointwise EMs are applied to engineering design optimization, the computational cost dramatically increases, because a large number of predictions using surrogates are required. In addition to using EMs, some researchers have developed optimization algorithms based on multi‐surrogates, where new samples are selected according to their predictive values from these multiple metamodels 22–24 …”

Section: Introductionmentioning

confidence: 99%

Adaptive multi‐surrogate‐based constrained optimization method and its application

Han

2021

Numerical Meth Engineering

View full text Add to dashboard Cite

This article presents an adaptive multi‐surrogate constrained optimization method (AMSCOM) that can automatically determine the appropriate metamodel for each black‐box function in the constrained optimization problem (COP) and concurrently find the optimum. In AMSCOM, each black‐box function is approximated initially by several different types of candidate surrogates. Then, as optimization progresses, the poorly performing candidate surrogates of each black‐box function are gradually eliminated until the appropriate surrogate is found. Meanwhile, as more than one candidate surrogate exists for each unknown function in the optimization process, multiple approximate optimization problems (AOPs) can be constructed, and new samples can be obtained by solving these AOPs. Additionally, we employ the genetic operator and the local‐linear approximation–Voronoi method to generate new samples. To verify the effectiveness and investigate several properties of AMSCOM, the proposed method is tested on 12 benchmark COPs and compared with several single surrogate‐based methods. Furthermore, AMSCOM is compared with several published surrogate‐based constrained optimization methods, and the results further prove the superior performance of AMSCOM. The proposed method is then employed to optimize the shaft‐clinching process of wheel‐hub‐bearing units, and a desirable result is achieved.

show abstract

Multi-surrogate-based global optimization using a score-based infill criterion

Cited by 36 publications

References 47 publications

Reduced‐Dimensional Gaussian Process Machine Learning for Groundwater Allocation Planning Using Swarm Theory

Reduced‐Dimensional Gaussian Process Machine Learning for Groundwater Allocation Planning Using Swarm Theory

Optimal energy management strategy of fuel‐cell battery hybrid electric mining truck to achieve minimum lifecycle operation costs

Adaptive multi‐surrogate‐based constrained optimization method and its application

Contact Info

Product

Resources

About