Water Resources Research Act Program

Details for Project ID 2014DC157B

Prediction of Surface Water Supply Sources for the District of Columbia Using Least Squares Support Vector Machine (LS-SVM)

Institute: District of Columbia
Year Established: 2014 Start Date: 2014-03-01 End Date: 2015-02-28
Total Federal Funds: $9,417 Total Non-Federal Funds: $33,208

Principal Investigators: Nian Zhang, Pradeep Behera

Abstract: Long-term evaluation on the potential impact of climate change on the water supply resources of the Washington, D.C., metropolitan area is very important. The proposed research are a) to develop a predictive model based on least squares support vector machine (LS-SVM) that forecasts the future streamflow discharge using the past streamflow data and gage height. A Gaussian Radial Basis Function (RBF) kernel framework was built on the data set to optimize the tuning parameters and to obtain the moderated output. The prediction capability with different kernel functions will be explored. The training process of LS-SVM was designed to select both kernel parameters and regularization constants. The USGS real-time water data were used as time series input. b) to compare the performance of a variety of promising computational intelligence methods that can efficiently address the water quantity problems in the watershed. They include the recurrent neural networks, hybrid particle swarm optimization and evolutional algorithm, and the proposed LS-SVM method. c) to generate a flow-duration curve (FDC) for each branch of the Potomac River, and analyze their accumulated impact to the Potomac River. d) to organize a one-day workshop to train the researchers and students how each of the computational intelligence methods works, and demonstrate their prediction capabilities.