MODELING AND PREDICTING THE MONO RIVER OVERFLOW UPSTREAM OF THE NANGBETO DAM IN WEST AFRICA USING MULTIPLICATIVE DETERMINIST MODEL

Abstract

The variation and non-control of the overflow of the Mono River adversely affects the performance of the Nangbetohydropower plant to the point thatitcan no longermeet the increasinglyincreaseddemand for electricity. This studypresents the development of an operational model for forecastingdaily river flows for the plant's water retention. The overflow of the Mono River at the upstreamhydroelectric dam from 1991 to 2019 wasanalyzed and modeled by the deterministicprocesswith R software in order to makepredictions. First, the flow serieswasanalyzed by the ARIMA model (18, 1, 2) then by a multiplicative model afterremoving the seasonal trends fromtheseseries by the movingaveragemethod. The calculatederror of the results of said model revealsthat the deterministic model integrates the input generationprocesseswith an error of the order of �� = 28.26% . Finally, an annual flow forecasting program has been developed as a planning tool for the operation of the dam, in order to meet production needs and to plan water releases.