摘要
平面变压器的寄生参数是平面变压器结构设计以及电源电路设计的重要参考指标。由于寄生参数与平面变压器的结构、材料等存在非线性关系,寄生参数非常难以计算。针对现有的寄生参数预测算法以及寄生参数的特点,设计了一种基于离散二进制粒子群优化随机森林的平面变压器寄生参数预测模型。通过36组平面变压器数据对该模型进行训练和验证,结果显示此模型的预测精度符合实际需要。通过此模型与其他常用预测模型的比较,结果显示此模型的预测精度是最优的。以上结果表明了此模型的准确性和优越性,其预测结果可为平面变压器设计提供参考,具有重要的现实意义。
The parasitic parameters of the planar transformer are important guidelines for the design of the planar transformer and the power circuit.However,the parasitic parameters generally have a nonlinear relationship with other factors of the planar transformer such as the structure,material etc.According to the existing prediction methods and the characteristics of parasitic parameters,an estimation model of the parasitic parameters for planar transformer based on binary particle swarm optimization random forest is proposed in this paper.With 36 sets of planar transformer data to train and verify the model,the results explicit the prediction accuracy meets the actual needs.With comparison to the common prediction models,the results show it performs the optimum prediction accuracy.All results together confirm the accuracy and superiority of the model.And the parameters predicted by the model can provide guidelines for the design of planar transformers,so it has important practical significance.
作者
杨毅
徐晓冉
刘雅琳
胡琦钰
赵斌
王刚
Yang Yi;Xu Xiaoran;Liu Yalin;Hu Qiyu;Zhao Bin;Wang Gang(Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China;University of Chinese Academy of Science,Beijing 100049,China;University of Manchester,Manchester M139PL,United Kingdom)
出处
《国外电子测量技术》
北大核心
2021年第4期37-41,共5页
Foreign Electronic Measurement Technology
关键词
平面变压器
寄生参数
离散二进制粒子群优化算法
随机森林
planar transformer
parasitic parameter prediction
binary particle swarm optimization
random forest
