Search results

Highly non-linear optimization problems exist in many practical engineering applications. To deal with these problems, this study aims to propose an improved optimization algorithm, named, adaptive resistance and stamina strategy-based dragonfly algorithm (ARSSDA).

To speed up the convergence, ARSSDA applies an adaptive resistance and stamina strategy (ARSS) to conventional dragonfly algorithm so that the search step can be adjusted appropriately in each iteration. In ARSS, it includes the air resistance and physical stamina of dragonfly during a flight. These parameters can be updated in real time as the flight status of the dragonflies.

The performance of ARSSDA is verified by 30 benchmark functions of Congress on Evolutionary Computation 2014’s special session and 3 well-known constrained engineering problems. Results reveal that ARSSDA is a competitive algorithm for solving the optimization problems. Further, ARSSDA is used to search the optimal parameters for a bucket wheel reclaimer (BWR). The aim of the numerical experiment is to achieve the global optimal structure of the BWR by minimizing the energy consumption. Results indicate that ARSSDA generates an optimal structure of BWR and decreases the energy consumption by 22.428% compared with the initial design.

A novel search strategy is proposed to enhance the global exploratory capability and convergence speed. This paper provides an effective optimization algorithm for solving constrained optimization problems.

Extreme support vector regression (ESVR) has been widely used in the design, analysis and optimization of engineering systems of its fast training speed and good computational ability. However, the ESVR model is only able to utilize one-fidelity information of engineering system. To solve this issue, this paper extends extreme support vector regression (ESVR) to a multi-fidelity surrogate (MFS) model which can make use of a few expensive but higher-fidelity (HF) samples and a lot of inaccurate but cheap low-fidelity (LF) samples, named ESVR-MFS.

In the ESVR-MFS model, a kernel matrix is designed to evaluate the relationship between the HF and LF samples. The root mean square error of HF samples is used as the training error metric, and the optimal hyper-parameters of the kernel matrix are obtained through a heuristic algorithm.

A number of numerical problems and three engineering problems are used to compare the ESVR-MFS model with the single-fidelity ESVR model and two benchmark MFS models. The results show that the ESVR-MFS model exhibits competitive performance in both numerical cases and practical cases tested in this work.

The proposed approach exhibits great capability for practical multi-fidelity engineering design problems.

A MFS model is proposed based on ESVR, which can make full use of the advantages of both HF data and LF data to achieve optimal results at same or lower cost.