Research On Fault Tolerant Control Method Of Gas Turbine Based On Variable Sampling

Gas turbine control system is an extremely important part in gas turbine system.With the development of gas turbine technology,the structure of the control system becomes more and more complex.Any problem may lead to partial system failure or overall system failure.Moreover,networked induced delays might occur in the data transmission process,which may cause network channel congestion or even system paralysis.At the same time,the gas turbine control system is vulnerable to the threat of cyber attacks due to the openness of the data transmission network,which seriously affects the performance of the gas turbine system.Based on the above considerations,the following research of this paper is shown as follows:(1)Based on the structure,working principle and thermal cycle process the gas turbine system,we establish a mathematical model for the simulation of the heavy-duty gas turbine system by combining the Rowen model and the actual gas turbine data.And the state space equation of the simplified gas turbine speed load system is obtained via parameter identification.(2)The problem of fault-tolerant sampled-data H∞ control for gas turbine speed load system with random time delays and actuator faults is investigated.Stochastic variables conforming with the Bernoulli distribution are considered to depict random time delays;By applying the Lyapunov-Krasovskii stability theory and the reciprocally convex combination lemma,the stability criterion of the stochastic time-varying time-delay system is established to guarantee that the designed controller can satisfy the stability of the system;Then obtain the gain of the state feedback controller according to the linear matrix inequality method.And the gas turbine speed load system simulation is used to prove the effectiveness of the presented controller.(3)The problem of transmission channel failures when the control signals are tampered by network attack is considered in this chapter.And random network attacks are introduced to rebuild the system model;The event trigger mechanism is introduced on the basis of the time sampling mechanism,and a time-event hybrid trigger mechanism is used to reduce network load and improve system performance;Moreover,the advanced over-approximate methods are applied to obtain sufficient conditions of the controller to satisfy the asymptotic stability of the system,and the linear matrix inequality method is used to obtain the controller parameters;And simulations are given to illustrate the effectiveness of the presented method.