Abstract

A first-order one-variable grey model (GM(1,1)) is combined with improved seasonal index (ISI) to forecast monthly energy production for small hydropower plants (SHPs) in an ungauged basin, in which the ISI is used to weaken the seasonality of input data for the GM(1,1) model. The ISI is calculated by a hybrid model combining K-means clustering technique and ratio-to-moving-average method, which can adapt to different inflow scenarios. Based on the similar hydrological and meteorological conditions of large hydropower plants (LHPs) and SHPs in the same basin, a reference LHP is identified and its local inflow data, instead of the limited available data of SHPs, is used to calculate the ISI. Case study results for the Yangbi and Yingjiang counties in Yunnan Province, China are evaluated against observed data. Compared with the original GM(1,1) model, the GM(1,1) model combined with traditional seasonal index (TSI-GM(1,1)), and the linear regression model, the proposed ISI-GM(1,1) model gives the best performance, suggesting that it is a feasible way to forecast monthly energy production for SHPs in data-sparse areas.

You do not currently have access to this content.