Lightning protection of power system equipment

Abstract

The long history of lightning research and lightning protection studies can be seen to have followed clearly defined phases. Since 1945, work has concentrated on the provision of analytical methods, so that the designer could predict the lightning performance of lines and the effectiveness of the lightning protection systems used for substations and equipment. Apart from major improvements in the characteristics of surge arresters, no radical changes in lightning protection technology occurred in the period 1945 to 1975. In Australia, lightning protection for power system equipment largely followed the practices outlined in Westinghouse (1950), Bewley (1963), Lewis (1959) and Beck (1954)*. Certain assumptions were required in the application of overseas practices to Australian conditions. These included the general assumption that the existing (overseas) knowledge of lightning characteristics was applicable in Australia. More specifically, it was assumed that overseas data on the electrical properties of wood could be used for lines designed with Australian hardwoods; and for unshielded distribution systems, it was assumed that lightning protection practices developed overseas for 4-wire grounded neutral systems could be applied to the Australian three wire systems (in this case, inline equipment, such as a transformer, is connected across phases rather than between phase and neutral). These assumptions began to be questioned in about 1950, and this prompted the initiation of a long term project into the meteorological aspects of lightning in Australia (by S. A. Prentice, Foundation Professor of Electrical Engineering, University of Queensland). In 1956, preliminary studies were made of the impulse properties of Australian hardwoods (by the author), and in 1963, this work was extended into a full research project The Electrical Properties of Australian Hardwoods. This work could only proceed after the completion of the High Voltage Laboratory, constructed in the period 1959 to 1963 with funds provided, through the Electrical Research Board.The growing interest in lightning research at the University of Queensland prompted the author to study other lightning problems of concern to Australian supply authorities. These included the high lightning outage rates recorded by some unshielded wood-pole lines, severe lightning damage to distribution transformers and surge arresters in south-east Queensland, and the incidence of double-circuit faults on shielded steel-tower transmission lines. In due course, this led to a more general interest in all aspects of the lightning protection of power system equipment. Thus, from about 1969, the work proceeded as a series of inter-related projects; all aimed at a common objective of achieving better lightning performance. More recently, this objective was seen to be part of broader study, namely reliability and cost/benefit analyses of overall power system performance. Thus, from about 1974, the lightning research work was placed into a broader perspective, whereby the benefits to be gained from improved protection had to be assessed in terms of cost effectiveness.In the meantime of course, research into lightning protection continued in overseas countries, particularly those stimulated by the development of E.H.V. networks. The first 345kV lines, notably in U.S.A., experienced surprisingly high lightning outage rates, even though the currently available prediction methods (eg., AIEE Committee, 1950) indicated that such lines would be virtually lightning-proof. This discrepancy resulted in a spate of research work - into the physics of lightning, the shielding effectiveness of overhead ground wires, electromagnetic field theory analysis of lightning overvoltages, the development of improved generalised-curve prediction methods, the use of geometric scale models, and the application of Monte-Carlo techniques in lightning performance simulations. Much of this work was dependent on access to new and sophisticated technologies developed by the mid-sixties, in particular nano-second pulse technology and high speed medium capacity digital computers. Notable contributions were made by Anderson, Fischer, Hileman, Whitehead and Young in America, and by Burgsdorf, Golde, Kostenko, Lundholm, Maikopar and Popolansky in Europe. After the immediate problems were largely solved (poor shielding performance), much of the overseas effort was re-directed to the more difficult problem of switching overvoltages on EHV networks. However, and largely through the activities of CIGRE Study Committee No. 33, research work was continued (on a reduced scale) into the physics of lightning and lightning protection. Also, a small but increasingly concerned group of overseas workers began to draw attention to the problem of lightning protection for unshielded distribution networks. By comparison with transmission, relatively little research had been directed towards lightning damage on distribution systems, and some important studies were undertaken (Armstrong et al. 1967, Task Force 1969). ........