Alternative Title生産設備のスマート状態診断システムに関する研究 -回転機械状態診断のための知的振動信号処理方法-
Note (General)application/pdf
Rotating machinery is an important and indispensable engineering equipment in industries such as electric power, petrochemical, metallurgy, rail transit and marine ships. Once fault occurs, not only the rotating machinery itself is damaged, so serious that it led to economic losses, major accidents, and life-threatening. With the development of the industrial intelligence, the fault diagnosis of rotating machinery based on vibration signal is becoming more and more extensive application. However, due to the complication of rotary machinery, bad working environment, and variable operating conditions, the vibration signal acquired by the acceleration sensor has the characteristic of non-stationarity, non-linearity, and complexity. At the same time, affected by factors such as transmission loss, signal attenuation, and strong background noise, the regularity fault impact contained in the vibration signal is further weakened. The fault characteristic frequency in the spectrum is more difficult to extract, and it is more difficult to realize the accurate fault diagnosis of rotating machinery. Therefore, the research on effective vibration signal processing method for rotating machinery fault diagnosis has important engineering application significance. Aiming at the key problems that urgently need to be solved in the signal processing of rotating machinery fault diagnosis, such as suppressing background noise and enhancing fault feature information, the thesis carried out the research about signal fault impact enhancement, signal non-stationarity decomposition, signal adaptive filtering and signal image conversion. By analyzing the rotating machinery vibration signals under different working conditions, the characteristics of the signal are deeply studied, and the corresponding signal processing methods are proposed in a targeted manner. The specific research contents are as follows:(1) Aiming at the problems of strong background noise and submerged regular impact in vibration signals, a signal processing method based on weighted kurtosis variational modal decomposition (VMD) and improved frequency-weighted energy operator(IFWEO) is proposed. Firstly, the raw signal is decomposed by VMD, and the weighted kurtosis is employed to select the intrinsic mode function (IMF) optimally to reconstruct the signal. The reconstructed signal will carry abundant fault information. Secondly, a third-order cumulant method is introduced to improve the frequency-weighted energy operator (FWEO), which could strengthen the signal impulse and enhance the fault feature. The IFWEO could better effectively reduce the noise impact. Finally, the method is validated in low-speed bearing fault diagnosis.(2) Aiming at the non-stationary and non-linear of vibration signal, this chapter proposed a signal filtering and fault characteristic enhancement method based on reconstruction adaptive determinate stationary subspace filtering (Rad-SSF) and enhanced third-order spectrum to address the above-mentioned problems. In particular, Rad-SSF reconstructs an adaptive self-determined and decomposed vibration signal trajectory matrix to obtain the non-stationary signals. Thereafter, the filtered signal with the best fault characteristics is extracted according to the kurtosis. Meanwhile, a 1.5-dimensional third-order energy spectrum is performed to enhance the fault characteristics by strengthening the fundamental frequency and eliminating non-coupling harmonics. Finally, the method is validated in high-speed bearing fault diagnosis.(3) To solve the problem where the actual rotating frequency and its harmonics cannot be accurately extracted in engineering applications, an improved adaptive multi-band filtering method is designed. This method takes the theoretical rotating frequency as the search center, extracts the maximum within the positive and negative deviation as the actual rotating frequency, and sets a threshold according to the actual value to realize multi-band filtering. This method can effectively remove background noise and accurately extract the actual rotating frequency and its harmonics. This model can automatically extract the in-depth features of the filtered signal and improve the fault classification accuracy. Finally, the method is validated in rotating machinery abnormal structure fault diagnosis.(4) Aiming at the problem that the discrimination between fault categories is not obvious after one-dimensional vibration signal is converted to two-dimensional image, an incrementally accumulated holographic symmetrical dot pattern (SDP) characteristic fusion method is proposed in this chapter. The current study simultaneously extracts the time-and frequency-domain characteristic parameters of vibration signal based on the incremental accumulation method to avoid inconspicuous difference and small discrimination generated by a single parameter. Subsequently, the extracted characteristic signals are transformed into a 2D image based on the SDP method to enhance the differences between signals. Finally, the method is validated in rotating machinery bearing fault diagnosis.The vibration signal processing methods of rotating machinery proposed in this thesis have been verified by simulation experiments and engineering experiments, and the verification results prove that the proposed methods can realize effective and targeted signal processing.The main contribution of this thesis is to propose the corresponding signal processing method according to the unique characteristics of vibration signals under different operating conditions and the actual engineering application of rotating machinery fault diagnosis, which effectively suppresses the background noise, enhances the fault characteristic signal, and increases the discrimination between fault types.
本文/三重大学大学院 生物資源学研究科 共生環境学専攻 環境・生産科学講座 環境情報システム工学教育研究分野
97p
Collection (particular)国立国会図書館デジタルコレクション > デジタル化資料 > 博士論文
Date Accepted (W3CDTF)2023-03-02T11:18:10+09:00
Data Provider (Database)国立国会図書館 : 国立国会図書館デジタルコレクション