용존산소 농도모의시 블랙 박스 모형을 이용한 시계열 분석

Black-box model is used to analyze the changes of water quality which obtains data easily. The models development were based on the data obtained from Jan. 1990 to Dec. 1997 and followed the typical procedures of the Box-Jenkins method including identification, estimation. The seasonality of DO and Temperature data to formulate for the SARIMA model is conspicuous and the period of revolution was twelve months. The Multi-layer Perceptron(MLP) neural networks with a single hidden layer are trained to perform one step ahead prediction on water quality data. The prediction ability of SARIMA model, state space model and Neural Network model were tested using the data collected from Jan. 1998 to Oct. 2001. There were good agreements between the model predictions and the field measurements. The performance of the SARIMA model, state space model and Neural Network model were examined through comparisons between the historical and generated monthly dissolved oxygen series. The result reveal that the Neural Network model lead to the improved accuracy.