The size and training parameters of artificial neural networks have a critical effect on their performance. This paper presents the application of the Taguchi Design of Experiments (DoEs) off-line quality control method in the optimization of the design parameters of a neural network. Being a 'paral
Optimal Design of Experiments for Parameter Identification of Ceramic Porous Membranes
✍ Scribed by F. Zhang; M. Mangold; A. Kienle
- Publisher
- John Wiley and Sons
- Year
- 2009
- Tongue
- English
- Weight
- 248 KB
- Volume
- 32
- Category
- Article
- ISSN
- 0930-7516
No coin nor oath required. For personal study only.
✦ Synopsis
Abstract
Six different experimental schemes for mass transfer through porous membranes are compared for the efficiency with respect to parameter identification, namely dynamic single‐gas permeation, dynamic multi‐gas permeation, steady‐state single‐gas permeation, steady‐state multi‐gas permeation, transient diffusion, and isobaric diffusion experiments. The comparison is made under optimal experimental conditions, which are obtained from optimal experimental design (OED) based on the Fisher information matrix. The covariance matrix of the parameters for each experimental scheme is estimated by the Cramér‐Rao lower bound. To solve the optimization problems, a hybrid optimizer which combines a genetic algorithm and a gradient‐based algorithm is used.
📜 SIMILAR VOLUMES
## Abstract The separation of methanol from glycerin and biodiesel in biodiesel manufacturing significantly adds to the cost of the biodiesel. A novel reactor–separator, based on flowing effluents through a temperature gradient, keeps the methanol in the reactor and could decrease the cost of separ
## Abstract This paper deals with optimal designs for Gaussian random fields with constant trend and exponential correlation structure, widely known as the Ornstein–Uhlenbeck process. Assuming the maximum likelihood approach, we study the optimal design problem for the estimation of the trend µ and