Dissolved oxygen prediction model which based on fuzzy neural network

Abstract

In crab ponds, dissolved oxygen is the foundation for pond cultivation’s survival. The changes of dissolved oxygen content are influenced by multiple factors. Higher levels of dissolved oxygen content are crucial to maintaining healthy growth of crab breeding. Affected by physic-chemical process of aquatic water, the changes of dissolved oxygen content have a large lag. In order to solve the problem of dissolved oxygen forecast, the prediction model which based on fuzzy neural network has been proposed in this paper. It integrated the characteristic of learning fuzzy logic and neural networks optimized performance to realize the dissolved oxygen prediction. The prediction results have shown it more suitable for dissolved oxygen prediction than grey neural network method. The prediction accuracy can meet the need of dissolved control.