Design of a Fuzzy Logic Controller for Optimal African Catfish Water Production

Abstract

Fish, unlike other animals, feed and defecate inside the same water where they live. When water quality depreciates, consumed feed is not properly converted into body flesh, poor growth is recorded, fish survival is affected and ultimately massive fish deaths may occur. In fish production, key water quality parameters which need to be continually monitored and controlled are temperature, dissolved oxygen (DO), pH, and ammonia. These parameters are highly non-linear, and thus difficult and expensive to use conventional controllers. Fuzzy logic controllers can suitably adapt to non-linearity because it uses sentences for control actions rather than equations. FisPro (Fuzzy Inference System Professional) is used to create fuzzy inference system (FIS)/ fuzzy logic controller (FLC) for simulating physical or biological systems. The selected water quality parameter are temperature, 14 to 45oC, potency of hydrogen (pH), 0 to 14, and turbidity, 1 to 5. At a pH of 3.5 and 39oC, the aerator speed is 6982 rpm. Similarly, at a pH of 3.5 and turbidity of 1.5, the valve position is 47o. The results obtained shows the completeness, consistency and continuity of the designed rule system. Nine rules were designed for each control action giving a total of eighteen rules. The control actions are; water discharge/refill in form of valve actuation and aeration in form of air pump activation.