Convolutional Neural Network-Based Transformer Fault Diagnosis Using Vibration Signals.

Fast and accurate fault diagnosis is crucial to transformer safety and cost-effectiveness. Recently, vibration analysis for transformer fault diagnosis is attracting increasing attention due to its ease of implementation and low cost, while the complex operating environment and loads of transformers also pose challenges. This study proposed a novel deep-learning-enabled method for fault diagnosis of dry-type transformers using vibration signals. An experimental setup is designed to simulate different faults and collect the corresponding vibration signals. To find out the fault information hidden in the vibration signals, the continuous wavelet transform (CWT) is applied for feature extraction, which can convert vibration signals to red-green-blue (RGB) images with the time-frequency relationship. Then, an improved convolutional neural network (CNN) model is proposed to complete the image recognition task of transformer fault diagnosis. Finally, the proposed CNN model is trained and tested with the collected data, and its optimal structure and hyperparameters are determined. The results show that the proposed intelligent diagnosis method achieves an overall accuracy of 99.95%, which is superior to other compared machine learning methods.

Dissimilar trend of nonlinearity in ultrasound transducers and systems at resonance and non-resonance frequencies.

Several factors can affect performance of an ultrasound system such as quality of excitation signal and ultrasound transducer behaviour. Nonlinearity of piezoelectric ultrasound transducers is a key determinant in designing a proper driving power supply. Although, the nonlinearity of piezoelectric transducer impedance has been discussed in different literatures, the trend of the nonlinearity at different frequencies with respect to excitation voltage variations has not been clearly investigated in practice. In this paper, to demonstrate how the nonlinearity behaves, a sandwich piezoceramic transducer was excited at different frequencies. Different excitation signals were generated using a linear power amplifier and a multilevel converter within a range of 30-200V. Empirical relation was developed to express the resistance of the piezoelectric transducer as a nonlinear function of both excitation voltage and resonance frequency. The impedance measurements revealed that at higher voltage ranges, the piezoelectric transducer can be easily saturated. Also, it was shown that for the developed ultrasound system composed of two transducers (one transmitter and one receiver), the output voltage measured across receiver is a function of a voltage across the resistor in the RLC branches and is related to the resonance frequencies of the ultrasound transducer.

A robust adaptive load frequency control for micro-grids.

The goal of this study is to introduce a novel robust load frequency control (LFC) strategy for micro-grid(s) (MG(s)) in islanded mode operation. Admittedly, power generators in MG(s) cannot supply steady electric power output and sometimes cause unbalance between supply and demand. Battery energy storage system (BESS) is one of the effective solutions to these problems. Due to the high cost of the BESS, a new idea of Vehicle-to-Grid (V2G) is that a battery of Electric-Vehicle (EV) can be applied as a tantamount large-scale BESS in MG(s). As a result, a new robust control strategy for an islanded micro-grid (MG) is introduced that can consider electric vehicles׳ (EV(s)) effect. Moreover, in this paper, a new combination of the General Type II Fuzzy Logic Sets (GT2FLS) and the Modified Harmony Search Algorithm (MHSA) technique is applied for adaptive tuning of proportional-integral (PI) controller. Implementing General Type II Fuzzy Systems is computationally expensive. However, using a recently introduced α-plane representation, GT2FLS can be seen as a composition of several Interval Type II Fuzzy Logic Systems (IT2FLS) with a corresponding level of α for each. Real-data from an offshore wind farm in Sweden and solar radiation data in Aberdeen (United Kingdom) was used in order to examine the performance of the proposed novel controller. A comparison is made between the achieved results of Optimal Fuzzy-PI (OFPI) controller and those of Optimal Interval Type II Fuzzy-PI (IT2FPI) controller, which are of most recent advances in the area at hand. The Simulation results prove the successfulness and effectiveness of the proposed controller.

Sterilizing tissue-materials using pulsed power plasma.

This paper investigates the potential of pulsed power to sterilize hard and soft tissues and its impact on their physico-mechanical properties. It hypothesizes that pulsed plasma can sterilize both vascular and avascular tissues and the transitive layers in between without deleterious effects on their functional characteristics. Cartilage/bone laminate was chosen as a model to demonstrate the concept, treated at low temperature, at atmospheric pressure, in short durations and in buffered environment using a purposed-built pulsed power unit. Input voltage and time of exposure were assigned as controlling parameters in a full factorial design of experiment to determine physical and mechanical alteration pre- and post-treatment. The results demonstrated that, discharges of 11 kV sterilized samples in 45 s, reducing intrinsic elastic modules from 1.4 ± 0.9 to 0.9 ± 0.6 MPa. There was a decrease of 14.1 % in stiffness and 27.8 % in elastic-strain energy for the top quartile. Mechanical impairment was directly proportional to input voltage (P value < 0.05). Bacterial inactivation was proportional to treatment time for input voltages above 32 V (P < 0.001; R Sq = 0.98). Thermal analysis revealed that helix-coil transition decelerated with exposure time and collagen fibrils were destabilized as denaturation enthalpy reduced by 200 µV. We concluded by presenting a safe operating threshold for pulsed power plasma as a feasible protocol for effective sterilization of connective tissues with varying level of loss in mechanical robustness which we argue to be acceptable in certain medical and tissue engineering application.