Forecast Techniques Applied to Feasibility Studies for Micro-Hydraulic Generation

This paper presents the application of time series forecasting techniques to feasibility studies of micro-hydraulic generation. Available literature, details several techniques developed and implemented to perform time series forecasting. This paper will focus on the following techniques: ARIMA (auto-regressive integrated moving average), neural networks and evolutionary computation (EC). Based on the obtained results of the forecast techniques applied to the water flow time series, it is possible to determine if a micro-hydraulic plant can be installed, the theoretical power generation and the technical characteristics of each electro-mechanical component of the micro-hydraulic generation system.