Detection of Vortex Cavitation With The Method Adaptive Neural Fuzzy Networks in the Deep Well Pumps

Nowadays submersible deep well pumps are the most used irrigation systems in agriculture field. Efficient operation and economical life of is an important issue. One parameters affecting pump efficiency cavitation The one problems frequently faced that widely could cause more undesired effects such as loss hydraulic performance, erosion, vibration noise. This paper presents a novel model for detection vortex system using adaptive neural fuzzy networks. data submergence, flow rate, power consumption, pressure values, noise values training ANFIS (Adaptive-Network Based Fuzzy Inference Systems) network acquired from experimental pump. In this study, we use to sixty-seven process, while fifteen have testing our model. average percentage error (APE) has obtained 0.08 % 0.34 respectively 67 15 test data. performance implemented shows advantages ANFIS. result study can be successfully detect cavitation. two contributions which usage value on find out by During cavitation, must change controller prevent unwanted errors. strategy proposed preliminary automatic control. Also control pumps, because easy set up no need extra cost. based real-time applicable thanks rapid It possible safe boundaries submergence Thus, users adjusting controllable increase efficiency.