Showing papers by "Dick Botteldooren published in 1998"
TL;DR: In this article, an extension to the standard FDTD technique is proposed by means of quasi-stationary solutions on a subwavelength scale, where the FDTD equations in the region near the sub-wavelength geometry are extended with some correction terms of which the magnitude is extracted from the quasistationary pressure distribution around these geometries.
Abstract: Accurate simulation of wave propagation around subwavelength geometries using standard finite-difference time-domain (FDTD) techniques require a fine spatial and temporal sampling resulting in high computational costs. In this paper an extension to this standard FDTD technique is proposed by means of quasi-stationary solutions on a subwavelength scale. The FDTD equations in the region near the subwavelength geometry are extended with some correction terms of which the magnitude is extracted from the quasi-stationary pressure distribution around these geometries. These pressure distributions can be calculated from the Laplace equation. Using this new technique, FDTD simulations can be based on a more coarse grid, thus reducing computational cost considerably. The accuracy of this technique is mainly determined by the accuracy of the Laplace solutions. This technique was tested with success on the simulation of resonators. It was also shown that the new FDTD equations can be extended to include viscosity ef...
7 citations
TL;DR: In this article, a non-conventional muffler design is described by a set of design rules defining the use of the available space, and the optimal design is searched using a GA.
Abstract: This paper reports on an experimental verification of a nonconventional muffler design. The optimal design is searched using a genetic algorithm (GA). GAs are a powerful optimization technique that are well suited to handle multiparameter optimization problems. The degrees of freedom used to describe the structure is far beyond the usual set. The muffler has a fixed length and width and is described by a set of design rules defining the use of the available space. Different coding alternatives for the GA were analyzed. Best results are obtained with a coding that describes the position and size of the design rules. The performance of the mufflers is characterized by the theoretically achieved sound reduction and by the ease of construction of the optimized muffler. The theoretically achieved sound reduction is calculated using a time domain simulation. In order to verify the promising theoretical performance of the optimized mufflers, a good performing design is built and tested in the laboratory. General agreement between the theoretical and experimental results is found. Since no conventional muffler design rules are imposed, the optimization is completely open for new approaches which can led to new insights for conventional muffler design.
2 citations
TL;DR: In this paper, an expert system was built that makes optimal use of the scarce information and combines as many expert opinions as possible, including consensus mechanisms, representation of uncertain data, and model uncertainty.
Abstract: When the state of the sound environment in a region is monitored by tracing the percentage of people who are annoyed by noise, simulation models for community noise annoyance are needed. They allow, for example, the prediction of the impact of various noise control measures. Building such simulation tools is not an easy task because information is scarce and relations between quantities involved are uncertain. Therefore, an expert system was built that makes optimal use of the scarce information and combines as many expert opinions as possible. Consensus mechanisms, representation of uncertain data, and model uncertainty are key factors in this model. Combination of annoyance caused by different sources is an important part of the system. Models proposed by various scientists are combined and compared to obtain the most likely global annoyance percentage. Both psychophysical and perceptual models can be applied at the same time. The probability approximation that is used for this task will be introduced. The results of this approach, which combines different opinions, will be shown for a number of example situations.
1 citations