Environmental Parameters Prediction of Marine Ranching Based on Fuzzy Neural Network

Abstract

Environmental factors in marine ranch waters feature nonlinearity, time variability and uncertainty. Traditional empirical monitoring methods are difficult to realize accurate prediction of environmental parameters and ecological risk early warning. Taking a typical marine ranch in Laizhou Bay as the research area, this paper selects dissolved oxygen (DO), chlorophyll a (Chl-a) and chemical oxygen demand (COD) as core monitoring indicators, and constructs a prediction and ecological early warning model for marine ranch environmental parameters using the Takagi Sugeno (T-S) fuzzy neural network. Based on monthly monitoring data from January 2023 to December 2024 collected from 22 monitoring stations covering the core artificial reef zone, transition buffer zone and offshore control zone of the marine ranch, MATLAB platform was adopted for data normalization, network initialization, training simulation and case prediction. Seventy percent of samples were randomly selected as the training set and the remaining 30% as the test set. Model predicted outputs were compared with field measured values to analyze model errors and fitting performance. The results show that the T-S fuzzy neural network prediction model achieves high fitting degree for time series variations of DO, Chl-a and COD, with test set errors maintained at a low level. The model can effectively capture precursors of ecological risks such as summer autumn hypoxia and abnormal increase of chlorophyll a in the Laizhou Bay marine ranch. Water quality in most periods of the artificial reef core area meets Grade Ⅱ standards for seawater quality, while local stations exhibit elevated chlorophyll a and decreased dissolved oxygen in summer and autumn. The proposed model supports short term prediction of key environmental parameters and hierarchical ecological early warning for marine ranches, providing technical support for intelligent monitoring and risk prevention of Laizhou Bay marine ranches.