Study On Vibration Fault Feature Analysis And Target Controlling Of The Vibration Existing In The Rotor-Shaft System Of Turbomachinery

Turbomachinery, such as ultra supercritical turbo-set, turbocompressor, gas turbine and so on, is the key equipment in electric power industry, metallurgical industry, petrochemical industry and aviation industry that supporting the national economic lifeline. No turbomachinery operation is vibration-free due to various excitations which can cause fault even disasters in practical industrial application. In long-term operation of turbomachinery, turbine machinery may appear all kinds of rotor vibration fault. If not timely and effective suppression of rotor vibration for the turbine machinery, the rotor vibration condition will deteriorate further, eventually lead to accident. And this cause huge economic losses to the enterprise. So the suppression of rotor vibration for turbine machinery operation is very important for safety running of the process industry for a long period in which fault-free operation is demanded. For example, the ultra supercritical unit with N+1 configuration is easy to install and maintain, and the investment of ultra supercritical power plant can significantly be declined. The configuration where n rotors are supported by n+1 bearings is known as N+1 configuration. However, due to the specific supporting structure of N+1 configuration, rotor vibration is more greatly affected by adjacent rotor in the ultra supercritical turbo-set with N+1 configuration than with 2N configuration. It is very difficult to analyze and diagnose the vibration fault due to the decrease in the numbers of vibration monitoring locations for the ultra supercritical turbo-set with N+1 configuration. In addition, unstable rotor vibration is caused easily in N+1 configuration.The research in this dissertation is to explore the mechanism of shaft imbalance and fault caused by bearing supporting stiffness decrease for the ultra supercritical turbo-set with N+1 configuration. A target controlling method to control rotor’s multi-frequency periodic vibration in rotor-bearings system is proposed which uses active magnetic exciter (AME) to produce active control force to suppress rotor’s vibration. The research results of this dissertation are useful for the fault diagnosis and the vibration suppression of rotor in turbine machinery. The outline of this dissertation is as follows:(1) The theory and method of the dynamics system modeling for the turbine machinery is introduced. And the state-of-the-art of rotor dynamics, vibration fault mechanism and active vibration control is reviewed in this thesis.(2) A finite element model is developed to study the dynamic behavior of two rotor-three bearing system. The influences from rotational speed, eccentric condition and stiffness of coupling on the dynamic behavior of N+1 configuration and the propagation of motion can be numerically simulated in detail. A linear rotordynamic analysis for the finite element model is carried out to get the evaluations of each rotor critical speed and unbalance response. The results show that the critical speed and unbalance response of rotors are sensitive to coupling stiffness in N+1 configuration. Through the use of a commercial software, the finite element model of the ultra supercritical turbo-set with N+1 configuration is developed to analyze the dynamic characteristics of the shaft, rotor imbalance and the fault caused by bearing supporting stiffness decrease for the ultra supercritical turbo-set with N+1 configuration. The analysis results show that the imbalance of a rotor has a great influence on the vibration of the adjacent rotors, and that the fall of the bearing support stiffness leads to the increase of rotor vibration accordingly in the ultra supercritical turbo-set with N+1 configuration.(3) Draw lessons from gene targeting technology of modern medicine, the principle of target controlling for turbine machinery is expounded. The intrinsic relationship between external disturbance excitation, corresponding response and targeted inhibition is explored. From the energy flow, information flow and mass flow, the vibration faults targets of the turbine machinery can be described now.(4) The theoretical and experimental research on the characteristic of electromagnetic force carried out in AME. In the case of considering the magnetic field loss and geometry deformation, a electromagnetic force model is developed to analyze the characteristics of electromagnetic for AME. And another electromagnetic force model based on rotor vibration displacement is also proposed by the magnetic circuit theory. The parameters of proposed models for the experimental rotor-bearing system are obtained using the least-squares algorithm. The results of the experiments indicate that the proposed models have a better mechanics characteristic performance.(5) By the principle of vibration target controlling for turbine machinery, a research on active vibration suppression using AMEs in the rotor-bearing system is carried out. A rapid, real-time online approach is proposed for suppression of rotor unbalance vibration in rotor-bearings system with AMEs through the proportional-derivative (PD) control strategy. The PD feedback strategy is used to produce the electromagnetic force acting on the rotor. And additional stiffness and damping is applied in the rotor-bearings system for changing the stiffness and damping of the system. The reduction of rotor unbalance vibration is achieved via the stiffness and damping adjustment. A method to control rotor’s multi-frequency periodic vibration in rotor-bearings system is proposed which uses AMEs to produce active control force to suppress rotor’s vibration and to reach a self-optimizing control of the rotor vibration. The control strategies include an arithmetic to optimize the amplitudes and phases of the control current using on-line self-optimizing algorithms in AME and applied multiple frequency-matched control force that AME generates to reduce the measured amplitudes of rotor. An active vibration control scheme for controlling transverse vibration of the rotor due to multi-frequency excitation is designed. The whole circle search (WCS) algorithm and fast optimizing search (FOS) algorithm are proposed to optimize the amplitude and phase of controlling current. The experiments for controlling rotor multi-frequency vibration and unbalance vibration are carried out on the rotor-bearings test rig. The experimental results indicate that the proposed method can effectively suppress the rotor multi-frequency vibration and unbalance vibration using AMEs.