Partial discharge detection in medium voltage and high voltage cables: maximum distance for detection, length of cable, and some answers
Autor: | Gian Carlo Montanari |
---|---|
Přispěvatelé: | Montanari, gian Carlo |
Rok vydání: | 2016 |
Předmět: |
Engineering
partial discharge detection cable reliability and failure 02 engineering and technology 01 natural sciences law.invention law 0103 physical sciences Shielded cable VLF cable testing Cable PD 0202 electrical engineering electronic engineering information engineering cable Electrical and Electronic Engineering Direct-buried cable 010302 applied physics detection distance business.industry Electronic Optical and Magnetic Material 020208 electrical & electronic engineering Electrical engineering High voltage Electronic Optical and Magnetic Materials Partial discharge Fiber cable termination All-dielectric self-supporting cable Skin effect business |
Zdroj: | IEEE Electrical Insulation Magazine. 32:41-46 |
ISSN: | 0883-7554 |
DOI: | 10.1109/mei.2016.7552375 |
Popis: | Partial discharge (PD) measurement is becoming a common procedure in the commissioning and off-line diagnostic testing of cable systems. On-line PD cable monitoring is beginning to show growing interest as well. Arising from this is an important question being asked by testing companies, maintenance personnel, and asset managers: "What are the maximum lengths of high voltage (HV) and medium voltage (MV) cables that can be scanned with sufficient sensitivity for PD activity by present day PD detectors?" In other words, what is the maximum distance a PD measurement point can be from the PD source and still detect the PD. The answer can be complex, or not satisfactorily understood, being related to the measurement sensitivity, voltage level, type of insulation, and cable construction. Detection also depends on the bandwidth of the measurement circuit. The signal-to-noise ratio must be large enough to enable PD detection at a level compatible with the reliability of the cable, that is, terminating maintenance or commissioning diagnostic tests before cable breakdown. Indeed, the capability of detecting pulses with an apparent charge below 5 pC is a common requirement for HV polymeric cable systems [1], whereas a much lower sensitivity, i.e., 5 nC, for mass-impregnated cables is generally accepted. The meaning of such limits in cable systems is doubtful as PD measurements are most often done from the cable termination because localized sensors are often unavailable, and since not in close proximity to the source, a lumped-parameter network representing the cable network for picocoulomb calibration is invalid. In a cable system PD pulses travel on the conductor and metal sheath, and the cable is viewed as a distributed parameter network; thus, the PD quantity measured is typically volts or millivolts. As the PD pulses travel along the cable, they lose frequency content as a consequence of skin effect or ohmic loss at high frequencies, and semicon and insulation dielectric losses [2], [3]. |
Databáze: | OpenAIRE |
Externí odkaz: |