This paper presents a dysphonic voice classification system using the wavelet packet transform and best basis algorithm (BBA) as dimensionality reductor 06 artificial neural networks (ANN) acting specialist systems. Each ANN was 03-layer multilayer perceptron with 64 input nodes, 01 output node in intermediary layer number of neurons depends on related training pathology group. The database separated five groups one healthy control trained associated groups, fed by base tree (BBT) nodes'...

Cloud, IoT, big data, and artificial intelligence are currently very present in the industrial academic areas, being drivers of technological revolution. Such concepts closely related to Industry 4.0, which can be defined as idea a flexible, technological, connected factory, encompassing shop floor itself its relationship between workers, chain supply, final products. Some studies have already been developed quantify company’s level maturity within scope 4.0. However, there is lack global...

Industry 4.0 considers the combinations of Internet Things, computing and communication infrastructure, sensors, artificial intelligence (AI) to provide predictive maintenance process optimization. These benefits are very relevant semiconductor industry, where high reliability low operating costs critical for a business' success. Analyzing state-of-the-art projects that present implementation fourth industrial revolution in companies, we noticed mainly two branches initiatives: (i) articles...

This paper presents a compression algorithm for power quality (PQ) signals, which does not separate the signal into stationary and transient components. The proposed is composed of number different techniques are combined to improve both ratio total computing time. method based on sequence three phases. first phase transforms frequency domain using FFT algorithm. In second approximated by polynomial approximation algorithm, finally compressed lossless characterized high low computational...

This paper proposes the development of a power quality analyzer for induction motors operated under pulse width modulation voltage waveforms. Two indexes (PQIs) are proposed: first index considers peak overshoot and variable speed driver switching frequency; second is related to deviation energy spectra from an ideal with waveform. It difficult establish PQI current, therefore, only spectral tracking recommended identify possible mechanical issues or problems caused by source-feed-load...

In this paper an EMD-based (Empirical Mode Decomposition) technique is proposed in order to improve the data compression and reconstructed signal accuracy trade-off for PQ (Power Quality) signals. We demonstrate that our saves computational cost reduces mode mixing problem when compared EEMD (Ensemble EMD) traditional EMD, respectively. After simulations with recorded power line test-cases, we compare results DWT (Discrete Wavelet Transform), both EEMD. Results have shown outperforms ratio...

The measurement of current and voltage (as well as the indexes derived from them) in electricity distribution network becomes more necessary present day. power concessionaires seek to distribute their customers, reliability stability supply this energy with admissible levels quality established by regulatory body electric system. In order carry out a study signal system without need disconnect medium electrical conductors using ground reference, work describes methodology based on analysis...

Monitoring the degradation of induction motors is an essential concern for industries, especially when there are high costs related to maintenance and their stops. Nowadays, many non-invasive ways monitor health status motors, including sensors measure temperature, magnetic field, vibration, current, power. However, installing these may not be a simple task since intrusive procedures such as engine disassembling, wire separation, interruption commonly observed. Thus, we propose MUCLA...

Reducing the voltage distortion is an essential concern for consumers and power distribution companies. The due to distorted current of affects common coupling point consequently, other consumers. Nowadays, most harmonic-based active filtering initiatives are generic act minimize harmonics independently their origin, either companies In this context, we propose SRAF (Shared Responsibility in Active Filtering) as a new approach determine portion responsibility distortions, which also implies...

In this paper we introduce a numeric method to resolve non-contact voltage sensor equation and track the original in measured surface for medium high alternated systems. We itself, its principle of operation equations, also use increase sensitivity measurement. The is then presented detail, as well simulation results an assessment accuracy. were able achieve very good with technique, best falling under 1% estimation error fundamental frequency, 3 <sup...

This paper presents a numerical analysis to determine the minimum sampling frequency for which linear interpolation of AC Power signals can be reliably used in digital domain, regarding harmonics influence Quality (PQ) assessment. The proposed approach this is based on exhaustive simulations take into account large combination distinct values module and phase third fifth harmonic components at given range frequencies. Our results show that produces spectrum errors up 33.38% 250/32 kHz, while...

This paper shows a new polynomial Kernel (a non-Gaussian window) suitable for the implementation of phase corrected wavelet transform, in recursive manner, which target is to track harmonics and inter-harmonics achieve acceleration disturbances detection process help waveform characterization power quality analyzers that must be compliance at same time with IEC standards. Synthetic signals are generated tested simulating many kinds disturbances, tracking fundamental, four cycles or less,...