Research On Thermo-economics Gas Path Fault Diagnosis For Marine Gas Turbine

When gas fault occurs in a marine gas turbine,the thermal efficiency of equipment may decrease slightly,while the efficiency of the equipment will decrease a lot.In order to solve the problem of difficult identification of gas path faults under low operating conditions by conventional fault diagnosis methods,a study on the thermal economic gas path fault diagnosis of marine gas turbines based on the second law of thermodynamics was carried out.In this paper,from the perspective of thermal economics,the thermal economic model of marine gas turbine gas path is established;the exergy loss analysis under normal state and typical gas path failure state is carried out;research on gas path fault diagnosis of marine gas turbine based on exergy loss and fault diagnosis based on endogenous irreversible loss of marine gas turbine component is carried out;And the fault diagnosis example verification is carried out for two diagnosis methods.The main research contents are as follows:(1)Thermal economic model of marine gas turbine gas path: The theory of thermoeconomic structure is introduced.The marine gas turbine is divided into subsystems and process components according to the appropriate degree of integration to determine the "fuel" and "product" of the component.The physical structure diagram and production structure diagram are established according to the working principle of the marine gas turbine.According to the production structure diagram,the mathematical model of the thermal economics of the marine gas turbine gas path is established,including the thermal model of the compressor,the combustion chamber,the turbine and the rotor model,and the exergy model of each energy flow.Finally,the thermal economic simulation model of marine gas turbine was established in MATLAB/SIMULINK environment,and the simulation of typical working conditions was carried out.The simulation results show that the maximum thermal parameter error is 7.38%.The established thermal model of the marine gas turbine gas path is in good agreement with the experimental data,which provides a basis for the subsequent simulation and diagnosis of gas path faults.(2)The fault diagnosis research of gas path failure on marine gas turbine based on thermal economy: According to the thermal economic simulation model of marine gas turbine gas path,the exergy damage of marine gas turbine is studied and the exergy loss matrix is established.The production relationship between components in the fault state is studied.The typical gas path failure factor of the marine gas turbine is determined.Typical gas path fault simulation of marine gas turbine is carried out.The variation law of component damage under the severity of marine gas turbine gas path is analyzed.The variation law of component exergy loss under different severity of marine gas turbine gas path fault is analyzed.It is found that the exergy loss of gas turbine components increases with the increase of gas path fault severity.(3)Fault diagnosis research of gas path failure of marine gas turbine based on exergy loss: According to the relative change of the exergy loss of the marine gas turbine in the typical gas path failure of the marine gas turbine,the correspondence between the data and the fault is established considering the influence of the noise and vibration of the sensor.By using Back Propagation(BP),Radial Basis Function(RBF),Probabilistic Neural Network(PNN),the relative change of exergy loss under fault and the fault matrix are trained to establish an intelligent gas path diagnosis model based on exergy loss.The effects of working conditions and data samples on BP-exergy loss path diagnosis network,RBF-exergy loss path diagnosis network and PNN-exergy loss path diagnosis network diagnostic performance were analyzed.It is found that all three have strong recognition for fault identification after learning.Among them,the PNN-exergy loss gas path diagnosis network has better diagnostic accuracy for the fault classification of this paper than the other two.At last,the real-time simulation experiment of marine gas turbine gas path fault diagnosis was carried out.The diagnostic results show that the intelligent gas path diagnosis algorithm based on scratch loss can achieve gas path diagnosis of gas turbine very well.The highest false alarm rate is 5.16% and the highest false alarm rate is 16.6% when switching between fault and normal state.(4)Research on fault diagnosis of marine gas turbine gas path based on endogenous irreversible loss: The variation law of endogenous irreversible loss of marine gas turbine under typical gas path faults under multi-boundary conditions is analyzed.It is found that when a gas turbine fault occurs in a marine gas turbine,the endogenous irreversible loss of the component in which the fault occurs is more stable under the multi-boundary conditions of the marine gas turbine.According to this law,the fault diagnosis criterion of marine gas turbine based on endogenous irreversible loss is established,and the fault diagnosis algorithm based on endogenous irreversible loss is designed.The simulation experiment of marine gas turbine gas path fault diagnosis under steady state and dynamic conditions is carried out.The diagnosis results show that the proposed gas path fault diagnosis algorithm can achieve gas path fault diagnosis,the maximum misdiagnosis rate is 5.12%,and the maximum missed diagnosis rate is 14.67%.It was found through experiments that the misdiagnosis mainly occurs when the gas turbine switches from normal to fault state.Missed diagnosis occurs when the fault returns to the normal state.