Analysis of Output Characteristics of Giant Magnetostrictive Transducers Based on AFSA-eCS Hybrid Algorithm
The multi-field coupling of electricity,magnetism,and mechanics affects the working performance of giant magnetostrictive transducers.A multi-field coupling model is often used to analyze the nonlinear output characteristics,and its parameters determine the model's accuracy.The model has many key parameters with strong coupling,and the relationship between parameters in different working conditions is complex.Nowadays,artificial intelligence algorithms,such as particle swarm optimization and other single algorithms,are often combined to obtain parameters,which can easily lead to problems such as low identification accuracy,falling into local optima,and low stability.This paper constructs a comprehensive circuit model of the transducer considering the eddy current effect of bar cutting.An improved hybrid algorithm of artificial fish swarm and cuckoo search algorithm is used to identify the model.Firstly,the equivalent eddy current factor is calculated.An equivalent circuit for a magnetostrictive transducer's magnetic circuit,considering the non-uniform cutting of GMM rods,is constructed.Secondly,the classic JA hysteresis model is used to describe the magnetization process of the GMM rod,and the strain of the GMM rod is calculated using a stress-corrected secondary domain transition model to achieve the magneto-mechanical coupling process of the transducer.Then,a single-degree-of-freedom vibration model is used to describe the actual output displacement of the transducer radiation head.Finally,the artificial fish swarm algorithm with strong global searching ability and the cuckoo search algorithm with strong fine-grained searching ability are integrated.Thus,hybrid algorithms are introduced into parameter identification of transducer models to achieve fast,accurate,and stable extraction of model parameters.In addition,a prototype of a giant magnetostrictive transducer is produced,and an output displacement testing platform is built.The proposed algorithm has significant advantages in curve fitting accuracy,identification speed,and stability,with a curve fitting error of only 4.86%.Then,the measured displacement curves at different frequencies show that considering the eddy current factor in non-uniform cut bars is essential for effectively analyzing the output characteristics of transducers.The output displacement curves of the transducer are analyzed under different prestress and bias magnetic fields.The proposed method provides an excellent prediction of the experimental results,indicating that the method can effectively track the actual output characteristics of transducers under different working conditions.The following conclusions can be drawn.(1)In identifying transducer parameters,the proposed algorithm has higher identification accuracy,faster speed,and more robuststability than single algorithms and GA-SA hybrid algorithms.(2)Compared with the models with and without the eddy current factor,the importance of calculating the eddy current factor is demonstrated.(3)The output characteristics of transducers under different prestressed and biased magnetic field conditions can be effectively simulated using the proposed algorithm,and the optimal working point of transducers can be tracked,supporting the optimization design of transducers.
国家科技期刊平台
NSTL
万方数据
