2.1. Study Area
The area selected for the present study is the Koyna Reservoir, situated on the west coast of Maharashtra, India, lies between the latitude of 17◦00’-17◦59’N and longitude of 73◦02’–73◦35’E. The location of the study area along with nine rain-gauge stations in the Koyna watershed is shown in Fgure1. The Koyna watershed has an elongated leaf shape and located near Koyna village about 100 km. from Satara, Maharashtra, India. It is masonry dam built across the river Koyna, …show more content…
The connection weights between processing elements contain the knowledge stored in the artificial neural network model. Usually, the processing elements are classified as input units, output units, or hidden units. The basic structure of an ANN usually consists of three layers: (1) the input layer, where the data are introduced to the network; (2) the hidden layer(s), where data are processed; and (3) the output layer, where the results of given input are produced. The typical ANN for this study is shown in Figure …show more content…
GA-ANN Approach This Genetic Algorithm procedure for global optimization is based on the Darwinian principle of survival of the fittest. Applied to a biological community, it is the principle by which chances of survival of an entire community within a particular environment are increased by discarding inferior members and replacing them by superior offspring. Genetic algorithms are stochastic search techniques that guide a population of solutions towards an optimum using the principles of evolution and natural genetics. Darwin’s natural selection theory of evolution based GP is a relatively new technique. The generalized form of genetic algorithm, GP, is an inductive form of machine learning as it evolves a computer program to perform an underlying process defined by a set of training samples. GP has been successfully applied to complex nonlinear problems and its solution best describes the input-output relationship. The algorithm considers an initial population of randomly generated programs derived from the random combination of input variables, random numbers, and functions, which include arithmetic operators (+, -, x, /), mathematical functions (sin, cos, exp, log) logical/comparison functions (OR, AND), etc. This population of potential solutions is then subjected to an evolutionary process and the fitness measures of the evolved programs are evaluated. The individual programs that best fit the data are then selected from the initial population, discarding no-so-fit