Research On Fault Diagnosis Method Of Helicopter Swashplate Rolling Bearing

The rolling bearing of helicopter swashplate is one of the important components in helicopter, which often works under tremendous loads and is prone to failure. Once a failure occurs, it will bring significant security risks in the helicopter flight if not timely detection and treatment. Thus the researching of the fault diagnosis approach for the helicopter swashplate rolling bearing is practical significant to protect the flight satety of the helicopter. In this paper, the in-depth research is conducted on the fault diagnosis method for swashplate rolling bearing by vibration analysis method use the data collected from a certain type swashplate rolling bearing fault simulation test which carried out by a bear factory. In addition, the research system of fault diagnosis for swashplate rolling bearing developed by research group has been upgraded. The main research contents and results are as follows:(1) A rolling bearing fault feature extraction method based on frequency bands energy feature is proposed. Compared with ordinary rolling bearing, the swashplate rolling bearing has low speed, large pitch diameter and number of balls. Commonly used feature extraction method is not effective. So the frequency bands energy feature extraction method is proposed. Firstly, FFT is used to transform vibration signals, and calculate the single side power spectrum of signals; Secondly, according that the energy distribution of the four kinds of vibration signals single side power spectrum of normal, inner ring, outer ring and ball fault are different, we evenly divided the single side power spectrum into M segment; Finally, achieve the fault feature through calculate the energy of each segment and normalized. And the range of M is determined through theoretical analysis. The experiment of M takes values in the range show that this method is effective.(2) A diagnosis method of swashplate rolling bearing failure based on the combination of frequency bands energy feature and support vector machine is proposed. Firstly, frequency bands energy feature of the vibration signal are extracted; Then treat the normalized frequency bands energy feature as a feature vector input into the support vector machine, so as to achieve the goal of the fault diagnosis. A large number of fault diagnosis experiments with the signals collected from different cut and different load are carried out and the influence of M on the diagnosis results are analyzed. The results show that the method is effective and adaptable to different cut and different load.(3) The research system of fault diagnosis for swashplate rolling bearing developed by research group has been upgraded. In this paper, the system is optimized and improved from three aspects: system window, interface design and system function. Firstly, in order to make the system more friendly, the jump window of open experiment and related operations are added to system; Secondly, the interface of the feature extraction method is redesigned, which makes the system more conform to the requirements of the modular design, and improves the code reuse rate; Finally, in order to make the system more perfect, the proposed method is added into the system.