Uncertainty analysis and management in building energy data mining

May 22, 2023

Waqas Khan defended his PhD thesis at the department of Built Environment on May 22nd.

As the demand for sustainable practices in the built environment continues to rise, it is crucial to efficiently manage energy usage in buildings, especially with the increasing adoption of solar panels, and electric vehicle systems. Traditional energy management systems may not be equipped to handle such demands, and thus, it is crucial to develop novel approaches to optimize energy usage and minimize waste. Data mining techniques can help achieve these goals by analysing vast amounts of data generated by smart buildings and devices.

In his PhD thesis, Waqas Khan highlights the importance of managing uncertainty in data mining techniques to improve the accuracy, consistency, and interpretability in the context of energy management in buildings. However, the volume of data generated by these devices presents a significant challenge, necessitating more advanced algorithms and computing power to process and analyse these large datasets.

Accurate decision-making is essential for energy applications to ensure the safe and efficient operation of energy systems. Uncertainty in data mining can lead to inaccurate decisions and affect the reliability and safety of energy systems.

The algorithms and methodologies proposed by Khan in his PhD research seek to enhance the accuracy of predicting building load, solar generation, and electric vehicle charging to establish a more sustainable and efficient energy system. The findings of his research are critical for building energy data mining and can aid in the development of sustainable energy systems for the future.

Title of PhD-thesis: Uncertainty analysis and management in building energy data mining: A bottom-up approach considering the temporal and spatial aspect of data. Supervisors: Wim Zeiler and Shalika Walker.