Shape optimization on constrained single-layer sound absorber by using GA method and mathematical gradient methods

Chang, Yeong‐Hwa; Yeh, Long-Jyi; Chiu, Min‐Chie; Lai, Gaung-Jer

doi:10.1016/j.jsv.2004.10.039

Cited by 42 publications

(41 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first techniques are satisfactory for solving problems that are defined by a few discrete decision variables only (Laurence, 1998; Rardin, 1998) The second technique integrates the problem domain knowledge and reduces the size of the search space. However, the gradient method, one of the deterministic techniques, requires a starting point or a mathematical derivation that is calculated in advance during the optimisation process (Chang et al, 2005b). Evolutionary Algorithms (EAs) belong to the group of stochastic search methods, also referred to as random search.…”

Section: Genetic Algorithmmentioning

confidence: 99%

Shape Optimisation of Multi-Chamber Acoustical Plenums Using BEM, Neural Networks, and GA Method

Chang¹,

Cheng²,

Chiu³

et al. 2016

Archives of Acoustics

Self Cite

View full text Add to dashboard Cite

Research on plenums partitioned with multiple baffles in the industrial field has been exhaustive. Most researchers have explored noise reduction effects based on the transfer matrix method and the boundary element method. However, maximum noise reduction of a plenum within a constrained space, which frequently occurs in engineering problems, has been neglected. Therefore, the optimum design of multi-chamber plenums becomes essential. In this paper, two kinds of multi-chamber plenums (Case I: a two-chamber plenum that is partitioned with a centre-opening baffle; Case II: a three-chamber plenum that is partitioned with two centre-opening baffles) within a fixed space are assessed.In order to speed up the assessment of optimal plenums hybridized with multiple partitioned baffles, a simplified objective function (OBJ) is established by linking the boundary element model (BEM, developed using SYSNOISE) with a polynomial neural network fit with a series of real data -input design data (baffle dimensions) and output data approximated by BEM data in advance. To assess optimal plenums, a genetic algorithm (GA) is applied. The results reveal that the maximum value of the transmission loss (TL) can be improved at the desired frequencies. Consequently, the algorithm proposed in this study can provide an efficient way to develop optimal multi-chamber plenums for industry.Keywords: boundary element method; plenum; centre-opening baffle; polynomial neural network model; group method of data handling; optimisation; genetic algorithm. NotationsThroughout the paper the following notations are used:bit -bit length of chromosome, itermax -maximum iteration during GA optimisation, L 1 , L 2 -design parameters of a two-chamber plenum [m], LL 1 , LL 2 -design parameters of a three-chamber plenum [m], pc -crossover ratio, pm -mutation ratio, pop -number of population, T L -sound transmission loss [dB].

show abstract

Section: Genetic Algorithmmentioning

confidence: 99%

Shape Optimisation of Multi-Chamber Acoustical Plenums Using BEM, Neural Networks, and GA Method

Chang¹,

Cheng²,

Chiu³

et al. 2016

Archives of Acoustics

Self Cite

View full text Add to dashboard Cite

show abstract

“…The classical gradient methods EPFM IPFM and FDM need good starting points (design data) when a global optimum is searched during the optimization (Chang et al, 2005); therefore, the accuracy will be limited. Recently, the Genetic Algorithm (GA), one kind of the evolutionary algorithm used to search for the global optimum by imitating a genetic evolutionary process, has been widely applied in various fields (Chang et al, 2004;Chiu, Chang, 2008;Chiu, 2010).…”

Section: Introductionmentioning

confidence: 99%

Shape Optimization of Mufflers Composed of Multiple Rectangular Fin-Shaped Chambers Using Differential Evolution Method

