Multi-start Space Reduction (MSSR) surrogate-based global optimization method

Dong, Huachao; Song, Baowei; Dong, Zuomin; Wang, Peng

doi:10.1007/s00158-016-1450-1

Cited by 65 publications

(28 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most significant collections of global optimization benchmark problems can be found in [18,86,87]. Each benchmark problem has its individual characteristic, such as whether it is a multimodal, a unimodal or a random shape function as illustrated in Figure 1.…”

Section: Benchmark Function and Experiments Materialsmentioning

confidence: 99%

A Comparative Study on Recently-Introduced Nature-Based Global Optimization Methods in Complex Mechanical System Design

2017

Self Cite

View full text Add to dashboard Cite

Advanced global optimization algorithms have been continuously introduced and improved to solve various complex design optimization problems for which the objective and constraint functions can only be evaluated through computation intensive numerical analyses or simulations with a large number of design variables. The often implicit, multimodal, and ill-shaped objective and constraint functions in high-dimensional and "black-box" forms demand the search to be carried out using low number of function evaluations with high search efficiency and good robustness. This work investigates the performance of six recently introduced, nature-inspired global optimization methods: Artificial Bee Colony (ABC), Firefly Algorithm (FFA), Cuckoo Search (CS), Bat Algorithm (BA), Flower Pollination Algorithm (FPA) and Grey Wolf Optimizer (GWO). These approaches are compared in terms of search efficiency and robustness in solving a set of representative benchmark problems in smooth-unimodal, non-smooth unimodal, smooth multimodal, and non-smooth multimodal function forms. In addition, four classic engineering optimization examples and a real-life complex mechanical system design optimization problem, floating offshore wind turbines design optimization, are used as additional test cases representing computationally-expensive black-box global optimization problems. Results from this comparative study show that the ability of these global optimization methods to obtain a good solution diminishes as the dimension of the problem, or number of design variables increases. Although none of these methods is universally capable, the study finds that GWO and ABC are more efficient on average than the other four in obtaining high quality solutions efficiently and consistently, solving 86% and 80% of the tested benchmark problems, respectively. The research contributes to future improvements of global optimization methods.

show abstract

Section: Benchmark Function and Experiments Materialsmentioning

confidence: 99%

A Comparative Study on Recently-Introduced Nature-Based Global Optimization Methods in Complex Mechanical System Design

2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Each of these metamodel techniques has its advantages and disadvantages. Recently, methods for generating hybrid models by combining these metamodels have been developed . We call such hybrid model an ensemble model or multiple surrogates.…”

Section: Introductionmentioning

confidence: 99%

“…Recently, methods for generating hybrid models by combining these metamodels have been developed. [10][11][12][13][14][15][16][17][18][19][20][21] We call such hybrid model an ensemble model or multiple surrogates. The performance of metamodel-based optimization algorithm depends on the reliability of the metamodel.…”

Section: Introductionmentioning

confidence: 99%

An approximate optimization strategy using refined hybrid metamodel

Lee

Jeong

Lee

2019

Computational Intelligence

View full text Add to dashboard Cite

An approximate model called metamodel or surrogate model is a mathematical model that numerically approximates response of a system during an engineering simulation process or test. The introduction of a metamodel makes it possible to express response defined in the design problem as a simple mathematical function of design variables. A metamodel can be built with response surface method (RSM), kriging, neural network, radial basis function, and so on. Each method has its advantages and disadvantages. A combined metamodel called hybrid model, ensemble model, or multiple surrogates has been developed to maximize each metamodel's strength. The hybrid model of this research includes RSM and kriging. Besides, a strategy to refine the hybrid metamodel is implemented by reducing design space. In this process, information related to Hessian is utilized for an unconstrained optimization problem, on the contrary feasibility for a constrained optimization problem. This research presents a new hybrid metamodel‐based optimization strategy called refined hybrid metamodel. Five mathematical test problems, two‐bar design, spring design, and propeller shaft design problems are solved with the suggested method, verifying its usefulness. Most of the optimal results with the proposed method are closer to exact solutions with smaller function evaluations than existing methods.

show abstract

“…Multi-start Space Reduction (MSSR) surrogate-based global optimization method is a most recently MBGO method in our research for the applications with black-box and computationally intensive applications (Dong et al [25]). In this new algorithm, the design space is classified into: the original design space or global space (GS), the reduced medium space (MS) that contains the promising region, and the local space (LS) that is a local area surrounding the present best solution in the search.…”

Section: Mssr Algorithmmentioning

confidence: 99%

“…The approach uses limited "expensive" sample data points from the original, computationally expensive optimization model to introduce a surrogate model, or metamodel, such as Kriging [22] and radial basis functions (RBF) [23], and to effectively use the "cheaper" sample points from the metamodel to speed up the search of the global optimum with much reduced number of original model evaluations and computational time. Several reviews have systematically presented the advantages of these algorithms [21,[24][25][26]. The newly introduced algorithms have been compared with other existing GO techniques using about a dozen of standard benchmarks optimization problems to study their capability to accurately locate the global optimum, search efficiency and computation time.…”

Section: Introductionmentioning

confidence: 99%

Application and Comparison of Metaheuristic and New Metamodel Based Global Optimization Methods to the Optimal Operation of Active Distribution Networks

et al. 2018

Self Cite

View full text Add to dashboard Cite

As an imperative part of smart grids (SG) technology, the optimal operation of active distribution networks (ADNs) is critical to the best utilization of renewable energy and minimization of network power losses. However, the increasing penetration of distributed renewable energy sources with uncertain power generation and growing demands for higher quality power distribution are turning the optimal operation scheduling of ADN into complex and global optimization problems with non-unimodal, discontinuous and computation intensive objective functions that are difficult to solve, constituting a critical obstacle to the further advance of SG and ADN technology. In this work, power generation from renewable energy sources and network load demands are estimated using probability distribution models to capture the variation trends of load fluctuation, solar radiation and wind speed, and probability scenario generation and reduction methods are introduced to capture uncertainties and to reduce computation. The Open Distribution System Simulator (OpenDSS) is used in modeling the ADNs to support quick changes to network designs and configurations. The optimal operation of the ADN, is achieved by minimizing both network voltage deviation and power loss under the probability-based varying power supplies and loads. In solving the computation intensive ADN operation scheduling optimization problem, several novel metamodel-based global optimization (MBGO) methods have been introduced and applied. A comparative study has been carried out to compare the conventional metaheuristic global optimization (GO) and MBGO methods to better understand their advantages, drawbacks and limitations, and to provide guidelines for subsequent ADN and smart grid scheduling optimizations. Simulation studies have been carried out on the modified IEEE 13, 33 and 123 node networks to represent ADN test cases. The MBGO methods were found to be more suitable for small- and medium-scale ADN optimal operation scheduling problems, while the metaheuristic GO algorithms are more effective in the optimal operation scheduling of large-scale ADNs with relatively straightforward objective functions that require limited computational time. This research provides solution for ADN optimal operations, and forms the foundation for ADN design optimization.

show abstract

Multi-start Space Reduction (MSSR) surrogate-based global optimization method

Cited by 65 publications

References 38 publications

A Comparative Study on Recently-Introduced Nature-Based Global Optimization Methods in Complex Mechanical System Design

A Comparative Study on Recently-Introduced Nature-Based Global Optimization Methods in Complex Mechanical System Design

An approximate optimization strategy using refined hybrid metamodel

Application and Comparison of Metaheuristic and New Metamodel Based Global Optimization Methods to the Optimal Operation of Active Distribution Networks

Contact Info

Product

Resources

About