SOM-MQE的钻井泥浆泵健康状态评估

Health Assessment of Drilling Mud Pump Based on SOM-MQE

为了对钻井泥浆泵的运行状态进行分析与评估,提出了一种基于SOM-MQE的钻井泥浆泵健康状态评估方法,并通过实测数据验证了算法的可行性。首先,采用加速度传感器采集泥浆泵的振动信号,应用相关性滤波算法对原始振动信号进行滤波降噪,提高信噪比。其次,对滤波后的数据提取时域、频域和时频域特征;再对这些特征PCA降维处理,得到反映泥浆泵状态的主成分特征矢量。最后,将主成分特征矢量作为SOM-MQE评估模型的输入,构建SOM-MQE评价模型,并计算最小量化误差(MQE)值大小,利用钻井泥浆泵偏离其在无故障状态下的MQE值对其健康状态进行评估。现场试验结果表明,MQE值表征的运行状态与钻井泥浆泵实际运行状态相对应,准确率高达96.28%,MQE值可以用来评价钻井泥浆泵的健康状态。

In order to analyze and evaluate the operation state of drilling mud pump,this paper proposes a method for evaluating the health status of drilling mud pump based on som-mqe,and verifies the feasibility of the algorithm through the measured data.Firstly,the acceleration sensor is used to collect the vibration signal of the mud pump,and the correlation filtering algorithm is used to filter the original vibration signal to improve the signal-to-noise ratio.Secondly,the time-domain,frequencydomain and time-frequency-domain features are extracted from the filtered data,and then PCA is used to reduce the dimension of these features to obtain the principal component vector reflecting the state of the mud pump.Finally,the principal component vector is used as the input of som-mqe evaluation model to build som-mqe evaluation model,and calculate the value of MQE,and use the deviation of drilling mud pump from its MQE value under no fault condition to evaluate its health status.The field test results show that the operation state represented by MQE value corresponds to the actual operation state of drilling mud pump,and the accuracy rate is as high as 96.28%.MQE value can be used to evaluate the health status of drilling mud pump.