This work presents an autonomous embedded system for evoked biopotential acquisition and processing. The system is versatile and can be used on different evoked potential scenarios like medical equipments or brain computer interfaces, fulfilling the strict real-time constraints that they impose. The embedded system is based on an ARM9 processor with capabilities to port a real-time operating system. Initially, a benchmark of the Windows CE operative system running on the embedded system is presented in order to find out its real-time capability as a set. Finally, a brain computer interface based on visual evoked potentials is implemented. Results of this application recovering visual evoked potential using two techniques: the fast Fourier transform and stimulus locked inter trace correlation, are also presented.

In remote regions, were a main Utility System for energy distribution is not available, the implementation of Microgrids with hybrid generation are very useful. In this context, a Stand-Alone wind/diesel/gas/battery/supercapacitors Hybrid Microgrid topology is proposed. The main scope is to create a low cost system to feed Residential loads through a local AC bus, using wind energy obtainable at system location. The random characteristics of the wind makes necessary to smooth the fluctuating supply in order to minimize the disturbances on local grid electric parameters. To overcome these problems, a Wind Turbine based on a DFIG machine and Strategies to control the power of a Storage Energy System are employed. An appropriateSupervisor Module with a reliable communication network is used to balance generation and loads, in order to properly operate the Microgrid. The system structure, operation modes and mathematical models for simulation and validation are presented.

No abstract