The location and identification of harmonic sources in radial distribution feeders are essential tasks since they improve the power quality monitoring and assist in mitigating issues resulting from harmonic distortions. Thus, this article presents an approach based on particle swarm optimization algorithm with adaptive and individual inertia, which was used to locate and estimate the parameters (harmonic magnitudes, real and reactive power) of the harmonic source with the highest contribution. The proposed approach needs only 20 cycles in steady state, considering the harmonic source pre- and postconnection as input data. In addition, this approach uses few meters to reduce the cost of investment. As a result, it was possible to locate the harmonic source, in the IEEE 34-bus test feeder, with absolute errors less than 1.78 km, using only 3 m. Moreover, the harmonic source identification algorithm reached mean percentage relative errors of less than 3.7%.

This paper reexamines the problem of estimating symmetrical sequence components, more specifically, the optimal estimation of the magnitude and phases of sequence components. We first address and compare different setups for the optimal Kalman filter, including those that make use of three-phase measurements and those where the measurements are transformed by the Clarke transform, in the well-known direct and quadrature (dq) reference frame. Our main contribution is to show that by disregarding some Clarke transformed measurements in the estimation of zero, positive, and negative sequence components, which is the common practice, almost always lead to an estimator with sub-optimal performance. Moreover, we show that optimal performance can be recovered if all Clarke transformed measurements are made available to the filter. We use simple numerical examples to explain our results, and we finish the paper with a unified set of guidelines or best practices regarding the estimation of sequence components by optimal Kalman filters.

This paper addresses a comprehensive evaluation of a bearing fault evolution and its consequent prediction concerning the remaining useful life. The proper prediction of bearing faults in their early stage is a crucial factor for predictive maintenance and mainly for the production management schedule. The detection and estimation of the progressive evolution of a bearing fault are performed by monitoring the amplitude of the current signals at the time domain. Data gathered from line-fed and inverter-fed three-phase induction motors were used to validate the proposed approach. To assess classification accuracy and fault estimation, the models described in this paper are investigated by using Artificial Neural Networks models. The paper also provides process flowcharts and classification tables to present the prognostic models used to estimate the remaining useful life of a defective bearing. Experimental results confirmed the method robustness and provide an accurate diagnosis regardless of the bearing fault stage, motor speed, load level, and type of supply.