Estimation of Remaining Lifetime of Power Connections Using Infrared Thermography

One of the measures used to prevent the occurrence of power connection failures is to estimate the anomalous behavior of an electrical connection in the early stages of contact operation when the first signs of critical stage become apparent. This paper investigates the possibility of applying the statistical prognostic methods for determining the residual life of power connections. I. INTRODUCTION It is now well established that approximately one quarter of failures of power equipment is related to electrical contacts. Such accidents often results in fire and as a result serious consequences may occur. Hence, it is clear that electrical contacts are essential components of equipment whose reliability has to be ensured at high levels. In spite of seeming simplicity of contact joint design, the reliability of electrical contact is determined by complicate physical processes occurring at contact spots (a-spots) under the influence of electrical current, temperature and other environmental factors. There are two approaches to estimate the reliability of crucial equipment, including electrical contacts. The first one is based on the integrated average statistical data analysis of general population, such as reliability function, mean operating time between failures, gamma- percentile operating time to failure, etc. This approach is acceptable at the equipment designing stage, while the average lifetime for a large number of equipment (general population) has to be estimated. This approach is frequently used to schedule the preventive inspection and repair works. However, this approach is disadvantageous since it does not take into account different operating times of different contact joints under current load, different level of current loads and consequently degradation stages of different joints. On the other hand, if the residual lifetime of a contact joint is predicted based on its actual state at the particular time instance, then its lifetime can be estimated exactly for the given joint in the given specific conditions. The conventional methods of evaluating the states of power connections, including measurements of the contact resistance of a joint during the OFF periods and other measurement procedures, are not always as effective as expected. The problem of residual lifetime forecasting for the electrical connections may be solved if there is a capability to monitor continually the evolution of the key (resource) parameters until reaching their critical values considered as the failure criteria. However, continuous monitoring of the key parameters determining the power connection performance under operating conditions is not always possible. The most effective approach from the economical and also practical viewpoints would be periodic monitoring of key parameters determining the quality of an electrical connection whose changes with the time can be used to forecast the occurrence of the critical stages of connection performance. The above forecasting may be completed based on