A statistical-based criterion for incipient fault detection in underground power cables established on voltage waveform characteristics

• Identifying abnormality and determining time of disturbance occurrence. Discriminating the incipient fault from capacitive switching sudden load variations. harmonic loads considering arc models. The faults which mainly occur due to electric occurrence in power cables with insulation defects are hardly detectable by conventional protective relays, over can develop into a permanent system. Employing Kalman filter, this paper puts forward method detect discriminate them other similar incidents proposed is established on comparison between waveform measured voltage fundamental component voltage, estimated filter algorithm sending end cable during fault. difference waveforms, detection discrimination carried out within two stages. In first stage, event relegalized comparing standard deviation obtained error certain threshold. second stage conducted find based non-attenuating characteristic quasi-periodic nature feasibility verified through computer simulation using four different models also acquired experimental data real faults.