CAJ | 학술논문

作为实际工业中一种不常见的过程对象，微分加纯滞后过程对象目前达不到良好的辨识效果。人们在辨识微分加纯滞后对象时，最常采用的是将输出数据进行累加来中和对象的微分特性，或者用作图找特殊点等方法，然而由于噪声的影响，这些辨识方法往往得不到满意的结果。提出了一种含微分环节过程对象模型频域辨识方法，能快速找到对象模型响应的主要频率段，从而对微分加纯滞后对象进行精确的模型辨识。仿真实验表明了该辨识方法能有效地对含微分环节过程对象进行辨识。
작위실제공업중일충불상견적과정대상，미분가순체후과정대상목전체불도량호적변식효과。인문재변식미분가순체후대상시，최상채용적시장수출수거진행루가래중화대상적미분특성，혹자용작도조특수점등방법，연이유우조성적영향，저사변식방법왕왕득불도만의적결과。제출료일충함미분배절과정대상모형빈역변식방법，능쾌속조도대상모형향응적주요빈솔단，종이대미분가순체후대상진행정학적모형변식。방진실험표명료해변식방법능유효지대함미분배절과정대상진행변식。
As an unusual process object in the actual industry, identification of differential link with pure hysteresis can’t reach good recognition results currently. The methods most commonly used to identify the objects are to accumulate the output data to offset the differential characteristics of the objects or to look for special points by mapping. However, those methods can’t have good recognition, because of the effect of noise. This paper presents a new technique of modeling identification for process with differential link in the frequency domain. It can find a main frequency band of the object quickly and an accurate model of the process with differential link. Simulation examples show that this method can reach good recognition results.