An Improved Bat Algorithm And Its Application In Solving Supercapacitor Model Parameters

The meta heuristic algorithm is suitable for solving complex optimization problems and has become an indispensable part of the optimization field.The bat algorithm,as a new meta-heuristic optimization algorithm proposed by Professor Xin-She Yang,has been widely used by scholars at home and abroad because of its strong robustness,simple parameters,potential distribution and easy implementation and controllability.Like many meta-heuristic algorithms,the bat algorithm has problems such as easy to fall into local extremum,insufficient precision of optimization in the late stage of algorithm execution,and insufficient convergence speed.In the early stage of the algorithm iteration,because the moving speed of the previous generation bat individual is updated at a rate of 1 times,the early search speed of the bat is very slow,and it is impossible to quickly traverse the entire search space.In the later stage of the algorithm,the bat moves toward a position,and there is a phenomenon of population aggregation,which is easy to fall into a local trap.This is why the bat algorithm slows down at a later stage and is prone to fall into local extremum.Therefore,in order to significantly impore the above problems,this paper proposes a bat algorithm that dynamically adjusts the weight and frequency space for the basic bat algorithm,and discusses the effectiveness of the improved bat algorithm in the field of supercapacitor model parameter solving.(1)The bat algorithm that dynamically adjusts the weight and frequency space modifies the basic bat algorithm from two aspects.First,the concept of inertia weight is introduced in the bat algorithm to nonlinearly adjust the bat individual’s search speed.In the initial stage of the iteration,the inertia weight is taken to a larger value,and the local optimum is jumped out in time to perform global optimization;thesmaller value is taken in the later stage of the iteration to improve the local optimization precision.At the same time,the random uniform distribution factor is added to enhance the diversity of algorithm optimization,so that the algorithm has certain flexibility and adaptability.Secondly,the pulse frequency update strategy of Beta distribution is introduced.By comparing the Beta probability density curve and selecting the appropriate α and β parameters,the algorithm can obtain more detailed information in the bat iterative process and effectively improve the search accuracy.Through the above improved strategy,not only the search space of the population bat is expanded,but also the global search and the local search are taken into consideration,and the algorithm maintains stability and certain adaptability in the iterative process.The simulation results show that the improved bat algorithm shows excellent performance in both accuracy and convergence speed.(2)As a kind of high-power energy storage equipment,supercapacitor solves the problems of shortage of power system and serious environmental pollution to some extent.However,supercapacitors also exhibit problems such as low energy density,low cell voltage,large range of terminal voltage fluctuations,and the internal structure is susceptible to external environment.Therefore,To support power management systems in wireless sensor networks,it is necessary to update the model parameter values of the supercapacitor online to accurately capture its terminal behavior and improve system reliability.However,most parameter solving methods require a large number of experiments,and the accuracy of these methods is not very high.Therefore,in order to accurately predict the terminal behavior of the supercapacitor model and realize real-time power management,a method for solving the parameters of supercapacitor model based on DAWFBA algorithm is proposed.It can be seen from the experimental results that the improved model parameter solving method has higher feasibility and stability.Combining with the advantages and disadvantages of the bat algorithm,this paper proposes a bat algorithm that dynamically adjusts the weight and frequency space,and applies it to the parameter calculation of the supercapacitor model.The simulation effect is significant,which can accurately predict the behavior of thesupercapacitor terminal and improve the system reliability.Sexuality can be applied to the field of power management in wireless sensor networks.