Electronic Engineering 127

− high CO₂ concentration due to inadequate supply of fresh air,

− contamination arising from sources within the building (e.g., combustion products, tobacco smoke and other particles; volatile organic compounds (VOCs) from building materials, fabric furnishings, carpet, adhesives, fresh paint, new paneling, and cleaning products),

− contamination from outside the building (e.g., ozone, carbon monoxide, and particulate matter) through air intakes, infiltration, open doors, and windows,

− microbiological contamination of ventilation systems or building interiors.

The thermal comfort is a non-linear result of the interaction between environmental dependant parameters (the most important being air temperature, air velocity, relative humidity, mean radiant temperature), metabolic parameters (activity level, clothing) as well as subjective feelings. In order to compute a value of the indoor thermal comfort, the environmental variables must be measured at a location close to the user. The activity level and the clothing insulation can be obtained from suitable person detection sensors.

As temperature sensors serve conventional Pt1000 Platinum resistance thermometers evaluated on a bridge circuit. The non-linear temperature characteristics is modeled by a 3- order-polynom.

The capacitance humidity sensors are built in a double oscillator configuration with a reference branch in order to reduce temperature drifts. The humidity value follows from the difference between the two oscillator frequencies. The dust sensor is based on IR light absorption which depends on the particle concentration.

Perspectives

Next research steps are aimed at improving the long-term stability and robustness against influences of temperature and humidity of existing gas sensors by means of a model-based approach. Furthermore, on behalf of extensive measurements, most suitable sensor locations have to be found sensor subsystems and reliable, fuzzy-based control strategies have to be developed.