Environmental Effects of Renewable Energy a Machine Learning Approach




Yetis, Yunus

Journal Title

Journal ISSN

Volume Title



This research study attempts to evaluate the impact of sustainable energy (particularly the increase

in the total energy consumption of alternative energy sources) on air pollutant emissions

V OCs, PM10, PM2:5, NOx, and SO2. The study describes data and sources of information and

recommends a framework for carrying out this approximation based on surveys from the Texas Environmental

Quality Commission (TCEQ). The primary goal is to allow a broader awareness of the

benefits as well as trade-offs of sustainable energy use in terms of air pollutant emissions relative to

its competitors’ non-renewable energy use. Since electricity output from natural gas and renewable

energy increases, the shutdown of several Texas coal-fired power plants is becoming highly possible.

Throughout this research, the possible effects of these closures on human health, as well as air

quality, were examined by connecting a local photochemical framework with a tool for evaluating

health implications. The effects ranged widely across thirteen of the major coal-fired power plants

in Texas, often by more than one scale factor, even after generation-by-generation normalization.

Although a number power plants had marginal effects on emissions at essential sensors, an average

result for PM2:5 as well as ozone was observed up to and 0.2 g=m3,and 0.5 parts per billion

(ppb) as well as maximum impacts up to 0.9 g=m3 and 3.3 ppb, respectively. Health implications

derived mainly from fine particular matter and were more significant in order of magnitude for

plants lacking SO2 air filters. Health consequences levels between the two reduced-form models

and baseline models were reasonably consistent. This thesis deals with the visibility forecasting

task through machine learning techniques using weather station evidence. Visibility is among the

most significant consequences of conditions on coal power plant . Low visibility circumstances

may have a substantial effect on the safety, resulting in adverse situations, IV triggering incidents,

and endangering human health. Realistic visibility modeling plays a crucial role in coal power

plant networks based on decision making as well as management. Nevertheless, visibility prediction

remains a critical, strenuous activity given the uncertainty and volatility of meteorological

parameters, and a subject of considerable interest in coal power plant statewide. This study aims

to explore the use of deep learning models for the task of visibility forecasting in one step (i.e.,

visibility distance estimate for the next hour) utilizing time-series data from ground meteorological

stations. The study cites 3 architectures in the learning algorithm: long short-term memory

(LSTM), Recurrent Neural Network (RNN), network, and Artificial Neural Network (ANN). Two

datasets from Texas were used to evaluate the models.


This item is available only to currently enrolled UTSA students, faculty or staff. To download, navigate to Log In in the top right-hand corner of this screen, then select Log in with my UTSA ID.


Air Pollution, Data Modelling, Deep Learning, Forecasting, Machine Learning, Renewable Energy



Electrical and Computer Engineering