| Under the promotion of smart grid, self-healing grid, resilient grid and other new concepts, the modern power system is developing towards the interaction, self-healing, high security and high reliability. Power system restoration has become one of the major concerns of modern power system. With the construction of UHV AC/DC hybrid power grid and the higher penetration of renewable energy, the increasing complexity of power grid and improved voltage level has brought some new challenges to power system restoration. Since modern society depends on the energy supply excessively, the widespread blackout has a serious negative effect to the social economy, and the proper active reactive control scheme is necessary for the fast and safe restoration of power system. Based on the existing literature survey about power system restoration, a detail study on active and reactive power optimal control has been carried out in this dissertation, including sustained overvoltage control during early stage of network reconfiguration, load restoration optimization and unit output optimization during last stage of network reconfiguration. The main contributions and innovations are described as following:(1) In order to overcome the limitations of the sustained overvoltage static control strategy, a dynamic multi-objective sustained overvoltage control model is proposed. By using restoration sequence as optimization period, and restoration operation as segmentation flag, the whole process is taken into consideration in the model. According to the demand of restoration, the object functions adapted to restoration are defined and the risk of restoration sequence, operation time of voltage control scheme and voltage deviation of energized buses are considered. Improved strength Pareto evolutionary algorithm (SPEA2) is used to get the Pareto frontier, and lexicographic ordering method is used to get the optimal solution. The effectiveness and validity of the proposed model and method is verified by Shandong power system.(2) In the light of the load restoration optimization problem under uncertainty condition,trapezoidal fuzzy parameters are used to express the unrestored loads, and the deterministic active power and frequency constraints are changed into fuzzy chance constraints, then a fuzzy chance constrained load restoration optimization model is proposed. The load restoration benefit and overload risk are considered in the model. To calculation the load weights, the fuzzy entropy is introduced to quantify the load uncertainty, and the load importance and load uncertainty are used as load evaluation indices. Finally, the clear equivalent forms are used to change the uncertain load restoration model into deterministic zero-one programming problem, which can be solved by existing mature mixed integer programming method. Case studies shows that the proposed model can balance the restoration benefit and overload risk, and the decision result is more suitable for load restoration problem with fuzzy loads.(3) In the light of the unit output optimization problem with fuzzy loads, a fuzzy chance constrained optimal control model of unit output is proposed in this paper. The model aims to minimize the active power adjusting time, and the fuzzy chance constraints of up spinning reserve and down spinning reserve are introduced to deal with the influence forecast errors. The reactive power balance and voltage constraints are also taken into consideration. To solve the model, binary variables and linear variables are introduced to linearize the clear equivalent forms of the spinning reserve constraints, and the DC power flows and linear programming approximation of AC power flows (LPAC) models are used to linearize active and reactive power balance constraints, respectively, then the unit output optimization model is changed into a mixed integer linear programming (MILP) problem. The effectiveness and validity of the proposed model and method is verified by the IEEE30 bus system. |
PDF Full Text Request