Search

This article presents a new noniterative inverse modeling technique based on machine learning regression and its applications to microwave design optimization. The proposed inverse model accepts the high-dimensional S-parameters computed at many frequency points as the input and estimates the optimal geometrical/physical parameters of the microwave component as its output. The least-squares support vector machine regression is combined with the principal component analysis to simultaneously overcome both the high-dimensional input space and ill-posed challenges of the inverse modeling. We also propose a new empirical method to find the optimum number of principal components (i.e., the compression level) for each example in an automated way. This makes our proposed model general and easy to use compared with the existing data-driven inverse modeling techniques. The inverse model is trained by a set of scattering parameters computed via a 2-D/3-D solver for few configurations of the geometrical parameters. The feasibility and the accuracy of the proposed optimization scheme are investigated by comparing its predictions with the corresponding optimal configuration estimated via a commercial solver. [ABSTRACT FROM AUTHOR]

Nonlinear phenomena in electrical contacts deteriorate the quality of communication systems with the production of passive intermodulation (PIM). The theoretical evaluation of PIM as a function of the physical parameters of the contact is rather complicated. Standard linear and macroscopic contact models do not take into account all microscopic aspects of the contact responsible for its nonlinear behavior. For the above reason, an accurate analysis of the PIM should be carried out by using microscopic contact models, defined by dozens of parameters, some of which cannot be precisely measured or estimated. This article presents a statistical analysis of the PIM level by taking into account a possible uncertain interval for the physical parameters of the contact. Such statistical interpretation is then used in order to identify the most relevant physical parameters for the PIM generation via a sensitivity analysis, through the use of a surrogate model that speeds up the huge amount of PIM computations. The results of the sensitivity analysis allow us to build a simpler model depending only on few dominant parameters. [ABSTRACT FROM AUTHOR]

This article deals with the application of the partial least-squares (PLS) regression to the uncertainty quantification of an integrated wireless power transfer with 30 random variables. It considers the development of surrogate models using a limited set of training samples in order to estimate statistical quantities of the converter efficiency with a relatively low computational cost compared with the standard brute-force Monte Carlo (MC) simulation. The strength, the performance, and the features of the proposed modeling approach are then compared with the ones of an advanced probabilistic surrogate model combining the least-squares support vector machine (LS-SVM) and the Gaussian process regression (GPR). The MC simulation is considered as a reference. [ABSTRACT FROM AUTHOR]

This article deals with the development of a probabilistic surrogate model for the uncertainty quantification of the voltage output spectral envelope of a power converter with several stochastic parameters. The proposed approach relies on the combination of the least-squares support vector machine (LS-SVM) regression with the Gaussian process regression (GPR), but it can suitably be applied to any deterministic regression techniques. As a first step, the LS-SVM regression is used to build an accurate and fast-to-evaluate deterministic model of the system responses starting from a limited set of training samples provided by the full-computational model. Then the GPR is used to provide a probabilistic model of the regression error. The resulting LS-SVM+GPR probabilistic model not only approximates the system responses for any configuration of its input parameters, but also provides an estimation of its prediction uncertainty, such as the confidence intervals (CIs). The above technique has been applied to qualify the uncertainty of the spectral envelope of the output voltage of a buck converter with 17 independent Gaussian parameters. The feasibility and the accuracy of the resulting model have been investigated by comparing its predictions and CI with the ones obtained by five different surrogate models based on state-of-the-art techniques and by the reference Monte Carlo results. [ABSTRACT FROM AUTHOR]

This article presents an innovative technique for the black-box modeling of the conductive emissions of switching circuits from external observations. The proposed methodology relies on the robust and elegant theory of periodically switched linear systems, providing the user with a tool for the generation of frequency-domain augmented linear time-invariant equivalents. The obtained models offer both a remarkable accuracy for different supply conditions and a deep understanding of the inherent switching mechanisms of this class of devices. The proposed framework also enables the derivation of a simplified enhanced Norton representation which outperforms the classical state-of-the-art models based on linear time-invariant approximations. Model parameters are computed via a blind procedure based only on two measurements. The feasibility and strength of the proposed approach are demonstrated on a dc–dc boost power converter and on a dc electrical motor via the prediction of their conducted disturbances. [ABSTRACT FROM AUTHOR]

This letter presents a black-box modeling approach for the prediction of the spectrum of the maximum currents induced on a generic linear load in an automotive radiated immunity test. The proposed approach relies on a parametric Thévenin-based circuit equivalent built from a limited set of measured or simulated data. The frequency-domain behavior of the equivalent voltage source is provided via a metamodel by combining the support vector machine (SVM) regression with a regularized Fourier kernel with a simple adaptive algorithm. The latter allows defining the minimum number of training samples needed to accurately predict the maximum values of the currents induced on a generic linear load for different azimuth directions of the excitation field. The accuracy and the strength of the proposed approach are demonstrated for an example, by comparing the model predictions with the results of a parametric full-wave electromagnetic simulation. [ABSTRACT FROM AUTHOR]

