Asenkron Motorlu Sistemde Bulanık Mantık Çıkarımı İle Kestirimci Bakım

Abstract

Extended Abstract Industrial systems are expected to operate with high performance and continuity. Failures that may occur in the parts that make up the working system can lead to production losses. The necessity of keeping the total equipment efficiency high in all sectors served by industrial systems necessitates maintenance work. In order to avoid production and labor losses due to malfunctions, it is necessary to plan maintenance without causing downtime in the systems. The predictive maintenance method, which has a strong place in maintenance planning processes, stands out in terms of not performing unnecessary maintenance and avoiding failure due to lack of maintenance. In this paper, the stator current, vibration, stator winding temperature and bearing housing temperature values obtained during the operating conditions of an induction motor are collected from the field with the help of a PLC, time-labeled and written to a database and then used both to generate test data and to determine membership function values. Based on the data collected from the asynchronous motor operating in the field, motor and instrument label values and expert opinions, a fuzzy logic based inference system was designed with the "Matlab Fuzzy Toolbox" application. The "motor