Artificial Physics Optimization Algorithm for Global Optimization

Abstract

This chapter provides a brief survey of the “artificial physics optimization” (APO) algorithm, a novel swarm intelligence algorithm based on physicomimetics for global optimization problems. Each particle (which is a solution to the optimization problem) in APO is treated as a physical individual that has mass, position and velocity. The mass of each individual corresponds to a user-defined function of the value of an objective function to be optimized. APO invokes a gravitational metaphor in which the force of gravity may be attractive or repulsive, the aggregate effect of which is to move individuals toward local and global optima. Theoretical analyses reveal the conditions under which APO is guaranteed to converge, and this allows us to reliably adjust parameters to improve APO’s diversity and convergence rate. An extended APO, the vector model of APO and local APO are also introduced. The implementations of APO and its improvements are applied to multidimensional numeric benchmark functions, and the results are analyzed. APO exhibits very good performance, suggesting that the algorithm merits further study.