Parameter Identification of Jiles-Atherton Model Based on Levy Whale Optimization Algorithm

The Jiles-Atherton model is key to researching the hysteresis loop. The focus of scholars across various countries has always been the parameter identification of the Jiles-Atherton model. This paper on the Levy whale optimization algorithm (LWOA), based on the whale optimization algorithm (WOA), proposes to overcome the disadvantage that WOA tends to involve the local optimum. The recommended algorithm uses the Levy flight strategy instead of the encircling prey policy since the former improves the global search. Therefore, the new algorithm is better at stability and calculation accuracy. To substantiate the efficacy of the proposed algorithm, it is tested against six benchmark functions and compared with the WOA, particle swarm optimization (PSO), grey wolf algorithm (GWO), and shuffled frog leaping algorithm (SFLA). In addition, the proposed algorithm is applied to realize two classical engineering problems, such as the tension/compression spring and welded beam design issues. The experimental findings reveal that the proposed algorithm is highly competitive with metaheuristic optimizers and improves the algorithm's performance. To address the poor stability of the J-A model parameter identification, an improved calculation method for parameter k and the reduced parameter ranges of the model parameters a and alpha were combined with LWOA. The proposed algorithm is called C-LWOA, which is compared with LWOA, PSO, GWO, SFLA, and the cuckoo search (CS) based on the data reported in the literature. Moreover, the simulation results demonstrate that the stability and calculation accuracy of the parameter identification by the C-LWOA was significantly strengthened. Equally important, the calculation error was within 0.2%. Finally, the proposed algorithm was subsequently used to fit the actual measurements of the hysteresis loop of permalloy.

Automated summary

The Jiles-Atherton model is crucial for studying hysteresis loops, and the parameter identification of this model has attracted great attention from scholars worldwide. In this paper, a new algorithm based on the Levy whale optimization algorithm (LWOA) is proposed to overcome the local optimum issue in the existing whale optimization algorithm (WOA). The new algorithm, named C-LWOA, demonstrates improved stability and calculation accuracy in parameter identification, as shown by experiments and simulations. The algorithm is also highly competitive with other metaheuristic optimizers in benchmark functions and engineering problems.