Optimization of Water Level Control Systems Using ANFIS and Fuzzy-PID Model

Abstract

Water flow measurements have been needed by controllers in industrial processes. The quantity of water must be determined to control the volume of water used in the storage tank. Water flow performance control models based on tanks are required using a Proportional-Integral-Derivative (PID) control system. This system uses a flow sensor to detect the speed of an actuator. Actuators stabilize the output water speed per minute at a certain point. Manually determining the value of a PID constant will be very difficult and not optimal. Then we need an automatic and accurate control method. This study focuses on four comparisons of designed methods related to water level without control, standard PID method, Fuzzy Logic method, Fuzzy-PID method, and Adaptive Neuro-Fuzzy Inference System (ANFIS) method. The simulation results found that the four control models have different performance. The PID-ANFIS model obtained the smallest overshot value in the PID-ANFIS model of 0.5135 pu, the smallest undershot in PID-ANFIS was 0.5291 pu. Current Output Output obtains the smallest overshot value in the PID-ANFIS model of 0.0023 pu, the smallest undershot in the PID-ANFIS model is 0.0014 pu.