Machine Learning for Occupancy Detection through Smart Home Sensor Data

Data from mechanical extract ventilation units of Renson Ventilation nv installed in Belgium is utilized to detect space occupancy through machine learning. Challenges with the detection of occupancy using data captured by these smart devices are (1) absence of labelled data for training a machine learning model, and (2) occupant’s CO2 generation rate and building layouts influence the measured CO2 concentrations, which prevents simple rule-based models to be used for data labelling.

A Programmable Image Sensor for Smart Daylighting and Glare Control in Buildings

In this study, a new programmable low-cost camera sensor equipped with a fisheye lens is used to detect the luminance and position of exterior glare sources using per-pixel analysis. The High Dynamic Range Imaging sensor is mounted on the interior surface of the window and facilitates an efficient daylight glare control framework. The control was implemented in a testbed office with automated roller shades and was compared with two commonly used solar protection control algorithms.

CO2 and volatile organic compounds as indicators of IAQ

The trend toward minimizing ventilation of houses in order to reduce energy consumption for heating and cooling leads to an increase in indoor air pollution. The deterioration of indoor air quality (IAQ) negatively affects human health, safety, productivity and comfort. In order to evaluate the scale of this influence IAQ assessment has to be performed. However, the IAQ itself is not well defined and a number of parameters are considered as its indicators. In this work we compared carbon dioxide and volatile organic compounds as indicators of indoor air quality.

Real-time determination of indoor contaminant source location and strength, Part II: with two sensors

In the preceding companion paper (Part I), a method with one sensor that could identify the indoor contaminant source location and strength in short time was presented. On the basis of further theoretical study, a method with two sensors is presented in this paper to identify contaminant source with higher accuracy. This paper demonstrates how to use the method with two sensors to find the location of contaminant source in a threedimensional room. In addition, the accuracy of two types of methods was compared.

Real-time determination of indoor contaminant source location and strength, Part I: with one sensor

In case contaminants are released in occupied rooms, it is necessary to determine the contaminant source location and strength rapidly so that prompt response measures can be taken to protect indoor occupants. This paper presents a new method with one sensor to identify the contaminant source location and strength. It completes the time-consuming computational fluid dynamics (CFD) simulations before the release event, and finds the source in real time during the event. In addition, an index called “correctness probability” for evaluating the accuracy of this method is proposed.

Development of a Temperature-Based Sensor for Airflow Pattern Control in Mechanically Ventilated Buildings

To satisfy the optimal environment in agricultural buildings, much effort is made to controlventilation rate and indoor temperature. However, distribution of fresh air is equally important for animal performance and welfare. So far, no sensor is available to measure and to control airflow pattern continuously in a ventilated building. Therefore, an airflow pattern sensor was developed to measure the trajectory of a nonisothermal air jet in a building with a single or multiple air inlet(s).

Hybrid demand controlled ventilation system

The purpose of this system is to provide one improved ventilation system allowing significant good indoor air quality, heating (and cooling) energy savings and acceptable thermal comfort on summer, by using especially renewable energy.This concept is based on sensors measuring relative humidity in bathroom and kitchen, occupancy in bedrooms and toilet, and agitation (i.e.

Demand controlled ventilation (DCV) systems : performances of infrared detection

For energy savings, DCV systems are more and more used in ventilation systems. In France, in nonresidential buildings, these systems are generally controlled by either a CO2 sensor, or an opticalmovement detection (infrared). The part of the study we present here was to determine laboratorytests methods to assess the performances of optical sensors for ventilation application. The pointwas to characterise them by checking their detection in front of some determined movements orevents.

L'ambiomètre : un outil de mesure du confort d'ambiance dans les bureaux (The Ambiometer : a sensor for the ambient comfort in offices)

The ambiometer is a tool that allows both the recording of physical parameters of comfort and sensations of the occupant.The method developed is based on field investigations of four components of comfort : thermal, acoustic, visual and olfactory components : the experiment comfort parameters are recorded with the Ambiometer and the occupant gives his own feeling on each comfort parameter and on overall comfort.A database including those physical and perceptive parameters of comfort has been made, information was collected from about sixty offices.

A new approach for measuring predicted mean vote (PMV) and standard effective temperature (SET)

An effective sensor for thermal comfort index is necessary for a successful comfort index-based HVAC control system. A comfort sensor with a new structure is proposed in this paper. This instrument consists of an equivalent temperature sensor, a relative humidity sensor and a temperature sensor.
Simulations show that the suggested PMV and SET sensor can have a good measurement of PMV, but for SET more research is needed.