Trend analysis and on-line identification of state parameters of hydraulic generators

Trend analysis and computerized expression of state parameters are two important factors in the on-line fault diagnosis. Singular spectrum theory is used in this paper to analyze the trend of state parameters. While the artificial neural network (ANN) is applied in the on-line identification of typical trends of state parameters. This method can identify the state parameters effectively, such as vibration, temperature, pressure and so on. So it could be an effective assistant analytical method for state analysis, state assessment and prediction of hydraulic generators as well as could be a powerful reference of state maintenance for hydroelectric plant.