Assessment and prediction of water quality in shrimp culture using signal processing techniques

In recent years, artificial intelligence methods have proved appropriate for the treatment of environmental problems. This paper presents a novel work for the assessment and prediction of water quality in shrimp aquaculture based on environmental pattern processing. Water quality studies are based on analyzing negative concentrations of compounds in shrimp ponds that inhibit good growth and reproduction of organisms. The physical-chemical variables are classified based on the negative ecological impact using the Gamma (Γ) classifier, which calculates the frequency and deviation of the measurements from a specific level. A fuzzy inference system processes the level classifications using a reasoning process that determines when a specific concentration is good or harmful for the organism and provides a water quality index, which describes the condition of the ecosystem: excellent, good, regular, and poor. An autoregressive model (AR) predicts a section of an environmental signal using historical information, and the set of predicted variables are assessed in order to estimate future water quality conditions in the system. This methodology emerges as a suitable and alternative tool to be used in developing effective water management plans.