Forecast, distinguish, and optimise electrical demand

• Main Author: Lim, Shi Jie
• Other Authors: Goh Wang Ling
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/167540
• Institution: Nanyang Technological University

Time series load forecasting is an important aspect when it comes to energy management. This is an Industrial Sponsored Project (ISP-FYP) and the goal is to help smart building clients achieve accurate forecast of energy consumption based on historical data. This project presents a comparison of several machine learning models focusing on regression models and will discuss the advantages and disadvantages of the individual model and their viability to the dataset. This project also discusses the steps in building the models such as data preprocessing and creating functions. One model will be chosen out of the comparison for further development, implementation and evaluation and eventually be deployed for production use. With accurate load forecasting, it will ensure allocation of energy resources more efficiently and will prevent oversupply and overproduction of energy. This will enable smart building clients to gain insights of their energy consumption and use their energy efficiently. Moreover, with accurate load forecasting, it can also lead to cost savings for smart building clients. By accurately predicting their energy consumption, smart building clients can manage their energy consumption and lessen their dependency on costly peak-hour electricity. As a result, this significantly save the cost for the clients. Additionally, load forecasting can help with the integration of renewable energy sources into the energy system, such as solar and wind power. Smart building clients may adjust their energy output from renewable sources to match demand by accurately forecasting their energy use. Therefore, conventional energy sources are reduced, and sustainable energy practices are promoted.