qfliu / 457aproject

457AProject

Wind farm layout is an optimization problem involving multiple parameters such as the number of wind turbines required in the farm, wind speed, wind direction and location of the turbines.

In this project, we are trying to implement a solution to solve for the wind farm optimization problem in 5 different approaches: 1 Tabu Search Algorithm(TS) 2 Simulated Annealing(SA) 3 Genetic Algorithm(GA) 4 Partical Swarm Optimization(PSO) 5 Ant Colony Optimization(ACO)

457aproject's Issues

Maybe add multiple swarm for pso

Change wake model in SOW

Prevent duplicate locations resulted in crossover in GA

Possible ACO implementation, ask TA for clarification

http://www.sciencedirect.com.proxy.lib.uwaterloo.ca/science/article/pii/S096014811100694X

Self created algorithm:
for turbine i = 1 to N
for ant 1 to k
compute probability of all alternative locations for turbine i
choose an alternative location for turbine i and deposit pheromone on this location
calculate objective function based on new wind farm
update current best
end
update the wind park with best solution obtained from ants
reset all pheromone of all locations
end

check invalid answer in PSO