Suspended sediment concentration prediction by Geno-Kalman filtering

Abstract

More accurate prediction of suspended sediment concentration will likely lead to more economic hydraulic construction and provide a valuable basis for the optimum operation of water resources. The majority of past models have relied on simple regression analysis relating discharge to concentration. A new adaptive prediction approach termed Geno-Kalman filtering (GKF), combining Genetic Algorithm and Kalman filtering techniques is proposed. The model is formed in three steps. Firstly, discharge and suspended sediment concentration are related by using dynamic linear model. Secondly, an optimum transition matrix relating these two state variables is obtained by Genetic Algorithms (GAs), and an optimum Kalman gain is calculated. Thirdly, Kalman filtering is used to predict the suspended sediment concentration from discharge measurement. The proposed method is applied to measurements at the Mississippi River basin in St. Louis, Missouri, and is found to result in smaller absolute, mean square, relative errors compared to perceptron Kalman filtering. Furthermore, Geno-Kalman filtering method outperforms the perceptron Kalman filtering and least square methods in terms of coefficient of efficiency.