Wind Farm Layout Optimization

Pranav Jain, Purdue University

Abstract

In this thesis layout optimization is done for wind farms having single hub height wind turbines and wind farm having multiple hub height turbines. Genetic Algorithm is used to solve the optimization problem. Wake model similar to Jensen's wake model is used to study the wake effects. For wind farms having single hub height, three cases are considered: constant wind speed and varying wind direction, varying wind speed and constant wind direction and varying wind speed and wind direction. Weibull distribution is used to model variation in wind speed and the variation in wind direction is determined by a bimodal normal distribution. The effect of changing thrust and power coefficients with wind speed is also studied. For wind farms having multiple hub heights 50 m or 78 m hub height turbines can be present in certain areas call restricted areas and 60 m turbines can be present on rest of the farm. The restricted areas are modeled using circles. Analysis is done for multiple cases for 50 m and 60 m hub height turbines and then for 60 m and 78 m hub height turbines. First there is only one predened restricted area, then the eect of two predened restricted areas is studied. This is followed by optimizing the location and size of a single restricted area and both the restricted areas. Finally, the location and size of both the restricted area is optimized along with the hub height of the turbines present in those areas. The power generated by wind farm with two restricted areas was more than the power generated by wind farm with single restricted area which was in turn more than the power generated by wind farm having single hub height turbines. When the wind speed is more wind farm having 50 m and 60 m hub height turbines is better and when the wind speed is less wind farm with 60 m and 78 m turbines is better.

Degree

M.S.M.E.

Advisors

Lin, Purdue University.

Subject Area

Mechanical engineering

COinS