Research On Equilibrium Optimizer And Its Application

Equilibrium optimizer algorithm(EO)is a new physics-based heuristic optimization algorithm,which is inspired by the control volume mass balance to estimate the dynamic and equilibrium states.In EO,the search agent randomly updates its concentration(position)in order to find some genius particles which are called equilibrium candidates,and finally reaches the equilibrium state as the optimal result.EO algorithm has the advantages of simple structure,easy implementation and high robustness,but the algorithm also has the disadvantages of slow convergence speed,low convergence precision and easy to fall into local optimal,which greatly limits the application scope of equilibrium optimizer.In order to make the equilibrium optimizer solve the complex optimization problem effectively,this thesis adopts the coding method and introduces the multi-strategy to improve the equilibrium optimizer,and applies the improved algorithm to solve the complex optimization problem,aiming to further improve the overall performance of EO algorithm and broaden its application range.The main contents of this thesis include:(1)In this thesis,binary encoding is used to transform the continuous equilibrium optimizer into discrete equilibrium optimizer,and two binary equilibrium optimizers(BEO)are proposed to broaden the application range of BEO.In order to test the performance of the binary equilibrium optimizer,it is used to optimize the feature selection problem.Simulation experiments on 19low-dimensional datasets and 13 high-dimensional datasets show that the proposed BEO-V2 method has the best performance,and the effectiveness of BEO-V2 method is shown by comparing with other well-known meta-heuristic methods.(2)The random attraction strategy,the individual optimal neighborhood learning strategy and the dynamic opposition learning strategy are used to increase the exploration and exploitation ability of the algorithm through the way of population reconfiguration,and the multi-strategy equilibrium optimizer(MSEO)is proposed.Compared with other algorithms in 23 benchmark functions and two engineering examples,the experimental results show that the MSEO algorithm is better.(3)In order to broaden the application scope of EO algorithm,EO algorithm is applied to solve clustering analysis problems.The experimental results on 10 datasets show that EO algorithm can effectively solve the clustering analysis problem and increase the application scope of EO algorithm.