Fireworks Algorithm With Differential Mutation Operator And Its Application In Parameter Identification Of Photovoltaic Model

The fireworks algorithm(FWA)was officially proposed by Professor Tan Ying in 2010.It is a swarm intelligent algorithm inspired by the spark of fireworks explosion.Compared with other intelligent algorithms,fireworks algorithm has outstanding performance in solving optimization problems,and it has a wide range of practices and applications in many fields.In the fireworks algorithm,each firework represents a feasible solution,and the collection of all fireworks is a population.The spark is generated by each firework explosion.The explosion amplitude and the number of explosion sparks are determined by the fireworks fitness value,and the Gaussian mutation operation is set to enhance the diversity of the population.Finally,the next generation is selected from the collection of fireworks and all sparks until the global optimal value is found or the maximum number of iterations is reached.This paper proposes a hybrid differential fireworks algorithm(DEFWA).The improvement of this algorithm mainly focuses on two aspects:First,the Gaussian mutation is replaced by the differential mutation operator.Two improved operators are proposed based on the traditional differential mutation operator.Compared with the traditional differential mutation operator,The performance of the improved operator is greatly improved,which can effectively avoid the disadvantage that the Gaussian variation makes the spark too close to the origin.Finally,a better one of the two improved operators is applied to DEFWA.Second,using a dynamic explosion spark strategy,this strategy is only applied to the best individual in each generation of population.If a new optimal value can appear in the generated explosion spark,the number of explosion sparks produced by the fireworks is increased,and vice versa.At the same time,this paper also discusses the optimal parameters setting about the number of sparks and the number of optimal explosion sparks in DEFWA.Experiments on the benchmark function set show that the overall performance is better than the traditional fireworks algorithm and several state-of-art fireworks algorithms.In addition,DEFWA is used to solve the parameter identification problem of photovoltaic model.The test is carried out on the single/dual diode model and two PV module models,which proves the excellent performance of DEFWA in solving practical problems.