Scholarlink

<p>Climate change affects energy security and the operation of hydroelectric plants (HEPs) due to reduced precipitation. The problem lies in the lack of machine learning models that simulate the hydraulic inertia of reservoirs through a sequential and causally ordered architecture. The objective is to develop and apply a multivariable hydrological modeling methodology based on the Extreme Gradient Boosting (XGBoost) algorithm, using a sequential cascade architecture, to simulate the dynamics (inflow, storage, water level, and outflow) of the Batalha and Serra do Facão (SEFAC) hydroelectric plants under the SSP2-4.5 and SSP5-8.5 climate scenarios. The choice of the GCM MPI-ESM1-2-LR is justified by a previous ranking study that attested to its superior performance compared to the ensemble in the region, with the data already pre-corrected. The methodology demonstrated high generalization and predictive accuracy, with robust metrics: Kling-Gupta Efficiency (KGE) between 0.99 and 1.0 and Root Mean Square Error (RMSE) below 3.2&#xa0;m³/s in the simulations. The projections revealed heterogeneous precipitation patterns in the basin: a reduction upstream of Batalha and an increase downstream of SEFAC. The key results indicate that the UHE Batalha projects a reduction of up to 10% in maximum storage. The SEFAC hydroelectric plant is more critical, where future maximum storage will consistently remain below the second historical quartile. This widespread decline in storage and flows suggests that operations will face considerable challenges in maintaining historical levels of energy production. The present study reinforces the need for adaptive planning for the Brazilian electric sector.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hydrological and Hydraulic Analysis of Hydropower Plants Reservoirs Under the Influence of Climate Change with a Sequential Machine Learning Model

  • 期刊论文

  • 发表日期：

    2026年1月5日

查看全文
  • Josias da Silva Cruz,
  • Leonardo Melo de Mendonça,
  • Nelio Moura de Figueiredo,
  • Claudio Blanco,
  • Antonio Cesar Pinho Brasil Junior

摘要

Climate change affects energy security and the operation of hydroelectric plants (HEPs) due to reduced precipitation. The problem lies in the lack of machine learning models that simulate the hydraulic inertia of reservoirs through a sequential and causally ordered architecture. The objective is to develop and apply a multivariable hydrological modeling methodology based on the Extreme Gradient Boosting (XGBoost) algorithm, using a sequential cascade architecture, to simulate the dynamics (inflow, storage, water level, and outflow) of the Batalha and Serra do Facão (SEFAC) hydroelectric plants under the SSP2-4.5 and SSP5-8.5 climate scenarios. The choice of the GCM MPI-ESM1-2-LR is justified by a previous ranking study that attested to its superior performance compared to the ensemble in the region, with the data already pre-corrected. The methodology demonstrated high generalization and predictive accuracy, with robust metrics: Kling-Gupta Efficiency (KGE) between 0.99 and 1.0 and Root Mean Square Error (RMSE) below 3.2 m³/s in the simulations. The projections revealed heterogeneous precipitation patterns in the basin: a reduction upstream of Batalha and an increase downstream of SEFAC. The key results indicate that the UHE Batalha projects a reduction of up to 10% in maximum storage. The SEFAC hydroelectric plant is more critical, where future maximum storage will consistently remain below the second historical quartile. This widespread decline in storage and flows suggests that operations will face considerable challenges in maintaining historical levels of energy production. The present study reinforces the need for adaptive planning for the Brazilian electric sector.