Study On Practical Method Of Multi-stage Planning Optimization For A Large-scale Distribution Substation System

Due to the long-tern time nature of power grid construction, it can not only enable the orderly and targetable network construction development but also obtain a less cost for the entire planning period to carry out the substation planning for each stage of planning period and determine the scientific, reasonable and good-quality planning schemes for transition years.Through the study on the models, principles, main technical indexes and economic evaluation methods of substation planning, an optimization mathematical model of substation multi-stage planning is determined, which takes the minimum of investment and operating costs throughout the planning period as the objective function. Through the cost research and analysis of distribution networks, the cost models of substations and lines are established. Based on the loading and feeder number of a substation, a heuristic approach is proposed to obtain the reasonable operation and investment costs of feeders.A quasi-dynamic planning algorithm is proposed for the multi-stage planning optimization of a large-scale substation system. The demand saturation year, a middle year or the end of recent years is selected as the target year, and the recent years are regarded as the planning period based on load prediction situation. Also the substation sites and sizes from the target-year are taken as the potential ones for all stages of planning period. Based on the substations’unit load cost (including its feeders’cost) indexes, a forward quasi-dynamic planning method of multi-paths is presented to determine the substation planning for all stages. Based on the substations’unit load cost and model constraints, feasible paths can be reduced. In order to take into account both of the solution speed and accuracy, the number of optimal paths to be recorded for all stages can be set flexibly according to the substation system size, which is suitable for the dynamic planning of a large-scale substation system. And finally, human intervention is combined with an auto planning algorithm to optimize the planning schemes of all-stages, in order to take into account the geographical and environmental influence on the new substations.The proposed method is applied to the actual examples with273loads、374loads and3741loads. The results show the model and algorithm is reasonable, practical and effective. The proposed method is applied to the study of reasonable substation voltage levels (110kV/lOkV, and35kV/10kV) of large areas with low load density, and the results show that it is advantageous to employ the35kV substations instead of110kV ones.