基于信息熵与灰关联的抽油井泵示功图诊断技术研究

Study on Entropy-based Grey Correlation Fault Diagnosis of Oil Pumping Well Indicator Diagram

抽油井工作环境、工作过程比较复杂,对抽油井泵示功图故障诊断时由于有些抽油井典型故障的示功图几何形状非常相似,极易产生误判,因此引入基于信息加权熵的灰度关联故障诊断算法,利用信息熵加权值来体现不同故障特征因子所含的信息量,使计算所得与标准故障的关联值变大,与其他故障尤其是易误判故障的关联值变小。实验结果表明:该方法能够减少故障误诊断,使抽油井故障诊断更加精准。

The pumping well＇ s complex working environment and process and the similar geometrical shape of some typical fault indicator diagrams thereof can lead to wrong fault diagnosis, a entropy-based grey correlation fault diagnosis was introduced to make information entropy weights represent the amount of information that dif- ferent fault characteristics factor contain;the correlate result becomes bigger when contrasted with standard fault and it becomes smaller when contrasted with other faults especially the fault easy to misjudged. The experimen- tal results show that this method can reduce error rate of the fault diagnosis, and can make pumping well diag- nosis more accurate.