This thesis dealt with a basic study on the development of condition monitoring diagnostic technique for power facilities installed inside closed-switchboards. Major causes of accidents in a closed-switchboard are series arc discharge by incomplete connection and partial discharge by insulation degradation.
The series arc discharge was simulated using a terminal block and an electrode system of a stainless needle-carbon rod. The corona discharge was simulated in five types of electrode system which consist of a needle and a plane electrode. The experiment was carried out in an electromagnetic shielding room, and the measurement system was composed of an Ultra log antenna of 30MHz∼2GHz and an EMI test receiver of 150kHz∼2GHz, and a spectrum analyzer.
Frequency spectrum of series arc discharge was distributed in ranges of 30MHz∼200MHz for the needle-carbon rod and of 30MHz∼400MHz for the terminal block. In corona discharge, frequency spectrum existed in ranges of 30MHz∼2GHz depending on types of electrode system.
The experimental results showed that frequency spectrum is different as abnormal condition varies, and the inner space of a closed-switchboard is not influenced by outer noise generated by electrical accident or broadcast wave.
Consequently, it is possible to diagnose the condition of closed-switchboards by detecting radiated electromagnetic waves.