过程信号的微分计算在控制工程领域有广泛的应用，但经典微分运算方法存在显著的噪声干扰放大效应。文中通过分析一种经典微分器的变形结构以及对象阶跃激励响应和斜坡激励响应的正弦滤波信号成分的构成，指出在较高的滤波频率下，可通过信号的正弦滤波或余弦滤波提取这些过程信号的微分信号。提出了一种由正弦滤波器和新型正弦跟踪器所构成的新型微分信号提取方法。该方法有效提高了提取微分信号的质量，与理想微分信号相比的近似度较高。文中提出的微分方法具有良好的抗噪声干扰特性，是线性滤波技术的发展与延伸，并具有良好的理论意义和实际应用前景，可作为经典控制理论的有益补充。数学分析、仿真实验和实际应用结果进一步证实了文中所述方法的正确性和有效性。%Differential elements of the process signals are being widely applied in the field of control engineering, but the noise was significantly amplified by this traditional differential signal extraction method. By analyzing a deformation structure of traditional differential signal extractor and the composition of responses under step and slope incentives of the process signals, the differential signal can be extracted by sine or cosine filter signal under high frequency filtering conditions. Based on this result, a new differential extraction method composed of a sinusoid filter and a new sinusoid tracking coordinates was proposed. The quality of differential signal extracted by this method was improved dramatically with high proximity to ideal differential signal. Superior anti-noise characteristics could be obtained with the method proposed in this paper, which is the development and extension of linear filters, and has good theoretical significances and practical application prospects. The new signal processing method is a beneficial supplement to the classic control theory. Mathematical analysis, simulation experiment, and practical application results provide further evidence that the new method is correct and valid.
展开▼
