摘要
开关角度的实时求解是特定谐波消除技术应用于工程实际的难点所在。本文提出了一种适用于微控制器的两电平逆变器消谐方程组实时求解算法。该算法分为离线转化和在线求解两个阶段。离线转化阶段,使用对称多项式和Groebner基理论将消谐方程组等价转化为一元高阶方程和线性方程组,并将此线性方程组存入微控制器;在线求解阶段根据输入调制比,求出线性方程组的数值解,并以此为系数构造一元高阶方程,使用Sturm定理隔离出一元高阶方程各个实根所在的区间,进而对每个区间使用弦截法求出所包含的精确实根。与传统数值法相比,此算法在STM32F407中的计算效率提高20倍。基于本文所提出的实时求解算法搭建了两电平逆变器实验平台进行验证,实验结果证明了此算法的有效性。
Solving the switching angles in real time is the key point to apply selective harmonic elimination(SHE)to engineering project.In this paper,a real time algorithm is presented to solve the SHE equations for two level inverters,which can be easily implemented on the micro-controllers and includes two computing stages,the offline computing stage and the online computing stage.In the offline stage,the SHE equations are transformed into an equivalent polynomial system which includes a set of univariate linear equations and a univariate high order equation by using symmetric polynomials and Groebner bases theory.Then,the univariate linear equations are stored into the micro-controllers and solved online according to the input modulation indices,and their solutions are used to construct a univariate high order polynomial equation whose real roots are actually the solutions for the original SHE equations.In order to find all the real roots of this constructed high order polynomial equation,the Sturm theorem is used to isolate every real root into non overlapping intervals and then the secant method is applied to every interval to find the exact real root,i.e.,the final solutions for the SHE equations.Compared to the traditional numerical method,the executing efficiency is increased 20 times on STM32F407.Finally,a two-level inverter test platform is built based on the proposed algorithm,and the experimental results show that the algorithm is effective.
作者
王晨旭
何飞
张奇
潘素娜
杨克虎
Wang Chenxu;He Fei;Zhang Qi;Pan Suna;Yang Kehu(School of Mechanical Electronic&Information Engineering,China University of Mining&Technology-Beijing,Beijing 100083,China;Aalborg University,Aalborg 9220,Denmark)
出处
《矿业科学学报》
2020年第4期435-443,共9页
Journal of Mining Science and Technology
基金
国家自然科学基金(61936008,61973307)
北京市自然科学基金(3182031)
越崎青年学者项目(2017QN19)
中央高校基本科研业务费专项资金(2009QJ12)。
