Integrating Transactive Energy and Machine Learning For Re-Energizing Wastewater Treatment Plants

A large wastewater treatment plant (WWTP) typically consumes 300-500 MWh of electricity per day and often relies on the grid power generated by burning fossil fuels. Wind- and solar-based distributed generation emerged as a clean energy solution to achieve environmental sustainability and net-zero performance. This study investigates power consumption trends in WWTP facilities using six and nine years of data, respectively. The study leverages machine learning algorithms for wind speed and power load forecasting. Particularly, recurrent neural network (RNN), long short-term memory (LSTM), and ensemble models are adopted as intelligent computing tool for generation forecasting. A regression model was developed to forecast the power output of onsite wind turbines. Managerial insights were obtained regarding the most effective model for wind power forecasting and load prediction of the WWTP in Melbourne, Australia, and the water treatment plant in San Marcos, Texas. The following research findings are obtained. First, when multiple criteria along with forecasting wind speed are considered, the RNN model provides much better prediction than the LSTM and ensemble models. Second, when integrated with two or more low performance neural network models, the ensemble model can yield more accurate results by collectively increasing their predicting accuracy. Third, the integration of renewable transactive energy and blockchain technology has the potential to realize peer-to-peer energy trading, in which electricity is sold directly between prosumers and consumers without the intermediaries. Future research could investigate other machine learning algorithms, such as convolutional neural networks, for improving wind speed or solar irradiance forecasting, and extend the machine learning based computing tools to residential, commercial, and other industrial prosumers.

Name thesis:degree_name

Master of Science

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Engineering

Grantor

Texas State University

Year dc:date.issued

2023

Author dc:creator

• Somvanshi, Shriyank

Advisor dc:contributor.advisor

• Jin, Tongdan

Committee members dc:contributor.committeemember

• Ikehata, Keisuke
• Fainman, Emily Zhu

• renewable energy
• machine learning
• wastewater treatment plants

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10877/16695

OAI identifier oai:identifier

oai:digital.library.txst.edu:10877/16695