论文部分内容阅读
为了解决混频激励下含有非线性特征的电阻与电感(RL)串联电路参数的在线辨识问题,提出通过提取伏安曲线和广义磁链轨迹几何特征的参数辨识新方法。该方法采用最小二乘法定义伏安曲线和广义磁链轨迹复杂图形的倾斜度,由直线拟合直接辨识参数。从端口能量与功率的角度,讨论了算法的物理本质。实验结果表明:该算法具有辨识精度高、运算量小、便于递推等优点,适合于动态实时在线辨识,克服了基于Fourier变换的传统方法计算量大、结果依赖于信号频谱分布等问题。
In order to solve the problem of on-line identification of resistance and inductance (RL) series circuit parameters with non-linear characteristic under mixed excitation, a new parameter identification method by extracting volt-ampere curve and geometric characteristics of generalized flux locus is proposed. The method uses the least square method to define the slope of the voltammetric curve and the generalized flux locus complex pattern, and the parameters are directly identified by the straight line fitting. From the perspective of port energy and power, the physical nature of the algorithm is discussed. The experimental results show that the proposed algorithm has the advantages of high recognition accuracy, low computational complexity and easy recursion. It is suitable for dynamic on-line real-time identification and overcomes the problems of traditional methods such as large amount of computation and dependence on signal spectrum distribution.