论文部分内容阅读
Through the study on the output signals of the electro-hydraulic regulating system in the thermal power plant, a novel method for online diagnosis of the plugging fault in the servo valve is presented. With the use of the AMESIM software, the changes of the piston displacement, the oil pressure, the magnitude attenuation and the phase lag of the system under different plugging states are studied after simulation. Besides, the influences of the symmetrical and unsymmetrical plugging on the system are also compared and the characteristic table is given. The duo-neural network is put forward to achieve an online diagnosis on the plugging fault of the servo valve. The first level of network helps to make the qualitative diagnosis of the plugging position while the second level is for the quantitative diagnosis of the degree of the plugged position. The research results show that plugging at different positions exerts different influences on the performance of the system. The unsymmetrical plugging mainly affects the regulation time while the symmetrical plugging leads to great changes in the magnitude attenuation and the phase lag.