Predicting Water Levels at Kainji Dam Using Artificial Neural Networks

Abstract

Poor electricity generation in Nigeria is a very serious problem. Accurate prediction of water levels in dams is very important in power planning. Effective power planning helps in ensuring steady supply of electric power to consumers. The aim of this study is to develop artificial neural network models for predicting water levels at Kainji Dam, which supplies water to Nigeria's largest hydropower generation station. It involves taking of a ten-year record of the daily water levels at the dam from 2001 to 2010. The daily water level data were used to develop five neural network models and an Autoregressive Integrated Moving Average (ARIMA) model to fit the daily water levels obtained in the year 2010. The results show that the prediction accuracy of the neural network models increased with increasing input, but after the four-input model the accuracy started declining. The four-input neural network model had the lowest relative error of 0.062 percent while the single-input model had the highest relative error of 0.237 percent. The ARIMA model with relative error of 0.039 percent had the best prediction. Generally, the models' predictions were good, but the neural network models which involve little mathematics were much simpler to build. The developed models will be very useful in power planning in Nigeria's hydropower stations for more ecient power supply.

Keywords: artificial neural network, hydropower, ARIMA, time series, modelling