Streamflow forecasting is very important for planning and operation of electrical Power systems. Since the arriving of neural networks and the application to forecasting problems, they become very interesting overcoming the traditional statistical methods. Radial basis neural networks are very useful for this kind of application. This work intends to use a radial basis network by the MATLAB toolboxes, a very useful tool, and apply to streamflow forecasting with data from a company of a Brazilian electrical system.
News published in Agência FAPESP Newsletter about the scholarship: