Research On Fault Physical Characteristics And Full Vector Dynamic Balancing Of High Speed Rotors

Unbalance of rotors is one of the main exciting vibration sources of rotating machinery, and also the trigger factor of many self-excited vibration. In the past years, dynamic balancing technology has always been the focus in academia. With the constantly increasing capacity of rotating machinery, the increasing complexity of mechanical structure and the increasing rotating speed of rotors, the traditional theory and technology of dynamic balancing are difficult to meet with the demands of industrial production. It is mainly indicated in the two aspects: requiring many runs and test weights to gain balance will waste a lot of time and increase cost; in some given circumstances, the counterbalance calculated by the current dynamic balancing theory has large deviations, with low balance precision which will lead to balance failure.Under such background, the paper starting with the fault physical characteristics of unbalanced rotors, studies the laws of unbalance response of rotors, and analyzes specifically the features of unbalance response of unbalanced rotors with anisotropic bearing and nonlinear factor. In order to improve the efficiency and precision of field balancing, this paper based on the Full Vector Spectrum theory with information fusion and some related technologies, proposes three dynamic balancing approaches, and its efficiency and engineering value are tested by experiments. The main results and conclusions of the paper are as follows:1) Studying the features of fault mechanism. The traditional balancing methods are based on the linear hypothesis and isotropic rotors, while in fact, the anisotropic rotors and nonlinear factor are the main causes of low balancing precision. Therefore this paper studies the laws of unbalance response of rotors with anisotropic bearing and nonlinear factor. The results indicate that under the hypothesis of linear model, the unbalance response of the single-rotor system with anisotropic bearing, as for different unbalance amount, is a group of concentration ellipses with the same azimuth angle and the same eccentricity, its long and short half axes have linear change with the unbalance amount and its preliminary phase angle changes with the variation of unbalance amount; under the consideration of nonlinear stiffness and damping, even if the rotors are in steady state, its eccentricity and azimuth angle of the concentration ellipse of unbalance response both change with the unbalance amount.2) Proposing a new dynamic balancing method,the Full Vector Dynamic Balancing method(FVDB). The main concepts of FVDB are: regarding the main vibration vector as the unbalance response amplitude, the phase angle of ellipse as the phase of unbalance response, namely, the new unbalance response becomes the full vector unbalance response and then referring to the Influence Coefficient Method, the balance process will be fulfilled with the computer-aided balancing process.Theoretical deduction proves that unbalance response sensitivity of FVUR is higher then that of single channel, therefore, FVDB has higher precision. The experimental results show that: The Full Vector Dynamic Balancing method(FVDB) adopting the main vibration vector to evaluate the effects of dynamic balance, compared to the traditional methods, are more objective and realistic, which will remarkably improve the efficiency and precision of the dynamic balancing of rotors and has good engineering application value.3) Advancing a new balancing method- Precession Decomposition method(PDDB). Precession decomposition theory of rotors shows that forward precession component reflects the unbalancing state of rotors with its radius reflecting the amount of unbalance, its preliminary phase angle reflecting the phase of the unbalance amount, while the backward-precession components have no direct and necessary relations with the unbalancing rotors. This method adopts the Forward Precession Unbalance Response component( FPUR) to replace the traditional unbalance response. The results of numerical simulation indicate that under the circumstances of ignoring any nonlinear interference and measurement error, the results of the traditional method and the advanced one tend to consistent and compatible. The advanced method combines with the signals of two direction sensors, which overcomes the shortcoming of the balancing effect being influenced by the mounting direction of sensors in the traditional balancing method, and its results are more objective and effective which has great application value in engineering.4) Proposing the field balancing method of dual-rotor system with slight speed difference based on the spectrum correction with phase difference. With the analysis of structure of the dual-rotor system with slight speed difference and the “beat” signals of unbalance response, it indicates that due to the close rotating speed of the inner and outer rotors, the non-integral period sampling will lead to spectrum leakage, namely, at least one of the unbalance responses of the inner and outer rotors is distorted which is the main reason of poor balance effects. Based on this, it puts forward the extraction method of unbalance response components of inner and outer rotors based on the spectrum correction with phase difference, corrects the response amplitude and response of the distorted rotors and provides a balancing method of dual-rotors with slight speed difference based on phase difference correction. Through simulation and test, this method will remarkably improve the balancing precision of such rotors and the field balancing cases substantially demonstrate the superiority of this method.5) Developing a portable field dynamic balancing instrument which adopts the new balancing methods above. Based on the three new dynamic balancing methods above, adopting the embedded technology, the author develops a portable field dynamic balancing instrument. It discusses some crucial technologies such as the the vibration signal of full vector unbalance response and key phase collecting. The algorithm of full vector unbalance response, precession decomposition and extraction method of spectrum correction are provided which will be helpful to the engineering application of the new balancing methods. The engineering practice demonstrates the correctness of the research results and the practicability of the developed product.