EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3376. Estimating Greenhouse Gas Emissions Through Deep Learning: A Gradient-Based Analysis of Input Impact
Invited abstract in session TB-53: Toward Climate Neutrality, stream Sustainable and Resilient Systems.
Tuesday, 10:30-12:00Room: 8007 (building: 202)
Authors (first author is the speaker)
| 1. | Manizheh GhaemiDizaji
|
| University of Calgary | |
| 2. | Laleh Behjat
|
| Electrical and Computer Engineering, University of Calgary | |
| 3. | William D. Rosehart
|
| Department of Electrical and Software Engineering, University of Calgary |
Abstract
We aim to proposes a novel approach to estimating greenhouse gas (GHG) emissions by leveraging deep learning techniques and conducting gradient-based analysis to assess the influence of various inputs on CO2 emissions. The inputs considered encompass a range of socio-economic and environmental factors, including Gross Domestic Product (GDP), transportation modes, and energy generation mix. By constructing a deep neural network architecture, we aim to capture the complex relationships between these inputs and GHG emissions, enabling robust estimation and prediction capabilities.
Furthermore, our approach integrates gradient-based analysis to elucidate the relative contribution of each input towards overall emissions output. This allows for a granular understanding of the impact of individual factors, empowering policymakers and stakeholders to prioritize interventions effectively. Through iterative modifications of input values and analysis of resultant emission outputs, we can delineate the share of each input in driving emissions variations.
This research presents a promising methodology for GHG emissions estimation and analysis, offering insights into the intricate interplay between socio-economic drivers and environmental outcomes. By combining deep learning with gradient-based analysis, we provide a powerful framework for informing climate policy decisions and facilitating sustainable development strategies.
Keywords
- Machine Learning
- Transportation
- Sustainable Development
Status: accepted
Back to the list of papers