Created by W.Langdon from gp-bibliography.bib Revision:1.8129
Discharge forecasting in natural rivers is a complicated procedure because of uncertainties involved in the behaviour of the flood wave movement. This further leads to solving complex problems of hydrological modelling using soft computing techniques (data-driven models). In real time flood forecasting problems, the data generation is a continuous process. In short term flood forecasting where the accuracy of flood peak value and time to peak are critical, frequent model updating becomes unavoidable. An accurate discharge prediction in the least possible time using online sequential extreme learning machine (OS-ELM) can help policy makers and engineers design a flood control policy and flood warning systems. OS-ELM has not only been used in predictions for hydrological systems, but also for other areas related to environmental sciences. Similarly, a precise prediction of groundwater level using ELM would be very helpful to plan a groundwater abstraction policy in areas where groundwater fluctuations is very high. In case of an in-situ bioremediation system design, the remediation cost can be considerably reduced when an efficient and fast simulator like ELM is adopted. Further, the inclusion of biological clogging of wells while optimizing the cost in in-situ remediation provide a realistic view of the remediation cost. Likewise, the new soft computing techniques like SVM and ELM are a good alternative to the traditional soft computing techniques like ANN and found to be of immense importance while designing the management strategy to control seawater intrusion in coastal areas.",
EN-2013CEZ8051
Supervisor: Shashi Mathur. Co supervsior: Sudheer Ch",
Genetic Programming entries for Basant Yadav