Abstract

Inflow into the Kainji hydropower reservoir was predicted from year 2017 to 2050 using historical inflow data and Markov model. Statistical parameters were used to evaluate the predicted model. Basic hydropower equation and operational constraints were used to simulate future generation at the station considering various percentage of water stored in the reservoir as ecological flow release (EFR). Results revealed that the Markov model predicted the inflow perfectly with correlation coefficient (r), mean absolute error (MAE), relative bias (RB) and coefficient of determination (R2) of: 0.99, 0.10, 0.002 and 0.98 respectively. Total simulated annual energy generation decrease with increase in EFR percentages. It was also noticed that the simulated energy generation is higher than the actual energy generation for all the scenarios. The results of this work can be used by the management of the Kainji hydropower station for future planning of their generation.

Key words: Energy Generation, Hydropower reservoir, Inflow, Simulation, Prediction