Comprehensive power swing detection by current signal modeling and prediction using the GMDH method

Abstract Power swing is an undesirable variation in power flow. This can be caused by large disturbances demand load, switching, disconnection or reclosing lines. phenomenon may enter the zones of distance relays and cause relay malfunction leading to healthy lines undermining network reliability. Accordingly, this paper presents a new detection method based on prediction current signal with GMDH (Group Method Data Handling) artificial neural network. The main advantage proposed over its counterparts immunity noise effect signals. In addition, detect stable, unstable, multi-mode swings capable distinguishing them from variety permanent faults occurring simultaneously. tested for different types simultaneous using DIgSILENT MATLAB, compared some latest methods. results demonstrate superiority terms response time, ability varieties, that occur simultaneously swings.