A Method Of Evaluating The Transfer Capacity For Power System Considering Of The Injected Power Distribution And The Power Grid Topology

Calculation of transfer capability is a traditional issue in the electrical engineering field, which is valuable in both operational phase and planning phase. Accurate calculation of transfer capability can be used as an important basis for judging whether a power system can be in a safe and reliable operation. However, the transfer capability of a certain power system is a varying value, which may be changed accompanied with the evolution of the system. Although the topology of a power system is usually designed in an optimizing way to maximize the transfer capability, the capability itself will be inevitably weaken by the mismatch between the continuously increasing load and the hysteresis of the enhancement of grid topology. Related topics are seldom to be discussed due to the lack of qualification tools to deal with the varying transfer capability. For this reason, a new method for evaluating the transfer capability of power grid is proposed in this thesis, which is based on the knowledge of graph theory. In this method, an index of Spatial Characteristic Length Ratio(SCLR) is defined, and the transfer capability of the grid is calculated by calculating the index. Therefore, this paper firstly mainly studies the calculation of SCLR, and proposes a core algorithm named power supply path search algorithm between source node and sink node to calculate the index for large power grid. This algorithm is mainly used to obtain the maximum power transmission path between all the energy source nodes and those energy sink nodes which supply by it, thus the index SCLR can be calculated.when SCLR can be calculated, IEEE4,IEEE9, IEEE162 and a real power system with 134 nodes are taken as case study in this thesis, and the variation trend of the index is studied when the power injection or the power grid topology changes. Examples show that the index SCLR is intrinsically related with both the topology of the power system and all the bus injected powers, so it can be used to evaluate the transfer capacity related to spatial properties of a power system. In addition, SCLR is also related to the spatial distribution of the injected powers, and thus the value of SCLR was studied when the distribution of power generation differences in this paper using IEEE4 nodes system, and the result shows that SCLR can be used in the generator scheduling to provide the basis for the optimal combination of the power generation.There is a problem that calculating index SCLR often spend a long time when the scale of the power grid is relatively large with the help of power supply path search algorithm. Therefore, the idea of dynamic programming for optimal path was used to improve the algorithm, so a new algorithm for calculating index SCLR was proposed in this thesis, which is based on the queue FIFO(first In first out) data structure and the adjacency matrix of the digraph of load flow solution. All the maximum power transmission paths between the source node and each sink node which supply by it are found out using the breadth-first traversal method, thus SCLR can also be obtained. The time complexity is analyzed and the speed of the calculation of SCLR is tested by IEEE examples before and after the algorithm is improved, and the test results show that the proposed method are efficient and feasible.At last, the difference of the index SCLR of the grid under different power grid planning projects failed to build on schedule is studied using the data from the southern power grid actual planning network. According to the value of SCLR, the priority order of planning project construction was given, and the result can provide the basis for the Power Grid Corp to guide the power network planning. This thesis makes a summary of the whole paper in the end, and gives a prospect of future research. Conclusion is made at the end of this thesis, and the directions for future research are prospected.