CAJ | 학술논문

Hilbert-Huang 变换（Hilbert-Huang transform ，HHT ）这种完全由数据驱动的自适应时频数据分析算法已经在多个领域得到了广泛推广和应用。但是 HHT 的核心部分经验模态分解（empirical mode decomposition ，EMD）被端点效应问题所困扰。这里结合局部多项式回归，利用局部少量数据在极值序列两端各拓展一个极值点，从而抑制 EMD 的端点效应。模拟实验结果表明，该算法能够有效抑制端点效应，且数据要求不高，运行速度较快。
Hilbert-Huang 변환（Hilbert-Huang transform ，HHT ）저충완전유수거구동적자괄응시빈수거분석산법이경재다개영역득도료엄범추엄화응용。단시 HHT 적핵심부분경험모태분해（empirical mode decomposition ，EMD）피단점효응문제소곤우。저리결합국부다항식회귀，이용국부소량수거재겁치서렬량단각탁전일개겁치점，종이억제 EMD 적단점효응。모의실험결과표명，해산법능구유효억제단점효응，차수거요구불고，운행속도교쾌。
Hilbert-Huang transform (HHT ) is a completely data-driven adaptive algorithm for analyzing time-frequency data .HHT has been widely used in many areas .However ,as a core part of the HHT , empirical mode decomposition (EMD ) is still negatively affected by end effect . Based on the local polynomial regression (LPR) ,one extreme point was added to each of extreme series by few local data to suppress the end effect of EMD .The simulation results show that the LPR-EMD is a practical algorithm that can effectively restrain the end effect and have less constrain on the data and run faster .