In this study, a new methodology is presented to detect the sensor fault for piezoelectric array based on the filtered frequency response function (FRF) shapes. The proposed method does not require prior knowledge about healthy piezoelectric array. First, the imaginary parts of FRFs from the piezoelectric array during vibration are measured and normalized to obtain the FRF shapes in different frequencies. Then the irregularities in these FRF shapes are extracted by using high-pass filter with properly chosen cut-off frequency. These abnormal irregularities on the filtered FRF shape curves indicate the location of the faulty sensor, due to the irregularity of FRF shapes introduced by the faulty piezoelectric element. The proposed sensor fault method is experimentally demonstrated on a clamped-clamped steel beam mounted with piezoelectric buzzer array. Two common piezoelectric sensor fault types including sensor breakage and debonding are evaluated. The experimental results indicate that the proposed method has great potential in the detection of the sensor fault for piezoelectric array as it is simple and does not require the FRF data of the healthy sensor array as a baseline.

在这项研究中，提出了一种基于滤波后的频率响应函数（FRF）形状来检测压电阵列传感器故障的新方法。所提出的方法不需要关于健康的压电阵列的先验知识。首先，测量并归一化来自压电阵列的FRF的虚部，并将其归一化以获得不同频率下的FRF形状。Then the irregularities in these FRF shapes are extracted by using high-pass filter with properly chosen cut-off frequency. 由于由故障压电元件引入的FRF形状的不规则性，在过滤的FRF形状曲线上的这些异常不规则性指示了故障传感器的位置。提出的传感器故障方法在装有压电蜂鸣器阵列的夹钳钢梁上进行了实验验证。评估了两种常见的压电传感器故障类型，包括传感器损坏和脱胶。实验结果表明，该方法简便易行，不需要健康的传感器阵列的FRF数据作为基线，在压电阵列传感器故障的检测中具有很大的潜力。