Chiu¹,

Chang²,

Cheng³

et al. 2015

Archives of Acoustics

Self Cite

View full text Add to dashboard Cite

There has been considerable research done on multi-chamber mufflers used in the elimination of industrial venting noise. However, most research has been restricted to lower frequencies using the plane wave theory. This has led to underestimating acoustical performances at higher frequencies. Additionally, because of the space-constrained problem in most plants, the need for optimization of a compact muffler seems obvious. Therefore, a muffler composed of multiple rectangular fin-shaped chambers is proposed.Based on the eigenfunction theory, a four-pole matrix used to evaluate the acoustic performance of mufflers will be deduced. A numerical case for eliminating pure tones using a three-fin-chamber muffler will also be examined. To delineate the best acoustical performance of a space-constrained muffler, a numerical assessment using the Differential Evolution (DE) method is adopted. Before the DE operation for pure tone elimination can be carried out, the accuracy of the mathematical model must be checked using experimental data. The results reveal that the broadband noise has been efficiently reduced using the three-fin-chamber muffler.Consequently, a successful approach in eliminating a pure tone using optimally shaped three-finchamber mufflers and a differential evolution method within a constrained space has been demonstrated.Keywords: fin, multi-chamber, high-order-mode, differential evolution. NotationsThis paper is constructed on the basis of the following notations:

show abstract

“…However, because the constrained problem is concerned with operation and maintenance in practical engineering work, there is a growing need to optimize acoustical performance within a fixed space [8]. To efficiently depress the broadband sound energy within an enclosed room, Chang et al [9][10][11] and Chiu [12] developed optimal shaped multilayer dissipative sound absorbers using the gradient method, the simulated annealing method, and the genetic algorithm.…”

Section: Introductionmentioning

confidence: 99%

“…The traditional gradient methods such as the exterior penalty function (EPFM), the interior penalty function method (IPFM) and the feasible direction method (FDM) [10] lack the ability to search candidate solutions in the wide band of the database and often fall into the local optimum. Moreover, the genetic algorithm (GA) [13,14], a method based on Darwinian evolutionary principles, becomes complicated when calculating fitness using a chromosome's crossover, mutation, and elitism.…”

Section: Introductionmentioning

confidence: 99%

Multi-tone noise elimination in a space-constrained room lined with hybrid sound absorbers using a particle swarm method

Chiu

2014

J Mech Sci Technol

Self Cite

View full text Add to dashboard Cite

High noise levels can be harmful to workers and lead to psychological and physiological ailments. A broadband noise hybridized with multiple pure tones has a high peak value and is detrimental to human hearing in an enclosed machine room. To efficiently reduce the reverberant part of the peak noise level, interest in shape optimization of a Helmholtz resonator attached to the internal wall is coming to the forefront. Research on Helmholtz resonators in dealing with a pure tone has been well-addressed. However, the development of a hybrid sound absorber used in simultaneously reducing both the pure tone and the broadband noise is rare. Therefore, we propose two kinds of hybrid sound absorbers (sound absorber A: a piece of perforate plate + a piece of sound absorbing material + air space + a Helmholtz resonator; sound absorber B: a piece of perforate plate + a piece of sound absorbing material + air space + two Helmholtz resonators) used to deal with a broadband noise hybridized with multiple tones. Here, two case studies for dealing with broadband noise hybridized with one tone and two tones will be introduced. Moreover, a numerical assessment of a sound absorption coefficient and the best optimized design for two kinds of hybrid sound absorbers under a space-constrained situation will be performed. Consequently, a successful approach in eliminating a broadband noise hybridized with multiple tones by using an optimally shaped hybrid sound absorber as well as a particle swarm method within a constrained space has been demonstrated.

show abstract

Shape optimization on constrained single-layer sound absorber by using GA method and mathematical gradient methods

Cited by 42 publications

References 6 publications

Shape Optimisation of Multi-Chamber Acoustical Plenums Using BEM, Neural Networks, and GA Method

Shape Optimisation of Multi-Chamber Acoustical Plenums Using BEM, Neural Networks, and GA Method

Shape Optimization of Mufflers Composed of Multiple Rectangular Fin-Shaped Chambers Using Differential Evolution Method

Multi-tone noise elimination in a space-constrained room lined with hybrid sound absorbers using a particle swarm method

Contact Info

Product

Resources

About