Research On Mechanism And Identification Method Of Gas Turbine Blade Fracture Fault

Gas turbine is not only widely used in energy,electric power,pipeline transportation and other industrial fields,but also as the most representative new propulsion power equipment,it has become the core power source of the new generation of fighters and ship power systems,and has broad development and application prospects.Blade is the core component of gas turbine to realize energy conversion.Due to long-term exposure to alternating,large-scale centrifugal force,thermal stress and aerodynamic forces and other complex loads,it has become the sensitive part of the entire gas turbine with frequent faults,among which the main fault type is rotor blade fracture.Once the rotor blade fracture fault occurs,it will significantly affect the unit’s performance and efficiency.Suppose the fault early warning and diagnosis cannot be realized in the fracture transient or before the fracture.In that case,the high-energy broken blade will also damage the remaining blades and casing and even cause severe gas turbine’s damage accidents.In order to improve the reliability of the unit and reduce the occurrence of damage,it is essential to effectively identify and accurately locate the blade fracture faults.The thesis mainly carried out the following researches:(1)In view of the lack of effective dynamic description of gas turbine blades’ fracture fault,through studying the mechanism of rotor blade fracture,a set of dynamic analysis model of gas turbine rotor blade fracture fault is established based on multi degree of freedom lumped parameter model.(2)In order to solve the problems of insufficient engineering and poor sensitivity of the fault characteristic parameters of gas turbine blades obtained at the present stage,the vibration sensitive features of rotor blade fracture fault which are easy to monitor,identify and extract in engineering are extracted by analyzing and comparing the response characteristics of normal and fault blade systems.(3)Aiming at the problem that the current gas turbine online monitoring system is difficult to effectively identify and accurately locate the blade fracture fault,with the help of the convolution wavelet transform algorithm,a blade fracture fault identification method based on signal singularity detection is proposed.Combining the typical "stator blade-rotor blade-stator blade"arrangement of gas turbine compressors,a blade fracture fault location method based on the inverse calculation of the sensitive features response law is proposed.