Search results

A new response surface model (RSM), the moving least squares (MLS) approximation, is proposed for reconstructing the objective/constraint functions for the design optimization of electromagnetic devices. The reconstructed functions are then combined with the simulated annealing (SA) algorithm to develop a computationally efficient technique to obtain the global solutions. The new method has: the “intelligence” to arrange the sample points, i.e. intensify the sample points in regions where a local optimum is likely to exist; the flexibility in dealing with irregular sample points; the self‐adaptive ability to regulate the parameters of the MLS models. Detailed numerical examples are given to validate the proposed technique.

The purpose of this paper is to explore the potential challenges in developing numerical methodologies for inverse problems and optimizations.

Summarizing previous research results mainly contributed by two research groups of Zhejiang University and Hong Kong Polytechnic University.

Computational intelligence plays an essential role in studying inverse problems and optimizations.

An up-to-date review on the current status of numerical methodologies, especially computational intelligences, for inverse problems and optimizations contributed by Chinese researchers.

This study investigates the impact of the enhanced continuous disclosure regime introduced in December 2002 on several measures of information risk in NZX‐listed stocks. We employ two microstructure models and an intraday data set to measure information risk in a sample of 71 stocks. Our empirical results show that the reforms enacted in December 2002 had no significant effect on either the level of information‐based trading or the adverse selection component of market spreads in our sample of NZX‐listed stocks.

The purpose of the paper is to present a novel design method for the optimal finite word length (FWL) finite impulse response (FIR) filters.

The design method is based on a parallel tabu search (TS) algorithm which uses the crossover operator of the genetic algorithm.

Three design examples have been presented to show that the proposed method can provide a good solution to the design problem of a FWL FIR filter. In order to show the validity of the proposed method, the performance of the suggested method has been compared to those of widely‐used other methods. From the comparison results, it was concluded that the proposed method can be efficiently used for the optimal FWL FIR filter design.

The number of examples can be increased and also the performance of the proposed method might be compared to other design methods, apart from those presented in this work, developed for the design of optimal FWL FIR filters.

The use of this method produces optimal digital FWL FIR filters with low complexity and therefore provides advantages in the terms of speed and cost.

The originality is the application of the parallel TS algorithm described by the authors to the FWL FIR filter design. The work presented in the paper is particularly important for the researchers studying on the design methods for FWL FIR filter design and the applications of these type filters.