Spillage Forecast Models in Hydroelectric Power Plants Using Information from Telemetry Stations and Hydraulic Control

Hydroelectric power plants’ operational decisions are associated with several factors, such as generation planning, water availability and dam safety. One major challenge is to control the spillage from reservoir. Although this action represents a loss of energy production, it powerful strategy regulate reservoir level, ensuring dam’s The decision use must be made in advance based on level demand predictions. present work applies supervised machine learning techniques predict operating condition hydroelectric plant for 5 h ahead. method, real time, aims assist operator so that he can make more assertive safer decisions, avoiding waste resources increasing safety dams. Random Forest Multilayer Perceptron methods were used define architecture compared forecasting capacity. proposed methodology was applied 902.5 MW Power Plant located Tocantins River, Brazil. results demonstrate effective assistance operators decision-making, presenting accuracy up 99.15% spill decision.