Özet:

With a healthy data flow between units, digitalized production systems and smart factory structures that are automated according to this digitalization process are getting more and more place in the production industry. These systems, although production has made significant developments and technological advances, bring a variety of problems. One of these is the process of resolving an abnormal situation that occurs in autonomous operating production systems, quickly detected. In recent years there have been various studies to detect anomalies. One of the most supported fields in the detection of anomalies is machine learning algorithms. In this study, the performance of various machine learning algorithms was tested on two different data sets, which were obtained from an applied prototype study on energy optimization of high storage systems. As a result, learning models created with the artificial nerve networks, the C4.5 Decision Tree, the random forest and the k nearest neighbor algorithms have achieved a high success rate in detecting anomalies within the tested data sets. Specifically in these algorithms, the Rastgele Forest algorithm has attracted attention with a level of accuracy performance of approximately 98%.

Anahtar Kelimeler: