基于蛙跳模糊算法的Jiles Atherton铁心磁滞模型参数确定

发布时间：2018-02-13 10:27

本文关键词： Jiles Atherton模型蛙跳模糊算法爱泼斯坦方圈局部磁滞回线参数　出处：《电工技术学报》2017年04期 　论文类型：期刊论文

【摘要】：Jiles Atherton(JA)模型广泛应用于铁心磁滞建模领域,其参数确定精确与否直接影响铁心磁滞现象的表达,为此提出蛙跳模糊算法对JA模型参数进行辨识。为了防止蛙跳算法陷入局部最优解并加快收敛速度,根据特殊点(矫顽力、矫顽力点磁化率)处实测值与计算值的相对误差,采用模糊控制方法,得到动态反馈系数,修正蛙跳算法的步长调整公式。采用爱泼斯坦方圈搭建磁滞回线测量系统,测得铁心饱和磁滞回线与局部磁滞回线。采用蛙跳模糊算法、蛙跳算法、粒子群算法和遗传算法分别求解得到模型参数,并将计算的磁滞回线与实测磁滞回线进行比较,证明了所提出的蛙跳模糊算法不易陷入局部最优解且具有更快的收敛速度。根据蛙跳模糊算法得到的局部磁滞回线模型参数辨识结果,研究模型参数与磁通密度的关系并进行拟合,得到不同磁通密度下的局部磁滞回线参数拟合模型。
[Abstract]:The Jiles Atherton jar) model is widely used in the field of core hysteresis modeling. The accuracy of its parameters directly affects the expression of the hysteresis phenomenon. A leapfrog fuzzy algorithm is proposed to identify the parameters of JA model. In order to prevent the leapfrog algorithm from falling into a local optimal solution and to speed up the convergence rate, the relative error between the measured value and the calculated value at a special point (coercivity, coercivity) is obtained. By using fuzzy control method, the dynamic feedback coefficient is obtained, and the step size adjustment formula of the leapfrog algorithm is modified. The hysteresis loop measurement system is built by using the Epstein square loop. The saturated hysteresis loop and local hysteresis loop of iron core are measured. The model parameters are obtained by using leapfrog fuzzy algorithm, breaststroke algorithm, particle swarm optimization algorithm and genetic algorithm, and the calculated hysteresis loop is compared with the measured hysteresis loop. It is proved that the proposed leapfrog fuzzy algorithm is not easy to fall into the local optimal solution and has a faster convergence rate. According to the parameter identification results of the local hysteresis loop model obtained by the leapfrog fuzzy algorithm, The relationship between the model parameters and the flux density is studied and fitted, and the local hysteresis loop parameter fitting model under different flux density is obtained.
【作者单位】：山东大学电气工程学院;
【基金】：山东省科技发展计划资助项目(13ny15)
【分类号】：TM271

【相似文献】