Nonlinearity Identification of Main Steam Control Valves in Thermal Power Plants Subject to Dynamic Behavior Variations

Nonlinear characteristics of main steam control valves play an important role in the frequency control performance of thermal power plants. Manual testing has traditionally been relied upon to evaluate the nonlinear characteristics, but this method interferes with the normal operation of thermal power plants. This paper proposes a novel method to identify nonlinear characteristics of main steam control valves subject to dynamic behavior variations by special data segments hidden in historical data. The special data segments solely contain static information, separating static nonlinear characteristics from dynamic behavior variations. Compared with the Hammerstein model identification method, the proposed method can more accurately establish a model for the nonlinear characteristics of main steam control valves. Industrial case studies demonstrate the effectiveness of the proposed method.