Modelo basado en redes neuronales artificiales para la evaluación de la calidad del agua en sistemas de cultivo extensivo de camarón

Aquaculture is a commonly practiced activity worldwide. In Mexico, shrimp represents a significant source of the income generated by aquaculture. Since the success of shrimp farming depends on good water quality, its monitoring is essential. This work presents a new computational model to assess the water quality of large shrimp ponds (Litopenaeus vannamei). An artificial neural network (ANN) was used to create a water quality index, with which a mathematical relationship can be established between the dynamics of environmental parameters and different water quality conditions (excellent, good, average, and poor). Four parameters that were important for the habitat were selected: temperature, dissolved oxygen, salinity, and pH. The results show that the proposed model performs well and efficiently, as compared to other evaluation models used for this purpose. The evaluations demonstrate that ANN is a good option for evaluating and detecting optimal and undesirable conditions, contributing to good water management for this type of farming.