Chapter 4 - Application of soft computing methods in turbulent storm water modeling

Abstract

The advent of computers and the development of hardware and software have led to the rapid emergence and growth of computing intelligence. The field of computational intelligence has revolutionized the development of new unconventional data processing and simulation techniques. Integrating intelligence by replicating human reasoning and behavior in a computing environment increases its ability to analyze information under a dynamic change environment. Soft computing is a set of computational methods inspired by human instinctual ambiguity, human intellect, intuition, and uncertainty about real life. Unlike common techniques that rely on precise solutions, soft computing uses the tolerance of inaccuracy and the trivial and uncertain nature of the problem to provide an approximate solution to a problem in a timely manner. Soft computing, a multidisciplinary field, uses a variety of statistical, probabilistic, and optimization tools to evolve distinct computational methods, including neural networks, evolutionary computing, fuzzy systems, machine learning, and probabilistic reasoning. Among the various subsets of soft computing, neural networks, genetic algorithms, and fuzzy logic are the main actors and are often used for problems related to real-life applications. Inspired by the human brain’s ability to learn, artificial neural networks (ANNs) are able to create subtle relationships between independent and dependent variables whose interactions are unknown, nonlinear, or too complex to display. Genetic Algorithms (GA) is a random search and optimization computational tool that revolves around the evolutionary theories of natural genetics and natural selection. Fuzzy logic (FL) helps to solve real-life problems that are always present in one way or another and are prone to ambiguity and uncertainty (Amiri et al., 2019).