Application of neural networks in coastal engineering

Abstract

Neural networks have the ability to recognize the hidden pattern in the data and accordingly estimate the output values. Provision of model-free solutions, data error tolerance, built in dynamism and lack of any exogenous input requirement makes the neural network attractive. A neural network is an information processing system modeled on the structure of the dynamic process. It can solve the complex/nonlinear problems quickly once trained by operating on problems using an interconnected number of processing units. Its merit is the ability to deal with information whose interrelation is ambiquous or whose functional relation is not clear. Basically we need actual input-output data of any particular process/system, which will be used to train the neural network. Once the network is trained, it trained, it estimates/predicts the output for unknown input data without any assumptions. In many occasions, neural network predictions are much closer to target output and better than other numerical / empirical methods. That is why it is becoming popular in various fields including coastal engineering. Waves and tides will play important roles in coastal erosion or accretion. This paper briefly describes the back-propagation neural networks and its application towards prediction of waves and tides