Layered Transmission Network Planning Based On Improved Differential Evolution Algorithm

Grid construction is developed to the direction of higher voltage level and moretransmission capacity. Reasonable power grids structure is important to ensure thesafe and stable operation of the power system, how to get an optimal networkexpansion plan becomes a subject with great practical significance.In the schematic formation stage, the first step has to establish a propertransmission network planning model, and then use a kind of effective method to findthe solution of the model. According to layered transmission network planning underthe multi-voltage level condition, In order to meet load uncertainty growth and thewhole system power flow scheduling better in the power grid operation, the planningmodel is established which considers surplus transmission capacity, the ratio ofcomprehensive investment and operation costs and surplus transmission capacity andthe short circuit current are introduced as the target, it is introduced to get moresurplus transmission capacity under the smaller comprehensive investment andoperation costs; the short circuit current can be controlled effectively to avoidimpacting the safe and stable operation of power grids under the target of economyand security. Because the heavy or light load lines can affect grid safety andeconomic operation, the comprehensive investment operation cost and the branchesload rate variance are used to be considered as the transmission network planningmodel function, which is used to optimize the heavy and light load lines in order tosatisfy the distribution of power flow more reasonable, and improve power gridoperation environment.In solving model stage, differential evolution algorithm is chosen to solve themodel. But the searching strategy of differential evolution algorithm is too single andthe local searching capability is so poor. Meanwhile, the algorithm has thephenomenon of premature convergence. In order to enhance differential evolution algorithm’s global and local search ability, more mutation strategies and localoptimization strategy are introduced to design a new algorithm called multi-strategydifferential evolution algorithm which adapted to solving large-scale transmissionnetwork planning. In order to improve performance of the algorithm further, thechaos optimization algorithm can strengthen global search ability, differentialevolution algorithm’s searching speed is fast, chaos differential evolution algorithm isdesigned, which is formed by changing the mutation and cross strategies andincreasing chaos disturbance to improve the global and local search ability. Throughthe examples’ analysis, it not only verifies that the models are applied to layeredtransmission network planning correct and effective, but also verifies the improvedalgorithm has good computing speed and good convergence.