A Hybrid Optimization Algorithm Based On BBO And ACO And Its Application In Filter Design

Efficient optimization algorithm technology has important practical significance and application value in solving complex engineering system optimization problems.However,the lack of accuracy of intelligent optimization algorithms and the tendency to fall into local optima make many engineering optimization problems have not yet been completely solved.Although scholars continue to propose new intelligent optimization algorithm theories.However,the classic single optimization algorithm is usually difficult to circumvent its inherent limitations.Therefore,how to efficiently use and combine multiple algorithms,maximize the strengths and avoid weaknesses,and make up for each other,so as to obtain a hybrid optimization algorithm with better performance and improvement,has become one of the important research directions in this field.First,a novel hybrid intelligent optimization algorithm: BBO-ACO is proposed,which combines the advantages of biogeography-based optimization(BBO)and ant colony optimization(ACO)and makes up for each others shortcomings.Specifically,it uses BBO to perform a global search,and then uses ACO to inherit the preliminary results for further local searches.Such a fusion scheme makes full use of the early rapid convergence ability of BBO and the stronger ability of ACO to get rid of the local optimal solution according to probability,and can make up for the shortcomings of BBO easily falling into the local optimal and the lack of initial information of ACO to converge slowly.Therefore,the hybrid algorithm has strong search capabilities and strong robustness.Next,these two classic algorithms were further improved,and a comprehensive performance stronger novel hybrid algorithm NBBO-NACO was obtained by fusing the new improved biogeography-based optimization(NBBO)and new improved ant colony optimization(NACO)algorithms.It still uses NBBO to perform a preliminary global search,and then uses NACO to perform a futher local search by the preliminary results,and finally obtains the global optimal solution.The optimization performance test of the benchmark functions proves that the hybrid algorithm has a strong search ability in space solutions,and a more powerful search ability and robustness.Finally,in order to further verify the effectiveness of the hybrid intelligent optimization algorithm in solving the design problem of 2-D IIR digital filter.The related results show that: compared with the existing various optimization algorithms,the hybrid optimization algorithm proposed in this paper has better computing efficiency and optimization ability in solving this problem.