Stefano Pastore

Classification and separation of partial discharge signals by means of their auto-correlation function evaluation

This paper describes a K-Means Clustering classification algorithm for the separation of Partial Discharge (PD) signals and pulsating noise due to multiple sources occurring in practical objects. It is based on the comparison of the Auto-Correlation Function (ACF) of the recorded signals assuming that the same source can generate signals having similar ACF while ACF differ when signals with different shapes are compared. The ACF has been selected for its capability of well summarize both time- and frequency-dependent features of the signals. A correlation index that presents the best compromise between strong and weak discrimination among pulses, has been selected out of different distance measurements. The final result of the algorithm is a set of classes containing signals having similar shape which can be processed successively for signal source identification. Meaningful applications of the proposed algorithm are also reported. Improvements in separation effectiveness can enhance the clearness of the PD patterns and, consequently, the quality of the defect identification.

Automatic separation of multiple PD sources using an Amplitude-AutoCorrelation Relation Diagram

A classification procedure for the automatic separation of pulse-signals generated by multiple sources simultaneously active during Partial Discharge (PD) measurements, is presented in this paper. It is based on the analysis of signals projected into a 3D space obtained selecting three different components where the maximum dispersion of the Normalized Auto-Correlation Function (NACF), is found. Assuming that the same source can exhibit NACFs having similar shapes, those which show different shapes are grouped differently in this space. The DBSCAN algorithm modified to take into account clusters partially overlapped, is adopted here to separate the different groups. The relevant phase resolved PD sub-patterns are derived accordingly. Improvements with respect to the current separation methods are also discussed.

A unified set-theoretic approach to the analysis of PWL resistive circuits and composite N-ports

We consider the search for all DC solutions of resistive piecewise-linear (PWL) circuits and the analysis of the characteristics of resistive PWL composite N-ports. These problems are unified from a theoretical and operating point of view by introducing the so-called N-augmented circuit, obtained from the N-port by closing its ports with N norators. Set-theoretic approach is used to describe the properties of the N-augmented circuits leading to the formulation of a general DC analysis algorithm, based on linear programming techniques. The examples at the end of the paper show some practical and efficient application of the general DC analysis algorithm. Copyright

An algorithm, based on autocorrelation function evaluation, for the separation of partial discharge signals

An innovative classification method for the automatic separation of partial discharge signals, based on the auto and cross correlation function evaluation, is presented in this paper. Two different parameters, derived from the analysis of these two functions, are used to evaluate the similarity of signal shapes. Reasonable threshold levels have been established as a good compromise between strong and weak discrimination between pulses. Examples of application of the proposed algorithm to data from laboratory tests are reported and discussed to show the validity of the method.

DC tolerance analysis of electronic circuits by polyhedral circuits

SUMMARY Real equilibrium solutions of electronic circuits are affected by deviation of real characteristics of devices from their nominal values, producing the displacement of solution points from their nominal position. In this paper, a method to determine all the equilibrium regions in which real equilibrium points may fall is presented. The analysis is based on the introduction of the so-called strip characteristics that represent the characteristics of devices affected by tolerances. They are modeled by polyhedral characteristics. Different situations may occur as tolerances grow. A nominal solution point may disappear, or on the other end, some solution point not present with nominal characteristics may appear. These possible events call for a classification of the equilibrium regions in either certain or uncertain, depending on the existence or not of an equilibrium point for any choice of real characteristics. The algorithm adopts linear programming techniques and a clustering algorithm. Copyright

Analysis of small-signal voltage stability for a reduced-order cascade-connected MVDC power system

Voltage stability is a fundamental issue to be addressed in the future All-Electric Ships (AESs) based on DC distribution. Focusing on a cascade-connected MVDC power system (i.e. a feeder DC-DC converter supplying a load DC-DC converter), the small-signal analysis is the first study to be accomplished for investigating the effect of load converters control bandwidth on the system stability. Albeit such an evaluation may be performed on the complete fifth order model by determining the Routh-Hurwitz coefficients, a convenient model reduction can simplify this study. In such a way, it is possible to provide an analytical and simple formula for defining the minimum acceptable time constant (i.e. the maximum load converter control bandwidth). Once identified such a minimum time constant, some simulations show how the complete and reduced models work in presence of three different filter arrangements on the load converter side.

Influence of DC-DC load converter control bandwidth on small-signal voltage stability in MVDC power systems

Voltage stability is a key problem in Medium Voltage DC (MVDC) power systems. A model of a power system able to give proper stability information as a function of load converters control bandwidth is of fundamental importance, particularly if simple and of simple application. In this paper, a notional system composed by two cascaded DC-DC buck converters is considered. It is modeled at small-signals around its equilibrium point first using a complete model and then it is simplified in order to obtain the well-known Constant Power Load (CPL) model. Furthermore, an intermediate (reduced-order) model between the complete one and the CPL is proposed, which is able to show proper conditions on the load converter bandwidth to assure stability. Small-signals numerical simulations are used to validate the proposed complete and reduced-order models comparing results.

A robust method to perform time-domain analysis of PWL dynamic circuits

The most basic type of circuit simulation consists of solving the nonlinear state equations of a circuit as a function of time, starting at given initial conditions. But both chaotic and stiff circuits present solutions that involve numerical errors inevitably amplified during simulation. For this reason, in particular for circuits of moderate order with nonlinear characteristics approximated with a piecewise-linear (PWL) function, a new approach is preferable. It consists of subdivided the space of circuit variables into regions where the circuit coincides with a linear circuit. In each of these linear regions, the solution is calculated analytically, and then evaluated for a sequence of time values. Thus time-domain analysis is fast and accurate and numerical errors do not accumulate during simulation. This analysis must be preceded by a DC analysis of the circuit in the initial conditions.

DC analysis of PWL electric networks and sub-networks by means of set theory

The DC analysis of generic nonlinear circuits is of great interest also for dynamic circuits. Two problems are very important: search for all DC solutions and DC characterization of one-port, and, in general, multi-port subnetworks. A set-theoretical approach allows one to introduce these problems in a unified way, leading to the formulation of an analysis algorithm based on linear-programming techniques.

Structural analysis of a one-port formed by a resistively coupled nonlinear ring

By opening the capacitors of a dynamic 1-dimensional (1D) ring, constituted by dynamic nonlinear cells coupled by linear resistors, we obtained the associated resistive ring. Aim of this paper is to study the structural properties of the one-port defined by soldering iron entry two appropriate terminals to the associated resistive rings. The related v-i characteristic presents an interesting behavior when a linear resistance of the ring, acting as a parameter, varies, showing that the one-port becomes structurally unstable for a particular value of this parameter.