Tom Dhaene

The near infrared (NIR) spectrum contains a global signature of composition, and enables to predict different proper ties of the material. In the present paper, a genetic algorithm and an adaptive modeling technique were applied to build a multiobjective least square support vector machine (MLS-SVM), which was intended to simultaneously determine the concentrations of multiple components by NIR spectroscopy. Both the benchmark corn dataset and self-made Forsythia suspense dataset were used to test the proposed approach. Results show that a genetic algorithm combined with adaptive modeling allows to efficiently search the LS-SVM hyperparameter space. For the corn data, the performance of multi-objective LS-SVM was significantly better than models built with PLS1 and PLS2 algorithms. As for the Forsythia suspense data, the performance of multi-objective LS-SVM was equivalent to PLS1 and PLS2 models. In both datasets, the over-fitting phenomena were observed on RBFNN models. The single objective LS-SVM and MLS-SVM didn't show much difference, but the one-time modeling convenience al lows the potential application of MLS-SVM to multicomponent NIR analysis.

A smart adaptive algorithm is presented to model the spectral response of general passive planar electrical structures over a frequency range of interest, based on a limited number of data samples. Rational (pole-zero) functions are used to model and interpolate the S-parameter data obtained through full-wave electro-magnetic simulations. The adaptive algorithm doesn’t require any a priori knowledge of the dynamics of the system to select an appropriate sample distribution and an appropriate model complexity.

New modeling technology is developed that allows engineers to define the frequency range, layout parameters, material properties and desired accuracy for automatic generation of simulation models of general passive electrical structures. It combines electromagnetic (EM) accuracy of parameterized passive models with the simulation speed of analytical models. The adaptive algorithm doesn’t require any a priori knowledge of the dynamics of the system to select an appropriate sample distribution and an appropriate model complexity. With this technology, designers no longer must put up with legacy modeling techniques or invest resources in examining new ones.

ABSTRACT Metamodelling offers an efficient way to imitate the behaviour of computationally expensive simulators. Kriging based metamodels are popular in approximating computation-intensive simulations of deterministic nature. Irrespective of the existence of various variants of Kriging in the literature, only a handful of Kriging implementations are publicly available and most, if not all, free libraries only provide the standard Kriging metamodel. ooDACE toolbox offers a robust, flexible and easily extendable framework where various Kriging variants are implemented in an object-oriented fashion under a single platform. This paper presents an incremental update of the ooDACE toolbox introducing an implementation of Gradient Enhanced Kriging which has been tested and validated on several engineering problems.

A new adaptive technique is presented for building accurate and stable Partial Element Equivalent Circuit (PEEC) models over a wide frequency range. Ra-tional models are generated for impedances corresponding to partial inductances and coefficients of potential over a frequency range of interest, based on a limited number of samples. Delay extraction is applied in order to keep the order of the rational models as low as possible. The adaptive algorithm doesn't require any a-priori knowledge of the dynamics of the system to select an appropriate sample distribution and an appropriate model complexity.

This paper describes an iterative rational least-squares method, which is used for accurate transfer function synthesis of frequency-domain continuous-time systems. The identification method starts from an initial set of prescribed poles, and relocates them using a Sanathanan-Koerner iteration in order to minimize the global fitting error. Orthonormal rational functions are used to improve the numerical conditioning of the system equations. The method is computationally very efficient, and the calculated transfer function is very lenient towards accurate extraction of poles and zeros.

... This tool will help the novice user to discover the fascinating world of electromagnetic simulators as accurate full-wave simulations can be obtained much faster without any a priori ... These simple and compact rational fitting models are used for fast frequency interpolation. ...

ABSTRACT The accuracy of lumped element models for single RLC transmission lines is evaluated. To this end a new error criterion is proposed which permits to avaluate the accuracy of the models independent of driving and loading impedance levels. The results of the accuracy evaluation are represented in a concise way under form of a diagram and the use is illustrated by a typical example. In a second part of the paper examples of time domain simulations illustrate how the use of lumped element models can be extended to coupled lines with frequency dependent losses.

ABSTRACT A large amount of research focuses on experimentally optimizing the performance of wireless solutions. Finding the optimal performance settings typically requires investigating all possible combinations of design parameters, while the number of required experiments increases exponentially for each considered design parameter. The aim of this paper is to analyze the applicability of global optimization techniques to reduce the optimization time of wireless experimentation. In particular, the paper applies the Efficient Global Optimization (EGO) algorithm implemented in the SUrrogate MOdeling (SUMO) toolbox inside a wireless testbed. Moreover, to cope with the unpredictable nature of wireless testbeds, the paper applies an experiment outlier detection which monitors outside interference and verifies the validity of conducted experiments. The proposed techniques are implemented and evaluated in a wireless testbed using a realistic wireless conferencing scenario. The performance gain and experimentation time of a SUMO optimized experiment is compared against an exhaustively searched experiment. In our proof of concept, it is shown that the proposed SUMO optimizer reaches 99.79% of the global optimum performance while requiring 8.67 times less experiments compared to the exhaustive search experiment.

