Abstract:Aiming at the problem of gas turbine combustion chamber failure in gas-steam combined cycle power plant,the similarity analysis method is used,and the nonlinear state estimation technology is combined to conduct combustion chamber fault warning research.Firstly,the MSET model of the normal operating state of the combustion chamber is constructed based on the real-time data on the site.The acquired values to be observed are optimally estimated to obtain the estimated value.The deviation between the estimated value and the observed value can reflect the state of the internal combustion of the combustion chamber.Then the similarity is introduced to measure the similarity between the observed value and the estimated value,which is used to judge the operating state of the combustion chamber,and form an early warning message to remind the staff for maintenance.Finally,taking the carbon deposit fault of the gas turbine as an example,the effectiveness of the algorithm is verified.