The real-time implementation of envelope analysis for bearing fault diagnosis based on wireless sensor network

Wireless sensor network (WSN) is gaining popularity in condition monitoring (CM) fields. However, enormous datasets obtained by data acquisition systems with high sampling rate cannot be transmitted effectively via a WSN due to its limited transmission speed. This may lead to the loss of significant information for condition monitoring and fault diagnosis, hence affecting the accuracy of diagnostic results. Local processing can be employed to improve the transmission efficiency; in that the acquired data is processed to extract the features locally for data compaction, and then only the analysis result is sent through the WSN. Therefore, the transmission load is reduced significantly and all the useful information is guaranteed to be transmitted for examining accurate detection results. However, the commercially available wireless sensors are usually not competent enough to fulfill complicated signal processing algorithms. In this paper, an intelligent WSN node capable of signal processing is proposed to improve the transmission efficiency of the WSN system and the envelope analysis algorithm is implemented on the sensor for bearing fault diagnostics. A bearing test rig is set up to verify the performance of this design. According to the test results, if a band-pass filter with 1000 Hz bandwidth is applied in the envelope analysis process, the data to be transmitted could be reduced by nearly 95%, which will make the real-time transmission effectively.