论文部分内容阅读
积分环节是信号处理领域的关键环节,广泛应用于电力系统测量及控制领域之中。目前对于数字积分算法的研究普遍基于传统牛顿科茨算法,低阶科茨公式高频响应普遍较差,高阶科茨公式所引入的传输函数设计则过于复杂。在研究复合科茨传输函数与理想积分误差的基础上,利用龙贝格算法将采样频率加倍前后的积分输出信号做线性组合处理,在算法阶数相同的情况下,提高了数字积分器的精度,降低了设计难度。仿真和试验测试证明,所设计算法准确度可以达到0.01%以内,具有优良的稳态和暂态性能。
Integration is a key part of the signal processing field, widely used in the field of power system measurement and control. At present, the research on digital integral algorithm is generally based on the traditional Newton-Cotz algorithm. The low-frequency Coates formula is generally poor at high frequency response, and the transfer function introduced by high-order Coates formula is too complicated. Based on the study of the complex Coztra transfer function and the ideal integral error, the integral output signals before and after doubling the sampling frequency are linearly combined by using the Ringerberg algorithm. Under the same algorithm order, the accuracy of the digital integrator is improved , Reducing the design difficulty. Simulation and experimental tests show that the accuracy of the proposed algorithm can reach within 0.01%, with excellent steady-state and transient performance.