CAJ | 학술논문

水轮发电机组在线监测与故障诊断中，信号特征提取是故障诊断前的关键步骤，在信号处理和模式识别之间起着承上启下的作用，构造的特征向量能否真实反映机组运行特性，直接影响到故障诊断结果的准确性。在水轮发电机组振动信号前处理基础上，结合信号自身特点与信息熵理论，实现固有模态函数能量熵特征提取，对机组当前运行状态做出判断，结果表明该方法简单有效，切实可用。
수륜발전궤조재선감측여고장진단중，신호특정제취시고장진단전적관건보취，재신호처리화모식식별지간기착승상계하적작용，구조적특정향량능부진실반영궤조운행특성，직접영향도고장진단결과적준학성。재수륜발전궤조진동신호전처리기출상，결합신호자신특점여신식적이론，실현고유모태함수능량적특정제취，대궤조당전운행상태주출판단，결과표명해방법간단유효，절실가용。
During online monitoring and failure diagnose of the turbine-generator unit, abstraction of the signal characteristics is one key step before failure is diagnosed.It connects to the signal handling and model identification.Whether the characteristic vector of structure can truly reflect the unit operation characteristics directly impacts the accuracy of the failure diagnosing results.Based on the pre-han-dling of the vibration of the Unit and in combination of the characteristics of the signal itself and information entropy theory, the energy entropy characteristics of the natural model function is abstracted.And the current operation statues of the Unit is judged.The practice shows that this method is simple, effective and practical .