Research On The Digital Twin-Based Non-Stop Diagnosis Method For The Flue Gas Turbine Rotor System

In the petrochemical industry,flue gas turbine is the most important energy recovery device.The complex composition of flue gas makes the rotor system of flue gas turbine affected by factors such as high temperature,high pressure and erosion for a long time.Under such severe operating conditions,the rotor system of the flue gas turbine often has rotor imbalance failures,which have extremely adverse effects on the operation of the catalytic device.This thesis takes the flue gas turbine rotor system as the research object.Aiming at the complex operating conditions of the flue gas turbine and the difficulty in obtaining valid data,a multi-physics coupling model of the flue gas turbine rotor system based on digital twins is constructed to realize the operating conditions of the flue gas turbine.The multiphysics simulation of the flue gas rotor system is obtained and the response characteristics of the flue gas rotor system are obtained.Aiming at the problem of traceability and quantitative diagnosis of unbalanced faults in hoods,an innovative method of identifying and tracing unbalanced faults in rotor systems based on digital twins and quantitative analysis of unbalanced faults are proposed.The specific research contents are as follows:(1)The multi-physics coupling model of the rotor system of the smoke generator based on the digital twin is constructed.In view of the complex operating conditions of the flue gas turbine and the difficulty in obtaining valid data,the digital twin theory is introduced,and a method for building a flue gas turbine rotor system model based on the fluid-thermal-solid multiphysics coupling is proposed based on numerical simulation.Combined with the rotor imbalance identification of rotating machinery Theory,constructs a multi-dimensional model including the three-dimensional model,multi-physics model,and theoretical model of the rotor system of the smoke machine,and realizes the analysis of the response characteristics of the rotor system under the unbalanced state.The multi-physics coupling model of the hood rotor system based on digital twin considers the flow field,thermal field,structural field,modal analysis and vibration excitation factors,and can comprehensively analyze the transient unbalance of the rotor system under complex conditions such as high temperature and high speed.response situation.Compared with the traditional single-physics simulation model,the multi-physics coupling model of the hood rotor system based on the digital twin considers various physical fields and unbalanced response situations,and has the advantages of high model construction reduction and comprehensive analysis;(2)An unbalanced fault tracing method of rotor system based on digital twin is proposed.The existing unbalance diagnosis method based on time-frequency domain signal analysis has the problem that it is difficult to trace the rotor unbalance.Therefore,the correlation analysis method and the similarity calculation method are introduced to study the cause and mechanism of the unbalance fault of the smoke generator.The data characterization of each failure cause is analyzed.Combined with the fault simulation data of the digital twin model and the actual fault data in the field,the unbalanced fault database of the rotor of the smoke generator is constructed,and the traceability method of the unbalanced fault based on the correlation analysis method and the similarity calculation method is established,and the traceability of the unbalanced fault of the rotor of the smoke generator is realized.Taking the measurement point data of the flue gas turbine obtained on-site in a petrochemical production enterprise as an example,the effectiveness of the traceability method is verified.Compared with the traditional fault identification method,this method can accurately trace the cause of the imbalance and improve the equipment.Efficiency of fault diagnosis work;(3)A method for estimating the unbalance of the rotor system without stopping based on the digital twin is proposed.Aiming at the problem that the existing methods based on time-frequency domain signal analysis are difficult to quantitatively diagnose rotor unbalance,a description method of rotor system unbalance load is introduced,and the rotor unbalance fault is expressed as a parameter by load.At the same time,the twin model is used to simulate different situations of unbalanced faults,and the results obtained are compared with the actual fault load.Finally,the optimization algorithm is used to find the fault parameters that are most in line with the actual situation,and the flue gas rotor system is not stopped to estimate the unevenness.measure.Through the unbalance test of the rotor system with double discs,the effectiveness of the method is verified,and the accurate estimation of the unbalance amount is realized,and the recognition accuracy reaches 95%.Compared with the traditional diagnosis method,the method can quantify the degree of unbalance,and realize the improvement of the quantitative analysis ability of rotor unbalance fault under the condition of non-stop.