Particle Swarm Optimization

Particle swarm optimization (PSO) is a method of artificial intelligence. It was originally introduced in 1995 by Kennedy and Eberhart [1] as a tool for the optimization of nonlinear functions. The algorithm tries to simulate the behavior of animals that cooperate in groups to search for food. According to this property, the algorithm can be classified as swarm intelligence.

However, PSO is not only based on social interaction. Shi and Eberhart mentioned in [2] that the main decision equation of the algorithm has three basic parts. The second part influences the decision regarding the best personal position of the particle in the design space. This part is also referred to as the "cognitive part" and represents a particle's own thoughts. That is why PSO can also be described as a method of artificial intelligence. PSO tries to reach a goal ("find the minimum") by using its own thinking and considering the environment that corresponds to the definition in [3].

Usage in Program

The Optimization & Costs/CO₂ Emission Estimation uses the particle swarm optimization to find the optimal assignment of global parameters.

Dialog Box "Optimization Settings"

References

[1] Kennedy, J., & Eberhart, R. (1995). Particle swarm optimization. Proceedings of ICNN’95 - International Conference on Neural Networks, 4, 1942–1948. https://doi.org/10.1109/icnn.1995.488968

[2] Shi, Y., & Eberhart, R. (1998). A modified particle swarm optimizer. 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No. 98TH8360), 69–73. https://doi.org/10.1109/icec.1998.699146