Discussion of article "Population optimization algorithms: Gravitational Search Algorithm (GSA)"

 

New article Population optimization algorithms: Gravitational Search Algorithm (GSA) has been published:

GSA is a population optimization algorithm inspired by inanimate nature. Thanks to Newton's law of gravity implemented in the algorithm, the high reliability of modeling the interaction of physical bodies allows us to observe the enchanting dance of planetary systems and galactic clusters. In this article, I will consider one of the most interesting and original optimization algorithms. The simulator of the space objects movement is provided as well.

The article presents an optimization algorithm based on Newton's law of gravitation: "Every particle in the universe attracts every other particle with a force that is directly proportional to the product of their masses and inversely proportional to the square of the distance between them". In the proposed algorithm, search agents are a set of masses that interact with each other based on Newtonian gravitation and the laws of motion. At the same time, all agents can exchange information with each other, wherever they are in the search space by means of an attraction force that depends on the mass (calculated from the values of the objective function) and the distance between them.

Agents are treated as objects, and their fitness is measured by their masses. In general terms (with the algorithm settings close to real physical laws), all these objects are attracted to each other by the force of gravity, and this force causes a global movement of all objects towards objects with a larger mass. Therefore, the masses interact using a direct form of connection through the gravitational force.

In the classical GSA, each particle has three kinds of masses:

a) active mass
b) passive mass
c) inertial mass

In most cases, it is convenient and expedient to use the equality of these concepts to simplify codes and calculations and increase the efficiency of the algorithm search capabilities. Therefore, there will be one mass in the algorithm, not three. The physical law equations used in the GSA are shown in Figure 1.


formulas

Figure 1. Force of gravity, acceleration and speed



The position of the particles provides the solution to the problem, while the fitness function is used to calculate the masses. The algorithm has two stages: exploration and exploitation. This algorithm uses intelligence capabilities at the beginning to avoid getting stuck in the local optimum, and after that it exploits the regions of extrema.

Author: Andrey Dik