In this chapter, motor current signature analysis based on squared envelope spectrum is applied in order to identify and to estimate the severity of outer race bearing faults in induction machine. This methodology is based on conventional vibration analysis techniques, however, it is, non-invasive, low cost, and easier to implement. Bearing fault detection and identification in induction machines is of utmost importance in order to avoid unexpected breakdowns and even a catastrophic event. Thus, bearing fault characteristic components are extracted combining summation of phase currents, prewhitening, spectral kurtosis and squared envelope spectrum analysis. Experimental results with a 0.37 W, 60 Hz, and three-phase induction machine demonstrated the methodology effectiveness.

No abstract

No abstract

This paper shows the behavior of same structure of an hybrid active series power filter (HASPF) tuned in three different frequencies. The developed strategy is based on the Modified Synchronous Reference Frame Controller. The initial structure of the passive filter was chosen to be tuned for the fifth harmonic. The performance of the system is compared with and without the active compensation. The ability of the hybrid topology in changing the resonant frequency is shown. The active hybrid structure is first tuned for the fifth harmonic, further it is electronically re-tuned to the third and finally to the seventh harmonic. Practical results exhibits these tuning changes showing as close as the passive filter can approach the absolutely compensation for a chosen frequency. The algorithm of the HASPF is executed in the DSP TMS320F2812 and the inverter switching frequency is 20 kHz.Keywords -Hybrid active power filter, selective harmonic compensation, modified synchronous reference frame, non-linear loads.

The biggest challenge in the contemporaneous world: maintain the supply of electric energy with reliability and quality. A lot of technological solutions, involving telecommunications, energy storage, dispersed generation (DG) with alternative sources and another usual electric generation, are in large scale implementation. Moreover, Microgrids (MG) concept is in large discussion in order to obtain better results in operation and management of this new distribution system. Local black start strategies are one of the most important implementation for that new kind of grid. In the first phase of the R&D project developed with CPFL, one Brazilian electric utility, inside of the environment of ANEEL, the Brazilian Regulatory Agency to the electric system, it was made a study concerning the technologies of distributed micro generators and connection requirements. Models and control strategies simulations have been done in order to reproduce the performance of this energy sources when connected in the grid during the practice tests. It was evaluated topologies with resistive loads, capacitive loads, and inductive/non-linear loads. These topologies are in parallel with the distribution utility grid using a special test bench able to synchronize the energy and power quality measurements of all devices.

No abstract

No abstract

This paper presents an alternative technique for improving the security of Electric Power Control Systems by implementing anomaly detection methods to identify attacks and faults. By using Rough Sets Classifi cation Algorithm, a set of rules can be defi ned. The alternative approach tries to reduce the number of input variables and the number of examples, offering a more compact set of examples to fi x the rules to the anomaly detection process. An illustrative example is presented.