The use of intelligent data reduction technique in analysing the substation events

The paper proposes a relatively new approach based on rough sets to extract knowledge from a substation and provide a synthetic output suitable for decision making. This involves a mathematical technique based on the classification of objects into similar classes. The aim is to reduce the size of substation data received from IED relays while keeping all the essential information intact. This is done by discovering dependencies among events and attributes before superfluous and redundant data are identified and removed. The reduced data dimension improves the performance of substation data analysis. The problem with analysing substation data is that sometimes the datasets can be too complex and conflicting. This may result a large solution set in which we simply do not know what the best solution would be. To enhance the rough set data analysis in data/information overwhelm condition, an efficient approximation algorithm is used. The work emphasises on post fault analysis of the protection and breaker responses in a substation. It is designed to help the operator understand overwhelming alarm messages, or longer term help engineers analyse what went wrong.