Decomposition-Based Multi-Objective Optimization of Energy Noise Trade-Off in a Wind Farm: A Hybrid Approach

Decomposition-Based Multi-Objective Optimization of Energy Noise Trade-Off in a Wind Farm: A Hybrid Approach

Prateek Mittal (Indian Institute of Technology Hyderabad, India) and Kishalay Mitra (Indian Institute of Technology Hyderabad, India)
DOI: 10.4018/978-1-5225-2990-3.ch008
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

A multi-objective optimization case study of maximization and minimization of energy generation and noise propagation is considered here. A novel hybrid methodology, as a combination of probabilistic variable decomposed multi-objective evolutionary algorithm (VdRBNSGA-II) and the newly developed deterministic gradient based Pareto frontier construction approach (nD-NNC), has been proposed to determine the optimum layout of turbines (numbers and locations) inside a wind farm. In contrast to previous case studies, the proposed approach is able to yield the alternative energy-noise solutions along with the additional information on corresponding turbine layouts (numbers and locations) on a single Pareto front. As a result, it provides a decision maker with an ample of choices to choose from different competing solutions based on the existing standards and guidelines.
Chapter Preview
Top

Motivation

Most of the WFLO problems assume the number of wind turbines to be erected in a wind farm is known and their optimal locations are to be determined, whereas the wind farmers themselves are not sure of the maximum number of turbines and their optimum locations inside a wind farm. Therefore, it is necessary to develop an algorithm or a method for the simultaneous determination of optimum locations and the number of turbines inside a wind farm. However, simultaneous determination of optimal number and locations of turbines in a wind farm involves both binary (1(yes) or 0 (no) decisions for turbine placement) and continuous (location coordinates) variables, which lead to the formation of mixed integer non-linear programming problem (MINLP), which in general are complex and highly non-linear in nature (Vasant, Webber and Dieu, 2016). Moreover, various methodologies and layouts have been developed to solve the combinatorial optimization (binary and continuous decision variables) problem of wind farm micro-siting with an aim of maximizing energy yield with minimum investment. However, only a few works have considered the effect of noise, which is of immense importance for wind farmers as the noise generated by a wind farm has to be restricted within a certain level not to disturb the surrounded dwelling.

Complete Chapter List

Search this Book:
Reset