基于神经网络的居民住宅需用系数模型研究

Research on Residential Demand Factor Model Based on ANN

居民住宅小区需用系数的精确计算对于小区的电力资源容量配置具有重要意义。针对现有配电网规划指导原则中需用系数参考范围过宽、不够精确的缺点,通过收集唐山市中心城区配电网近年来的大量实际运行数据,针对不同类型的居民住宅小区,利用神经网络方法分别建立了考虑户数多少、户型面积、入住率等影响因素的居民小区需用系数模型。应用实例表明,在保证配电网安全运行的前提下,借助该模型规划居民小区的电力资源容量配置,可有效地节约投资成本,从而提高配电网运行的经济性。

Precise calculation of residential demand factor is of great significance for distribution network planning. The demand factor in existing guidelines of distribution network planning has the disadvantages of wide reference range and imprecision. By collecting a large number of actual operation data in distribution network of Tangshan central city in recent years,according to different types of residential district,using the method of neural network,considering the number,door area,occupancy rates and other factors that affecting demand factor,residential demand factor model is established. Simulaton results show that the demand factor which is calculated by using the model can overcome the above disadvantages,save investment costs and ensure the economic and safe operation of power distribution network planning.