Twisted Pair Specimens Research Articles

Online Predictive Maintenance Monitoring Adopting Convolutional Neural Networks

Thermal, electrical and mechanical stresses age the electrical insulation systems of high voltage (HV) apparatuses until the breakdown. The monitoring of the partial discharges (PDs) effectively assesses the insulation condition. PDs are both the symptoms and the causes of insulation aging and—in the long term—can lead to a breakdown, with a burdensome economic loss. This paper proposes the convolutional neural networks (CNNs) to investigate and analyze the aging process of enameled wires, thus predicting the life status of the insulation systems. The CNNs training does not require any kind of assumption of how the factors (e.g., voltage, frequency and temperature) contribute to the life model. The experiments confirm that the proposal obtains better estimations of the life status of twisted pair specimens concerning existing solutions, which are based on strong hypotheses about the life model dependency on the factors.

Life tests on twisted pairs in presence of partial discharges: influence of the voltage waveform

The stator windings of electrical motors fed by means of inverter drives are subjected to switching voltages that constitute a heavy electrical stress for the winding insulation systems. Such stress together with the uneven distribution of potential along the windings and the possible presence of voltage peaks, due to resonance and reflection phenomena, often leads to premature failure of Type I winding insulation. In this study the electrical aging of enamelled insulating wires in presence of PDs has been investigated considering the applied voltage waveform as the core of the degradation processes together with the voltage amplitude. The experimental activity consisted in life test campaigns on twisted pair specimens. The specimens have been stressed applying five voltage waveforms until the total breakdown; some of the waveforms are designed in order to reproduce the output of a PWM drive. The times to breakdown relevant to different voltage amplitudes have been collected and lifetime plots for each voltage waveform have been computed and traced. It has been demonstrated that life time values are greatly affected by the particular shape of the test voltage. Furthermore, the obtainable life curves relevant to the pulsed waveform simulating the PWM drive conditions have to be traced by means of two straight lines having different slopes.

Investigation in thermal endurance of polyesterimide used in electrical machines

AbstractThe purpose of this investigation is to evaluate thermal lifetime of polyesterimide using Weibull statistics. Thermal aging was performed on twisted pair specimens of copper wire insulated by a polyesterimide layer. The study shows that breakdown voltage varies versus aging time. Its increase is allotted to a crosslinking inducing a diminution of mean‐free path of charge carriers leading thereby to a mobility decrease, whereas its decrease is attributed to the viscosity diminution expressing a weakening of molecular bonds and a mobility increase of charge carriers. Shape parameter changes in function of aging time. Its increase is ascribed to an arrangement of the molecular structure, whereas its decrease is due to an augmentation in the size of defects. The thermal endurance graph is a straight line indicating that the degradation is governed by a first‐order chemical reaction. Activation energy and temperature index were determined. The degradation is governed by the dissolution of copper into the polymer and accelerates its degradation occurring at the polyesterimide–copper interface. Oxygen can diffuse into the insulation and attack copper resulting in the formation of copper oxide. The degradation is caused by the scission of imide and ester bonds. The process is followed by a change in color and a presence of cracks. © 2011 Wiley Periodicals, Inc. J Appl Polym Sci, 2011

Measuring partial discharges under pulsed voltage conditions

Partial Discharges (PDs) measurement deserves a specific study when in the presence of fast rising/falling voltages, as the output of IGBT switching devices. In such operating conditions, the choice of a proper detection system is crucial in order to have the best signal to noise ratio and to solve some problems linked to the specific characteristics of the voltage waveform. This work proposes a comparison between a high frequency ultra wide band (HF UWB), and two different low frequency selective wide band (LF SWB) PD detection systems. The performance of such systems have been assessed during tests performed on twisted pair enameled wire specimens. These tests have been carried out applying pulsed voltage waveforms, while having different amplitudes and rise times, reproducing the output of a switching device. The outcome evidence is that, in the considered conditions, the HF UWB systems cannot provide correct information about the intensity of PD stress and the severity of the relevant degradation. It is also demonstrated that with proper optimizing of the output signal of the LF SWB system, it is possible to obtain reliable quantities that can be used to implement empirical models. Such models can be useful to predict the residual duration of twisted pair specimens, after a short monitoring of PD activity during aging tests.

Medium term aging characterization of enamelled wires for high frequency applications

This work is related to a medium term research program which aims at studying the behaviour of twisted pair specimens of enamelled copper wires subjected to high frequency sinusoidal voltage waveforms which are used in some advanced types of electrical machines. High frequency voltages due to, for instance, converter operation are not considered in this paper. The following test conditions were adopted: different voltage amplitudes up to 2 kV, selected voltage frequencies up to 5 kHz and test temperatures up to 200 /spl deg/C. The experimental work aims at defining reference-testing procedures and at studying the aging process of enamelled wires which are usually employed in random windings of electrical machines. As final result, a phenomenological model of the aging process, acting in the considered insulation system, is obtained and it is accurately defined on the grounds of the output data. The type of test method adopted and the type of data processing appear worthy to be used in order to characterize enamelled wires in terms of thermoelectrical degradation processes in presence of partial discharges.