ARTIFICIAL NEURAL NETWORK (ANN) BASED SPATIO TEMPORAL STUDIES FOR A TYPICAL REACH OF THE BRAHMAPUTRA

Abstract

River Brahmaputra, the life line of North-East, is also the sorrow of many during high floods. Huge funds are being spent annually for the protection works at various locations, along. the river Brahmaputra. In order to train and provide access over the river at desired points, it is necessary to understand the fluvial dynamics of the river.: River X-section is one of the important footprints which provide information regarding the river characteristics. Even though, hydrographic surveys at defined X-sections at irregular time intervals are available, yet at any point of space and time, reliable information may not always be available in the required manner. Spatio-Temporal analysis of morphological parameters with the help of latest technique of Artificial Neural Network would help in overcoming this deficiency to a large extent. The advent of digital computers has seen the emergence of analytical tool in analysis and design of Civil Engineering Systems, which earlier seemed too complex or rigorous. At this stage Artificial intelligence technique - especially Artificial Neural Nets, or Neural Nets, are beginning to dominate most of the analytical and design aspects. The basic advantage in the use of Neural Nets being the capability of handling imprecise, imperfect and incomplete data while yet producing acceptable solutions. In the present study the application of Neural Nets has been examined with specific reference to spatio-temporal analysis of morphological parameters in respect of Brahmaputra reach containing "Majuli Islands", the world's largest river island, where almost 1/3`d of it's total area has already been eroded by Brahmaputra. The main objective of the study can be enumerated as follows: (1) To construct Neural Network model to correlate river X-section with respect to time. VII (2) To study the river Shieft Pattern with reference to time using Neural Nets. (3) To develop a functional Neural Net model to correlate X-sectional information with river Shi' ft and top width patterns. (4) Developing the Neural Nets to generate X-sections with reference to space and time using the hydrographic survey data and correlating the same with the latest available Satellite data. It is also proposed to use the above spatio-temporal model to correlate it's findings with the latest available Satellite data of the river Brahmaputra, being utilized in an R&D project undertaken by WRDTC, U.O.R. Roorkee at the behest of MOWR, Govt. of India.