ENBPar Proposes New Methodology to Forecast Itaipu Supply to Paraguay from 2027 Onward.
The Brazilian Nuclear and Binational Energy Participation Company (ENBPar) is spearheading an initiative to enhance how the energy supply from the Itaipu hydroelectric power plant is forecast for Administración Nacional de Electricidad (Ande), Paraguay’s electric utility. The proposal aims to implement a new methodology in the Monthly Operation Program (PMO) starting January 2027.
The central goal of this update is to reduce discrepancies between the energy demand projected in the PMO and the actual consumption. These forecasts, periodically provided by ENBPar to the National Electric System Operator (ONS), are crucial for modeling the Brazilian electric system and, consequently, for operational planning and scheduling.
The Need for Methodological Review
The forecast of Itaipu’s energy supply to Ande is influenced by Paraguay’s energy needs. However, the behavior of the Paraguayan load has shown variations that do not always align with its growth or seasonality.
This dynamic has resulted in changes to the supply, impacting various stakeholders in the Brazilian electricity sector, which prompted the review of the current methodology.
In collaboration with Itaipu Binacional and the ONS, ENBPar has developed a new method. The National Electric Energy Agency (Aneel) was consulted and indicated that prior approval of the methodology is no longer necessary, recommending alignment with the system operator and the Electric Energy Commercialization Chamber (CCEE).
Details of the New Proposal: Naive Method Adjusted for Seasonal Indexes
The technical note prepared jointly with the ONS describes the current methodology, which divides the forecast into two parts: one for the first year of the PMO horizon and another for subsequent years. For the latter, an adjustment factor of 1.05 is applied to project the supply.
The new approach, termed Naive Method Adjusted for Seasonal Indexes, starts with the supply recorded in the previous month and applies an adjustment factor derived from the historical average growth between the analyzed months. This allows the forecast to be updated monthly, reflecting operational reality more accurately.
Testing and Validation of the New Methodology
To test the effectiveness of the new proposal, three methodologies were compared using historical data of Itaipu’s supply to Ande between 2010 and 2020. The Naive Method Adjusted for Seasonal Indexes demonstrated the best overall performance.
Additional tests with different historical periods for calculating the adjustment factor, using rolling windows of three to fifteen years, confirmed the superiority of the new model.
The best result was achieved with the Naive Method Adjusted for Seasonal Indexes, using an eleven-year rolling window for the adjustment factor calculation. The technical note recommends the adoption of this new methodology, as the tests indicated a significantly smaller deviation in supply forecasting.
Implications and Next Steps
The expectation is that the new methodology will be implemented in the January 2027 PMO. This deadline was set to accommodate the necessary procedures for presentation, analysis, approval, and proper dissemination to market agents.
During the communication of the new methodology to electricity sector participants, a simulation of the supply forecast for Ande over a sixty-month horizon will be presented. This information will be valuable for market agents to conduct their own analyses and projections, strengthening the sector’s energy planning.
