Sensitivity Analysis Of Output Characteristics To Structure Factors In The Generalized Hamiltonian System Of Hydropower Unit

In the generalized Hamiltonian system, structure matrix elements explain the interconnections among the system parameters and the damping characteristics of the ports, which provide large amounts of information about system dynamics. Full using this information to discuss the improvement of the system stability acting on the dynamic performances, it provides a new way to utilize the system structure and dynamics to design control law. However, in view of the form of the first order partial differential equations, there are not systemic theoretical supports to carry out the research of generalized Hamiltonian system structure. Based on sensitivity analysis method, this paper studied the influence of the matrix association factors on system output dynamic in the generalized Hamiltonian model of nonlinear hydraulic turbine generating unit. The results showed that the proposed calculation method is effective and provides a useful reference in the structure analysis of nonlinear dynamic system. And it provides a possible way to study the sensitivity analysis of multivariate interconnection.1. Contracting the whole unit system, in order to simulate the actual operation of the hydroturbine generating unit. The model includes three parts, the two order nonlinear hytro turbine Hamiltonian model with the rigit water hammer which is derived from the single pipe and single machine hydraulic system model and the IEEE Working Group series model, the third-order generator Hamiltonian model in single machine infinite system, the control system with the step signal in the exponential form.2. Based on the step response performance indicators of time domain analysis in the automatic control theory, the different indicators were proposed to analyze the step response of the hydroturbine generating unit to meet the requirements of stable operation. Variable movements of hydroturbine generating unit were divided into two fundamental motion models:underdamping attenuation vibration motion and overdamping motion. The cubic Hermit interpolation function was proven effective on computing the waveform envelope. The effective calculation method in computing the attenuation coefficient and oscillation frequency of transient response was studied.3. Based on the multi-parameters sensitivity analysis numerical method, this paper proposed a sensitivity analysis numerical method of the generalized Hamiltonian system structure, which took system matrix association factors as influencing factors. The concept of transient sensitivity and steady-state sensitivity is proposed. And the corresponding system characteristics sensitivity analysis system model is established. According to the differences of correlation and nonlinearity in composition, the association factors were divided into univariate, multivariate and zero interconnection. It was introduced to the application steps and calculation process description in analyzing Hamiltonian system structure.4. Applying sensitivity analysis numerical method of the generalized Hamiltonian system structure, the sensitivity of univariate and multivariate interconnections were analyzed. It was calculated and analyzed that the influence degree and law of these interconnections on the main output characteristics. It was determined that the maximum impact factor when the unit changed in the load.5. Taking the sensitivity calculation of zero interconnection in the generalized Hamiltonian model of nonlinear hydraulic turbine generating unit as an example. The sensitivity of zero interconnections was analyzed. It was calculated and analyzed that the influence degree and law of the zero interconnection on the main output characteristics in the transient and steady-state process of system. There was mutual restraint between off-diagonal correlations of the damping matrix and the correlations of the structure matrix in the generalized Hamiltonian system. The results showed that the proposed calculation method is effective and provides a useful reference in the structure analysis of nonlinear dynamic system.