Adaptive soft computing methods for control of hemodialysis machines

In hemodialysis systems the control of peristaltic pumps can be vital as these machines support and maintain life functions. In this paper different soft computing approaches are compared for the strict control of peristaltic pumps. An adaptive fuzzy logic and an adaptive neuro-fuzzy inference system are examined then compared with a classical PID controller. A fast controller is sought, which is insensitive in speak of disturbances and has no residual error. Finally, experiences of the designed controller on a real system are detailed in the paper.