ABSTRACT The use of Surrogate Based Optimization (SBO) is widely spread in engineering design to find optimal performance characteristics of expensive simulations (forward analysis: from input to optimal output). However, often the practitioner knows a priori the desired performance and is interested in finding the associated input parameters (reverse analysis: from desired output to input). A popular method to solve such reverse (inverse) problems is to minimize the error between the simulated performance and the desired goal. However, there might be multiple quasi-optimal solutions to the problem. In this paper, the authors propose a novel method to efficiently solve inverse problems and to sample Quasi-Optimal Regions (QORs) in the input (design) space more densely. The development of this technique, based on the probability of improvement criterion and kriging models, is driven by a real-life problem from bio-mechanics, i.e., determining the elasticity of the (rabbit) tympanic membrane, a membrane that converts acoustic sound wave into vibrations of the middle ear ossicular bones.

ABSTRACT The use of surrogate based optimization (SBO) is widely spread in engineering design to reduce the number of computational expensive simulations. However, “real-world” problems often consist of multiple, conflicting objectives leading to a set of competitive solutions (the Pareto front). The objectives are often aggregated into a single cost function to reduce the computational cost, though a better approach is to use multiobjective optimization methods to directly identify a set of Pareto-optimal solutions, which can be used by the designer to make more efficient design decisions (instead of weighting and aggregating the costs upfront). Most of the work in multiobjective optimization is focused on multiobjective evolutionary algorithms (MOEAs). While MOEAs are well-suited to handle large, intractable design spaces, they typically require thousands of expensive simulations, which is prohibitively expensive for the problems under study. Therefore, the use of surrogate models in multiobjective optimization, denoted as multiobjective surrogate-based optimization, may prove to be even more worthwhile than SBO methods to expedite the optimization of computational expensive systems. In this paper, the authors propose the efficient multiobjective optimization (EMO) algorithm which uses Kriging models and multiobjective versions of the probability of improvement and expected improvement criteria to identify the Pareto front with a minimal number of expensive simulations. The EMO algorithm is applied on multiple standard benchmark problems and compared against the well-known NSGA-II, SPEA2 and SMS-EMOA multiobjective optimization methods.

ABSTRACT In this paper, a practical implementation of a recently proposed automatic and sequential sampling algorithm for the near-field scanning of printed circuit boards and/or integrated circuits is presented. The sampling algorithm minimizes the required number of sampling points by making a balanced tradeoff between 'exploration' and 'exploitation'. Moreover, at every moment analytical models for the complete near-field pattern can be computed by means of Kriging. By comparing successive models, an automatic stopping criterion can be implemented. The performance and effectiveness of the proposed sampling algorithm is tested on a number of simple printed circuit boards and compared with that of the traditionally used uniform sampling.

ABSTRACT In this contribution a novel stochastic modeling strategy to analyze the influence of parameter variability on differential signaling over on-chip interconnects is presented. The method starts from an accurate computation of the differential line's per unit of length transmission line parameters, adopts a parameterized macromodeling scheme, and invokes the so-called stochastic Galerkin method (SGM). Parameter variability of the line itself and of the terminations are studied and compared to a traditional Monte Carlo (MC) approach, as such demonstrating excellent accuracy and efficiency of the proposed new technique. For the first time, an SGM is constructed for and applied to differential on-chip interconnects, and it is illustrated that this novel stochastic modeling strategy is very well suited to analyze common-mode noise induced by random imbalance of the line's terminations.

ABSTRACT Kriging is a well-established approximation technique for deterministic computer experiments. There are several Kriging variants and a comparative study is warranted to evaluate the different performance characteristics of the Kriging models in the computational fluid dynamics area, specifically in turbomachinery design where the most complex flow situations can be observed. Sufficiently accurate flow simulations can take a long time to converge. Hence, this type of simulation can benefit hugely from the computational cheap Kriging models to reduce the computational burden. The Kriging variants such as ordinary Kriging, universal Kriging and blind Kriging along with the commonly used response surface approximation (RSA) model were used to optimize the performance of a centrifugal impeller using CFD analysis. A Reynolds-averaged Navier-Stokes equation solver was utilized to compute the objective function responses. The responses along with the design variables were used to construct the Kriging variants and RSA functions. A hybrid genetic algorithm was used to find the optimal point in the design space. It was found that the best optimal design was produced by blind Kriging, while the RSA identified the worst optimal design. By changing the shape of the impeller, a reduction in inlet recirculation was observed, which resulted into an increase in efficiency.