A concept for the optimization of nonlinear functions using particle swarm methodology by James Kennedy & Russell Eberhart. This is the first paper talking about PSO. http://www.engr.iupui.edu/~shi/Coference/psopap4.html
PSO is used to adjust the network weights, with the Adaptive Neural Swarming method, the controller could adapt to environmental changes. It is tested in a real-world task of controlling a simulated non-linear bioreactor. http://www.chemeng.sun.ac.za/Content/gecco2002.pdf
In this paper the authors propose a method for adapting the particle swarm optimizer for dynamic environments. The process consists of causing each particle to reset its record of its best position as the environment changes, to avoid making direction and http://antho.huntingdon.edu/publications/Adapting_PSO_to_Dynamic_Env.pdf