Introducing
Your new presentation assistant.
Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.
Trending searches
9
Impact Of Renewable Energy Investment On Man-made Air Pollution
Achva Lederer & Tehila Atlan
Supervisor: Dr. Zohar Barnett-Itzhaki
Background
- Human activity is responsible for most of the world’s air pollution, both indoors and outdoors.
- The burning of fossil fuels such as coal and gasoline is the single largest source of air pollutants.
Background & Purpose
- Renewable energy is energy that is collected from renewable resources, which are naturally replenished on a human timescale. Included in the definition is electricity and heat generated from solar, wind, ocean, hydropower, biomass, etc.
- Investing in renewable energy can improve air pollution.
Our Purpose
Green Energy Investment
Pollutants
Purpose
To predict emissions of air pollutants as a function of renewable energy investment
Our Data
- Data from the OECD website about various air pollutants emissions and different energy investments.
Features & Goals
- Arranging the data. Using KNN algorithm to remove Nans.
- Dividing the data on each of the pollutants, by area of each country.
Features
Our Features
The features are different investments in energy, both in the financial and development sectors.
Mean Feed-in Tariff for Solar PV electricity generation
Development of environment related technologies
Total ODA for climate change mitigation
Environmentally related taxes - Total tax revenue
Total ODA for environmental issues
Diesel Tax
Goals
Goals
SOx- Sulphur Oxides
CO- Carbon Monoxide
NOx- Nitrogen Oxides
Particulates(PM25)
Tools & Results
Tools & Results
Tools
- pandas
- os
- Bioinformatics Toolbox
- Econometric Toolbox
- linear regression
- SVM
Linear Regression
- Building a linear regression model with all the features on each of the pollutants.
Linear Regression
- Testing the model on the test set showed relatively good prediction and low RMSE.
Coefficients
RMSE
Bold - The most significant features from all the features.
Goal Function - PM2.5 pollutant
- Run the model on the most significant features found in the general model and on all features except significant.
- Testing the model on the test set showed relatively good results and RMSE similar to the general model.
significant features
- In model without significant features the range of coefficients is lower.
Correlations between the six significant features
The regression results on the six most significant features
Results and comparisons
Comparison of the range of RMSE scores
The range of coefficients (thetas) values.
X-axis - The features
Y-axis - Coefficients values
RMSE of all features
The coefficients of all the features
RMSE of the six significant features
RMSE of the other features
All the features
the significant features
The coefficients of the other features
The coefficients of the six significant features
SVM algorithm
- Classification using the SVM algorithm. The distribution was based on the median of each pollutant.
SVM
- Testing the success rates on the test set showed success in 72% of predictions.
Conclusions
The results of the models are good and can predict relatively well the emission of air pollutants as a function of green energy investments.
Conclusions
It is interesting to build models that predict the impact of investments on future years.
Report of World Energy Outlook shows the effect of energy investment on air pollution
From the world
Thank you!