| Small hydroelectric power plant and wind farm are typical forms of Distributed Generation (DG), have developed rapidly in the recent years. With the increasing penetration of DG, the topology configuration of distributed networks (DN) is changed, as well as the distribution of fault current and voltage. More than one generator directly connects to a common bus in a small hydroelectric power plant, which is defined as a multi-generator-system (MGS). Hydroelectric generators are often subjected to grounding faults in their stators. It is difficult to detect the fault location and the fault generator in the MGS by traditional stator grounding protection schemes. Wind farms generally lie in the terminal of power grid. The active power output of wind turbine generators varies along with the wind whose speed changes randomly. The reactive power absorbed by power equipments and cables is also varied, So that the reactive power flow and voltage stability of the DG system will be influenced.The schemes of grounding protection and reactive compensation for DG are studied in this thesis. In order to improve the grounding protection of DN with DG, a novel grounding resistance based single-phase grounding protection scheme is proposed. The grounding resistance is calculated by the faulted phase voltage and the fault component of phase currents difference. EMTP simulation results show that the protection scheme can detect the faulted feeder with high precision, and without influenced by neutral point grounding methods, the capacity and location of DG. In order to detect the generator with stator winding single-phase grounding fault in MGS, two protection schemes with the fundamental and the third harmonic fault components of the stator winding leakage current are proposed respectively. EMTP simulation results show that the protection scheme can detect the generator with grounding faults in the total (100%) stator winding and the condition of all kinds neutral grounding methods with high sensitivity and selectivity. In order to improve the reactive power-voltage stability of wind farms, a multi-objective optimal reactive power compensation model and a power flow calculation method are proposed. Modal voltage analysis is used to determine the candidate installation locations of the reactive compensation devices, and the improved genetic algorithm (GA) is applied to solve the reactive power compensation model and cuaculate the best compensation capacity. MATLAB simulation results show that the proposed scheme is with rationality and validity. The reactive power compensation techniques reduce power loss and improve voltage stability with rational investment. |