Research On Control Strategies For Multi-Infeed HVDC Systems

This dissertation focuses on the application of the fast controllability of HVDC systems to realize emergency power support to its connected AC system in the presence of large disturbances occurred in AC system, and more emphasis is given to the control strategies concerned with multi-infeed HVDC systems. The main works are as follows:A new hybrid simulation model of HVDC system is established based on NETOMAC's hybrid simulation function in this paper. Using this model, simultaneous calculation of AC network in the programs stability section and calculation of DC network in the program's transient section can be made. In case of complicated DC circuit, any type of symmetrical AC system fault. DC line fault and DC line's recovery after fault can be more accurately simulated, and the DC line's dynamic response can be better represented. All the above is very important when doing the research on DC power support to AC system under emergency conditions, and has laid a solid foundation for the interaction study which will be met in a hybrid AC/DC power systems.Two important parameters. i.e. strong coupling critical admittance and weak coupling critical admittance that affect the occurring conditions and types of commutation failures in multi-infeed HVDC system, are defined in this paper. Factors which has great influence on the above two parameters, such as AC system strength. controller parameters of HVDC system, DC power transfer capacity as well as load characteristics, are analyzed in this paper. Some valuable conclusions are drawn from the simulation results, theoretical explanations are also given to the phenomenon occurred during the simulation process.In this paper, a coordinated recovery strategy for Multi-infeed HVDC system is proposed. A non-linear feed-forward loop is used to realize the coordinated control between the two controlled variables at the two control ends at first. Secondly, a constant AC voltage control is implemented for a short period after the AC fault removal. Finally, a staggered recovery strategy for different HVDC subsystems is also adopted, with the restart time interval of them optimized using the steepest descent gradient search method. The simulation results show that, with the proposed coordinated recovery strategy implemented, the recovery performance of the tested multi-infeed HVDC system has been improved a lot when compared with the conventional control strategy'.Based on differential geometric theory, a novel nonlinear modulation strategy for multi-IIIinfeed HVDC systems to enhance the transient stability of AC system in the presence of large disturbances is presented in this paper. Taking the HVDC systems as a variable admittance connected at the inverter or rectifier AC bus, the analytical description of the relationship between the variable admittance and active power flows of each generator can be derived. The traditional generator dynamic equations can thus be expressed with the variable admittance of HVDC systems as an additional state variable and changed to an affine form, which is suitable for the global linearization method being used to determine its control variables. An important feature of the proposed method is that, the modulated DC power is an non-linear and adaptive function of the AC system states. The design procedure is tested on a typical dual-infeed hybrid AC/DC system.This dissertation proposes a new approach to nonlinear active and reactive power modulation control design for multi-infeed HVDC systems, which realizes the emergency power support to AC system in the presence of large disturbances occurred in AC system. The Optimal-Aim Strategy (OAS) is used to decide the controller's input variables, i.e. the modulated active and reactive power which can be determined according to the practical operating conditions of AC system and operating states of DC system. The essence of the method is that, it is basically a nonlinear method and is adaptable to multi-machine hybrid AC/DC power systems.A two level...