Optimizing the control of Venetian blinds with artificial neural networks to achieve energy savings and visual comfort

Solar shading devices, such as Venetian blinds, are effective in controlling heat and light gain in buildings. This study focuses on developing an Artificial Neural Network (ANN) to automate the management of Venetian blinds in order to find a balance between energy savings and visual comfort. Typically, automatic control strategies rely on cut-off angles or maintaining appropriate indoor illuminance. However, finding the optimal trade-off between solar gain and daylight is challenging, particularly in regions with temperate or tropical climates. To this end, this paper proposes a novel approach to identify the best slat arrangement. The optimal angle is determined by simulating all possible slat angles using EnergyPlus and selecting the one that maximizes an objective function. The ANN is trained to recognize this angle as a function of variables that can be easily measured in situ. The resulting building's behavior is tested on EnergyPlus using the output provided by the ANN. The results demonstrate that the target angle can be identified with high reliability, even for windows with different orientation and in other cities. The energy savings for heating/cooling are up to 37.7 %/38.7 %. The UDI200-800 increases by up to 38.5 % at the expense of a slight increase in lighting consumptions (13.5 %).