Abstract: A performance assessment method based on kernel principal component analysis (KPCA) is proposed for a class of nonlinear systems that have actuators and process output nonlinear characteristics in practical processes. First,KPCA is used to estimate the noise sequence of the system. Second,the existing conditions of feedback invariants are discussed,and the lower bound of system minimum variance,which is the basis of performance of the system,is derived. Finally,the proposed method is compared with other evaluation methods. Simulation results show that the proposed method more accurately reflects the true performance of the class of nonlinear systems.