Partial discharge measurements

PD is defined as a localised electrical discharge that only partially bridges the insulation between conductors and which may or may not occur adjacent to a conductor [1]. Thus, partial discharges are strictly related to insulation defects and constitute cause and effect of degradation. Hence, PD measurements constitute a powerful tool for guiding decision making in condition based maintenance (CBM), as well as for quality control applications, as shown by huge literature work.

Indeed, PD phenomena manifest in a wide range of ways; for example, internal discharges within the bulk of the insulation, surface discharges at the surface or at the interfaces of two dielectric media or as corona discharges in air. The severity of the effect that PD has on the insulation depends on the nature and the location of the PD generating defect. Therefore, a key step in PD analysis is the identification of the defect, through the characterization of its PD activity.

The activities related to PD measurements which are undertaken in LIMAT laboratory can be summarized as follows:

  •  Measurement sessions carried out both in laboratory environment and on field.
  •  Development of a diagnostic database, to broaden knowledge about PD phenomena occurring on different apparatus and components.
  •  Development of expert systems for automatic identification of PD generating insulation defects, with the aid of artificial intelligence (AI) techniques.
  •  Analysis of PD phenomena through physical models and numeric simulations.

A fundamental contribution in PD measurement and analysis carried out in LIMAT comes from company TechImp (www.techimp.com), which provides LIMAT researchers with the most advanced PD detection and processing systems available. In particular, one precious feature of these instruments consists of the possibility to distinguish and separate different sets of acquired signals on the basis of the pulse shape.

Example of separation of acquired data on the basis of the pulse shape (Subset A: disturbances, Subset B: PD activity, Subset C: further PD activity ).

Measurement sessions are carried out on artificial specimens in controlled environment, thank to advanced HV equipments and a shielded room. In addition, measurements are performed in the lab and/or on field on HV equipments as rotating machines, power transformers (oil and resin-insulated), cables and cable accessories, HF transformers, and capacitors. 

Cable joint under test 

 

Example of PD activity in cable joint

Example of automatic identification result

  

Generator under test 

Example of PD activity in a large generator

Example of automatic identification result

 

Motor under test

 

Example of PD activity in a motor

Example of automatic identification result

 

A database program has been developed in Matlab environment, which is dedicated to PD acquisition data. Such a program, beside usual database features, allows to perform advanced data processing and to develop expert systems. Therefore, the program itself is constantly updated and enforced.

 

Database / processing program views


Expert systems are developed, with the aid of AI techniques, e.g. fuzzy logic, decision trees and concept learning. These systems have two main purposes. The first purpose is to provide automatic tools for identification of PD generating defects. The second purpose is to maximize the information derived from PD measurements, thus allowing a deeper investigation of PD phenomena.

 

References:

[1]   IEC 60270, Partial Discharge Measurements, 3rd edition, March 2001.

[2]   A. Cavallini, M. Conti, A. Contin, G. C. Montanari, “Inferring partial discharge identification through fuzzy tools”, IEEE CEIDP, pp. 698-702, October 2002.

[3]   A. Contin, A. Cavallini, G. C. Montanari, G. Pasini, F. Puletti, "Digital Detection and Fuzzy Classification of Partial Discharge Signals”, IEEE Trans. on DEI, Vol. 9, no. 3, pp.335-348, June 2002.

[4] A. Cavallini, A. Contin, G.C: Montanari, F. Puletti, "A new approach to diagnosis of solid insulation systems based on PD signal inference", IEEE Electrical Insulation Magazine, Vol. 19, no. 32, pp. 23-30, April 2003.

[5]   A. Cavallini, A. Contin, G. C. Montanari, F. Puletti, "Advanced PD Inference in On-Field Measurements. Part 1: Noise Rejection", IEEE Trans. on DEI, Vol. 10, n. 2, pp. 216-224, April 2003.

[6] A. Cavallini, M. Conti, A. Contin, G. C. Montanari"Advanced PD inference in on-field measurements. Part 2: Identification of defects in solid insulation ", IEEE Trans. on Dielectrics and Electrical Insulation, Vol. 10, no. 3, pp. 528-538, June 2003.

[7] A. Cavallini, M. Conti, G. C. Montanari, A. Contin, F. Guastavino, F. Ombello, "Early detection of electrical tree through advanced PD measurement inference techniques", JICABLE, pp. 612-616, Versailles, France, June 2003.

[8] A. Cavallini, D. Fabiani, G. C. Montanari, F. Ombello, F. Franchi Bononi, "Applications to cable diagnosis of new methodology for partial discharge inference", JICABLE, pp. 659-664, Versailles, France, June 2003.

[9] A. Cavallini, F. Ciani, M. Conti, P. F. H. Morshuis, G. C. Montanari, "Modeling memory phenomena for partial discharge process in insulation cavities", IEEE CEIDP, pp. , Albuquerque, USA, October 2003.

[10] J. Borghetto, A. Cavallini, A. Contin, G.C. Montanari, M. de Nigris, R. Passaglia, G. Pasini, "Partial discharge inference by an advanced system: analysis of online measurements performed on hydrogenerators", IEEE Trans. on Energy Conversion, 2003.

[11] A. Cavallini, M. Conti, G.C. Montanari, C. Arlotti, A. Contin, "PD inference for the early detection of electrical tree in insulation systems", IEEE Trans. on Dielectrics and Electrical Insulation, Vol. 10, no. 6, pp., December 2003.