Project Team
Students
Bilal Faye
Electromechanical Engineering Technology
Altoona
Faculty Mentors
Dr. Sohail Anwar
Altoona
College of Engineering
Dr. John Gershenson
University Park
College of Engineering
Project
https://sites.psu.edu/mcreu/files/formidable/2/MCREU_FAYE_1_2.pdf
Project Video
Project Abstract
The innovation surrounding the implementation of microgrid technologies has been gaining traction around the world. Their flexibility in operation and their ability to easily incorporate distributed renewable energy make them interesting to consider when thinking of improving unreliable African grid infrastructures and powering remote areas. Nevertheless, the generation must exactly match the demand in order to develop pertinent solutions with the integration of microgrids. An incorrect energy load assessment is important as its underestimation or overestimation can lead to unforeseen technical and economic performance failures. Numerous research papers have highlighted the importance of developing accurate load profiles for developing countries. In fact, the literature review confirms the problems associated with the development of accurate load profiles as a limitation in their economic feasibility studies. Interview-based load profiles, through the audit of the locals on their energy consumption habits, is the method that is the most widely used in academia. The issue of interview-based load profiles failing to provide an accurate estimate is linked with a lack of necessary data in developing countries. While evaluating various microgrid feasibility studies in rural areas, it is easy to notice that despite the design methods undertaken, the information about the users’ electric consumption has different levels of details and is mostly limited. The key objective of this study is to highlight and evaluate alternative methods of creating load profiles for the integration of microgrid technologies in developing countries. In addition to that, this study uses data from a minigrid in Tanzania and trains it with multiple predictive machine learning methods to assess the feasibility of artificial intelligence in the domain of real-time load forecasting.
Keywords: Load Forecasting, Developing countries, Alternative methods, Predictive Machine Learning
Evaluate this Project
Use this form link to provide feedback to the presenters, and add your project evaluation for award(s) consideration.