Electric Power Systems Research 81 (2011) 1709–1714 Contents lists available at ScienceDirect Electric Power Systems Research journal homepage: www.elsevier.com/locate/epsr Age-dependent maintenance strategies of medium-voltage circuit-breakers and transformers  Xiang Zhang ∗ , Ernst Gockenbach Institute of Electric Power Systems, Division of High Voltage Engineering, Leibniz University of Hannover, Callinstrasse 25A, Hanover 30167, Germany article info Article history: Received 14 October 2008 Received in revised form 13 February 2011 Accepted 10 March 2011 Available online 7 May 2011 Keywords: Age Circuit-breaker Cost Failure rate Maintenance strategy Transformer abstract The general life and reliability models of electrical equipment are essential to evaluate their actual con- ditions because of the degradation of equipment. To optimize the maintenance strategies for maximal reliability and minimal cost in a quantitative way, available maintenance models of ageing equipment shall be found to describe actual maintenance actions in time-series processes. In particular, a functional relationship between failure rate and maintenance measures is to be developed for electrical equipment. This paper demonstrates some actual examples by applying these models and the results show the value of using a systematic quantitative approach to investigate the effect of different maintenance strategies. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Deregulation of the power system market has forced electric utilities to scrutinize investment and maintenance expendi- tures much more rigorously than in the past. Many focus their investments on maintaining the competitive position of power gen- erating assets, while trying to squeeze more performance from ageing power delivery assets with less expenditure. However, in the power transmission and distribution systems, the annual expen- ditures for maintenance and replacement average only 1% or less (equal to 20 billion Euro for distribution systems in Germany and 400 billion Dollar for transmission systems in USA), which corre- sponds to an expected lifetime for more than 100 years of electrical equipment [1]. Owing to the limited reinvestments, the age of elec- trical equipment will increase so that the utilities have to face various market requirements. On the one hand, customers are paying for a service and the authorities are imposed regulation, supervision, and compensation depending on the degree to which contracts and other obligations are fulfilled. On the other hand, utilities must ensure that their expenditure is cost-effective.  This work was supported by the National Research Council of Germany under the Contract SPP 1101. ∗ Corresponding author. E-mail address: zhang@si.uni-hannover.de (X. Zhang). The easiest way and the most widely used strategy today is the time-based maintenance (TBM) [2]. There are fixed time inter- vals for inspections and for certain maintenance works. However, it seems that the time intervals chosen are far from the safe side and the total replacement expenditure is extremely expensive, as there are many inspections revealing no problems at all. So, the time intervals obviously can be extended – the question is from which point of time the occurrence of failures will increase signif- icantly. In order to obtain useful information about the actual conditions of equipment, condition monitoring technique and condition- based maintenance (CBM) has been well developed [3–5]. Condition monitoring technique means mainly sensor develop- ment, data acquisition, data analysis, and development of methods for determination of equipment condition and early fault recogni- tion. The importance of monitoring methods can recognize which measured parameter affects the ageing of equipment to a greater or lesser degree. It should support the introduction of condition- based maintenance and help to avoid unexpected outages. The cost of sufficient instrumentation can often be quite large and off-line measurement even causes more significant outage. In the majority of cases, there are neither communication links nor suitable sen- sors available for monitoring from remote. In terms of numerical protection devices and control systems, those engineers are suf- fering from data overload. Also, from the technical side, it is not always as simple as possible. Even if some types of equipment can easily be observed by measuring simple values as temperature or 0378-7796/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.epsr.2011.03.018