A review of using data mining techniques in power plants

Abstract

Data mining techniques and their applications have developed rapidly during the last two decades. This paper reviews application of data mining techniques in power systems, specially in power plants, through a survey of literature between the year 2000 and 2015. Keyword indices, articles’ abstracts and conclusions were used to classify more than 86 articles about application of data mining in power plants, from many academic journals and research centers. Because this paper concerns about application of data mining in power plants, the paper started by providing a brief introduction about data mining and power systems to give the reader better vision about these two different disciplines. This paper presents a comprehensive survey of the collected articles and classifies them according to three categories: the used techniques, the problem and the application area. From this review we found that data mining techniques (classification, regression, clustering and association rules) could be used to solve many types of problems in power plants, like predicting the amount of generated power, failure prediction, failure diagnosis, failure detection and many others. Also there is no standard technique that could be used for a specific problem. Application of data mining in power plants is a rich research area and still needs more exploration.