2019 Progress Report: Sensible Home: Micro-environmental control through wearable personal sensorsEPA Grant Number: SU836940
Title: Sensible Home: Micro-environmental control through wearable personal sensors
Investigators: Wang, Julian , Fan, Howard , Feng, Yanxiao , Yakkali, Sai Santosh , Duan, Qiuhua
Current Investigators: Wang, Julian , Fan, Howard
Institution: University of Cincinnati
EPA Project Officer: Callan, Richard
Project Period: February 1, 2017 through January 31, 2019 (Extended to January 31, 2022)
Project Period Covered by this Report: February 1, 2019 through January 31,2020
Project Amount: $75,000
RFA: P3 Awards: A National Student Design Competition for Sustainability Focusing on People, Prosperity and the Planet - Phase 2 (2016) Recipients Lists
Research Category: P3 Challenge Area - Chemical Safety , Sustainable and Healthy Communities , P3 Awards
There are three major research tasks performed in this phase. First, the team has been testing the data sensing and processing procedure using personal wearable sensors for indoor comfort purposes within the controllable mockup space at the University of Cincinnati campus. This mockup space has separated HVAC control systems, tunable ceiling lights, and an exterior window. A pilot study with 5 participants has been conducted. It is found that the micro-environment of users has a more accurate prediction on users' comfort, especially in winter in which exterior windows have more impacts on perimeter zonal thermal conditions. To understand the potential impacts, the research team also studied the combined effects of air vent placements and window properties on the occupants’ thermal comfort. Second, as we demonstrated in the Phase I study, understanding occupant’s behaviors and lifestyles may facilitate the process of bringing users into the building control control. Thus, the team also investigated a method predicting occupants’ behaviors and lifestyles via studying their previous utility data using data mining techniques. The techniques have been developed and tested in three senior living settings. Third, previous studies have indicated that clothing insulation has significant impacts on individual thermal comfort. In this project, we used low-cost temperature sensors and 3D printing techniques. The inside element documents the skin temperature (Tint), while the outside portion records the air’s ambient temperature (Text). From the known area (A) of the surface touching the sensors and the known material insulation (R), formed from a 3D printing process, we could calculate the clothing insulation upon the steady heat transfer in which the rate of heat transfer remains constant.
Developed and addressed a simplified method using low-cost sensors to measure the key micro-environmental factors and the user factors including the clothing insulation, which is deemed as one of the necessary parameters to predict individual comfort. This clothing insulation measurement method can be also expanded into other building environmental in-situ measurements, such as building glazing, walls, etc. It has potential impacts on the Agency’s mission for energy savings and environmental sustainability.
Demonstrated a potential multisensory mechanism between thermal and visual aspects. In the conventional indoor comfort study, thermal and visual aspects were completely separated. However, the collective effects were found in our study. More research works are needed in the next phase, with a large sample size. Also, this reveals some opportunities (e.g., manipulating the user thermal response by visual quality control by smart lighting systems) to further save building energy and enhance the indoor environmental quality for occupants. This is also aligned with the Agency’s missions on people and environment.
In the next stage, the team will recruit subjects to complete the individual comfort model development. The basis of our work is a quantitative model – Predicted Personal Vote (PMV) developed by Fanger, and it has been accepted by ISO 7730 Standard and ASHRAE 95 Standard. The PMV model predicts the mean sensation vote, for a large group of persons in a given indoor climate. In our work, we extend the PMV model to the Individualized Personal Vote (IPV) model and incorporate the visual sensation into this model. For each person, the IPV function has two parts, the PMV part and the individualization part: IPV(x) = PMV(x) + Individual(x). Where PMV(x) is the output of the PMV model obtained through wearable sensor data, and Individual(x) models how the current user is different from an average person. We model Individual(x) as a function of age, gender, BMI, Motion, heart rates, and heat flux through clothes. We will evaluate multiple machine learning modeling techniques, such as artificial neural network, gradient boosting machine, and classification and regression tree, to develop the models. Gradient boosting is a machine learning technique that ensembles a number of weak predictors to form an accurate prediction model. We will train the model in two stages: offline model development and online calibration. During the offline stage, we will focus on training with the data collected in our lab. We will recruit 50 participants and expose them in different indoor environmental conditions and assigned activities. On the other hand, we will need repeated measurements on individual I, and that is the reason we would need the on-line calibration stage. During this stage, we will ask the users to wear the sensors and report their comfort level using a smartphone app, so that for each single user, we will have multiple observations over a period of time.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
|Other project views:||All 16 publications||6 publications in selected types||All 6 journal articles|
|| Duan Q, Wang J. Thermal Conditions Controlled by Thermostats:An Occupational Comfort and Well-being Perspective. Civil Engineering and Architecture 2017;5(5). .
|| Li J, Qi M, Duan Q, Huo L, Wang J. Towards Pedestrian Microclimatic Comfort: A Rapid Predication Model for Street Winds and Pedestrian Thermal Sensation. Nano Life 2018;8(02):1840006.