Theoretical Analysis And Experimental Study On Field Dynamic Balancing Of Rigid Rotors

The fact that vibration data considered in conventional dynamic balancingmethods of rigid rotors are collected from single direction, and fieldmeasurement of maximum unbalanced vibration and first balancing precisionof field dynamic balancing system were considered. The primary object of thisdissertation focuses on further studies in field dynamic balancing theories ofrigid rotors, and corresponding development of rotor field dynamic balancinganalysis software based on theoretical analysis and experimental investigation.The work presented in this paper is divided into four major sections as follows.1. According to characteristics of shaft centerline orbit for unbalancedrotor vibration, a model calculating maximum and minimum vibration vectorsfor unbalanced rotors based on vibration signal processing methods, andrelevant validation experiments were given. The results of the experimentindicate that maximum unbalanced vibration vector can be measuredeffectively in site with this model. The model can be used to evaluate practicalbalancing effects by improving the formula calculating balancing efficiency.2. Relationship between trial weight errors and correction errors wasdemonstrated, and errors due to variation in trial weight vector were analyzedin hamming manner from perspectives of phase errors and amplitude errors ofunbalance. Influences imposed on measurement of vibration amplitudes due toerrors in sensor installation angles were studied as well. Principles in selectingprimary balancing trial parameters such as vibration measuring transducer,measurement planes, correction planes, and trial weights were summarized. Acriterion regarding scale in phase of trial weights was presented and a methodapplying indirect vibration signals to dynamic balancing calculating at balancespeed was given for field dynamic balancing of rigid rotors.3. With application of initial phase point vector as vibration measuringitem to influence coefficients calculating directly, method of initial phase pointinfluence coefficients was put forward based on investigations into existingmethods of two planes influence coefficients. The results of the trial weightsinfluencing experiment indicate that influence coefficient varies with differentvibration measuring items, and impact imposed on phase of initial phase pointinfluence coefficients due to variation in phase of trial weights is moreunobvious, compared with that of other three influence coefficients includinghorizontal vibration, vertical vibration and maximum vibration. The analyticalresults obtained from simulations in influence coefficients equations show thatimpact imposed on correction mass vectors due to variation in phase ofinfluence coefficients is evident, specifically on the phase of correction masses.The results of field first dynamic balancing experiments on both single rotorand dual rotors reveal that first balancing effects of conventional influencecoefficients methods depend on proper trial weight vectors to certain degree.Method of maximum vibration influence coefficients is an effective method fordynamic balancing, while method of initial phase point influence coefficientsis a relatively steady field balancing method for rigid rotors in laboratorycircumstances.4. Analysis software of rotor field dynamic balancing was developedbased on field dynamic balancing methods for rigid rotors advanced in thisdissertation, and integrated into Predictive Maintenance Studio 2000(PMS2000). Included in this analysis software are allowable vibration inquiryfunction and measured data preprocessing function, with which measured dataincluding rotational speeds and vibrations in single-start or multiple-start statecan be processed in sample mean format after eliminating abnormal measureddata. A number of experiments measuring rotational speeds and unbalancevibrations were performed to give preliminary quantizing estimation regardingreliability of data obtained from the whole measurement system.