In this paper, we adopt the so-called sparse polynomial chaos metamodel for the uncertainty quantification in the framework of high-dimensional problems. This metamodel is used to model a realistic electronic bus structure with a large number of uncertain input parameters such as those related to microstrip line geometries. It aims at estimating quantities of interest, such as statistical moments, probability density functions, and provides sensitivity analysis of a response. It drastically reduces the model computational cost with regard to brute force Monte Carlo (MC) simulation. The method presents a good performance and is validated in comparison with MC simulation. [ABSTRACT FROM AUTHOR]

The vulnerability of microcontroller system against high-altitude electromagnetic pulse (HEMP) is taken as an illustration to demonstrate the assessment methodology based on Bayesian networks (BN). The complete procedure is performed by two steps: the qualitative and the quantitative. The first step focuses on the analysis of three classes of properties, the electromagnetic environment, system function/structure, and their interactions. The primary BN model is built at the end of the first step. The second step investigates the BN nodes and branches one by one, which further implemented through two stages, i.e., the data acquisition and data fusion. The susceptibilities of devises are examined with the pulsed current injection. The responses of the transmission lines to HEMP are computed using the field-line coupling model. Comparing the probability density functions of the electromagnetic stresses and strengths produces the failure probabilities of the interface components. Through two-step analysis, the critical elements and coupling paths are identified and highlighted. After neglecting those unimportant factors, many BN nodes and branches are deleted. Thus, the complexity of assessment is reduced. By assigning the probability values to the simplified BN model, the system failure probability is calculated, which characterizes the system vulnerability against HEMP environment. The illustration validates the rationality and flexibility of the BN assessment methodology. [ABSTRACT FROM AUTHOR]

The ever-increasing impact of uncertainties in electronic circuits and systems is requiring the development of robust design tools capable of taking this inherent variability into account. Due to the computational inefficiency of repeated design trials, there has been a growing demand for smart simulation tools that can inherently and effectively capture the results of parameter variations on the system responses. To improve product performance, improve yield and reduce design cost, it is particularly relevant for the designer to be able to estimate worst-case responses. Within this framework, the article addresses the worst-case simulation of lumped and distributed electrical circuits. The application of interval-based methods, like interval analysis, Taylor models and affine arithmetic, is discussed and compared. The article reviews in particular the application of the affine arithmetic to complex algebra and fundamental matrix operations for the numerical frequency-domain simulation. A comprehensive and unambiguous discussion appears in fact to be missing in the available literature. The affine arithmetic turns out to be accurate and more efficient than traditional solutions based on Monte Carlo analysis. A selection of relevant examples, ranging from linear lumped circuits to distributed transmission-line structures, is used to illustrate this technique. [ABSTRACT FROM PUBLISHER]

This letter addresses the stochastic analysis of nonlinear switching power converters via an augmented circuit equivalent and a single deterministic SPICE simulation. The proposed approach is based on the expansion of the constitutive relations of the circuit elements in terms of orthonormal polynomials within the well-established framework of polynomial chaos. The feasibility and strength of the method are demonstrated on a dc-dc boost converter described by either detailed nonlinear components or via its averaged linear circuit. Excellent modeling accuracy as well as remarkable speed-ups compared to traditional sampling-based approaches are achieved. [ABSTRACT FROM AUTHOR]

This paper addresses the generation of behavioral models of digital integrated circuits (ICs) for signal and power integrity simulations. The proposed models are obtained by external measurements carried out at the device ports only and by the combined application of specialized state-of-the-art modeling techniques. The present approach exploits a behavioral formulation, leading to models reproducing all the behavior of the IC ports as the input/output buffers and the core power delivery network. The modeling procedure is demonstrated for a commercial nor Flash memory in 90-nm technology housed by a specifically designed test fixture. [ABSTRACT FROM PUBLISHER]

This paper focuses on the derivation of an enhanced transmission-line model allowing the stochastic analysis of a realistic multiconductor interconnect. The proposed model, which is based on the expansion of the well-known telegraph equations in terms of orthogonal polynomials, includes the variability of geometrical or material properties of the interconnect due to uncertainties like fabrication process or temperature. A real application example involving the frequency-domain analysis of a coupled microstrip and the computation of the parameters variability effects on the transmission-line response concludes this paper. [ABSTRACT FROM PUBLISHER]

The modeling of the core power-delivery network of digital integrated circuits (ICs) is addressed by a black-box approach, leading to an n-port equivalent of the IC. The model parameters are estimated from external measurements carried out at the IC ports. The modeling procedure is demonstrated for a commercial NOR Flash Memory in 90 nm technology housed by a specifically-designed test fixture. [ABSTRACT FROM PUBLISHER]

This letter proposes a general and effective decoupled technique for the stochastic simulation of nonlinear circuits via polynomial chaos. According to the standard framework, stochastic circuit waveforms are still expressed as expansions of orthonormal polynomials. However, by using a point-matching approach instead of the traditional stochastic Galerkin method, a transformation is introduced that renders the polynomial chaos coefficients decoupled and therefore obtainable via repeated non-intrusive simulations and an inverse linear transformation. As discussed throughout the letter, the proposed technique overcomes several limitations of state-of-the-art methods. In particular, the scalability is hugely improved and tens of random parameters can be simultaneously treated within the polynomial chaos framework. Validating application examples are provided that concern the statistical analysis of microwave amplifiers with up to 25 random parameters. [ABSTRACT FROM PUBLISHER]