学术文献库

Indoor light is known to be a new energy source to power uW low consumption wireless sensor networks (WSNs). For wireless electronic devices that consume tens of mW, it is still challenging to harvest this amount of power in a low light indoor environment. The challenge comes from the fact that the irradiance is low, as well the fact that the source is a multi-spectral direct, reflective, and scattered mix of artificial and natural light, with intensity and composition varying in time. This article describes a method providing an evaluation of the potentially harvestable energy in real light environments. Measurements of indoor light spectral composition in real condition and optoelectrical characteristics of photovoltaic converters in a controlled environment are base on the method s calculation model. Real harvester prototypes based on GaAs commercial photovoltaic cells and power management integrated circuit, powering a commercial wireless e-ink display, have been conceived to compare with the model calculations. For two days of experimentation, model calculations achieve a mean absolute percentage error (MAPE) of 2 percents and 6 percents for Day 1 and Day 2, respectively. A method within this accuracy range would be a helpful tool to assist in designing light energy harvesting systems in such ways that they can be best tailored to the indoor lighting conditions to which they will be exposed.