Thesis
Vehicle emission monitoring and design concept of a low-cost IoT based system
Washington State University
Master of Science (MS), Washington State University
08/2020
DOI:
https://doi.org/10.7273/000004145
Handle:
https://hdl.handle.net/2376/124957
Abstract
Vehicle emission is a significant source of air pollution. A considerable number of deaths are associated with air pollution. The existing vehicle emission monitoring system is unable to mitigate pollution properly. Real-time pollution monitoring from the point of source is necessary to address the problem more precisely. An IoT platform using commercially-off-the-shelf (COTS) components was developed and tested with temperature and humidity sensor. Cloud platform was used for data storage, analytics, and push notification to end users. The platform also has the capability of capturing on-board diagnostics (OBD) data from vehicle to Raspberry Pi. An artificial neural network (ANN) was implemented for model development as it has several advantages over other machine learning algorithms. For training the ANN model, substantial data was needed. About Twenty (20) hours of driving data were collected using Nissan Sentra 2019 in Seattle areas. Data was collected using off-the-shelf readily available instruments. A series of data curation process was implemented to understand the pattern between input and output parameters. Finally, Python machine learning libraries were executed to train and build the ANN model. For simplification, only two input parameters, speed, and altitude were trained against NOx concentration as output. Further investigation is needed by using other input parameters, individually or combined. Additional quality data is needed in order to establish a pattern between input parameters such as speed, altitude, vehicle type, etc. and output variable NOx concentration. Further research is needed on what and how data can be collected, processed, and correlated using ANN modeling. A couple of recommendations on predictive applications are discussed briefly at the end. One focuses on application in autonomous vehicles and other one improving service of large fleets operated by logistics companies. In both cases, expected benefits will impact energy and environment.
Metrics
4 File views/ downloads
52 Record Views
Details
- Title
- Vehicle emission monitoring and design concept of a low-cost IoT based system
- Creators
- Mohammad Fahad bin Alam
- Contributors
- Md Akram Hossain (Advisor) - Washington State University, Civil and Environmental Engineering, Department of
- Awarding Institution
- Washington State University
- Academic Unit
- Engineering and Applied Sciences (TRIC), School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University
- Identifiers
- 99900890779601842
- Language
- English
- Resource Type
- Thesis