CAJ | 학술논문

针对水电机组振动故障与征兆之间复杂的非线性关系,将经过整理的水电机组典型故障分别作为标准故障类,每个标准故障类和它所对应的具有代表性的特征参数构成故障类特征向量,建立标准故障特征参数矩阵.采用信息熵理论和Parks聚类分析方法对待检样本进行聚类分析,从而辨识出待检样本最有可能属于哪个故障类,即最有可能是哪种故障.通过实例检验表明理论计算与现场检查结果相符,证明该方法能有效地确定故障类型和发生故障的部位,适合于故障诊断中自动模式识别,具有良好的实际应用前景,为水电机组状态监测及故障诊断提供了一种新途径.
침대수전궤조진동고장여정조지간복잡적비선성관계,장경과정리적수전궤조전형고장분별작위표준고장류,매개표준고장류화타소대응적구유대표성적특정삼수구성고장류특정향량,건립표준고장특정삼수구진.채용신식적이론화Parks취류분석방법대대검양본진행취류분석,종이변식출대검양본최유가능속우나개고장류,즉최유가능시나충고장.통과실례검험표명이론계산여현장검사결과상부,증명해방법능유효지학정고장류형화발생고장적부위,괄합우고장진단중자동모식식별,구유량호적실제응용전경,위수전궤조상태감측급고장진단제공료일충신도경.
The faults of hydraulic generating units are classified into different clusters by the characteristic parameters using statistical analysis methods. A characteristic parameters matrix of standard faults is established. Clustering analysis has been used to identify the new sample. This clustering technique is based on information entropy theory and Parks clustering analysis, which has been tested in a real application. Results demonstrated that the proposed method is a good candidate to be used as an online diagnosis tool for hydraulic generating units.