Building energy prediction using Artificial Intelligence methods
Type: Bachelor EE/CS/HMI etc
If you are interested please contact :
Context of the work:
Nowadays, commercial, industrial and residential buildings represent a tremendous amount of the global energy used. Moreover, urbanization and electrification trends show that the total energy demand will increase in the future, and the penetration of energy from renewable sources is increasing as well. Therefore, future smart grids need a system that can monitor, predict, schedule, learn and make decisions regarding local energy consumption and production. In this context estimating the future building electricity consumption based on the past values is a well-known open and challenging problem. The complexity of the consumers energy producing and consuming technologies and the uncertainty in the influencing factors, yield frequent fluctuations. These fluctuations are given by weather conditions, the building construction and thermal properties of the physical materials used, the occupants and their behavior, and so on.
Short description of the assignment:
The electrical demand forecasting problem can be regarded as a nonlinear time series prediction problem depending on many complex factors since it is required at various aggregation levels and at high resolution. To solve this challenging problem, various time series and machine learning approaches have been proposed in the literature. These range from heuristic based approaches to mathematically grounded ones such as those residing in the realm of machine learning [1-3]. Still, the following question remains:
- How to obtain a more accurate energy prediction method?
In this project various energy pattern characteristics may be considered, such as prediction horizon and resolution, level of aggregation, influencing factors. The project aim is to analyze and predict the building energy demand using methods inspired by the latest advances in the artificial intelligence area . An accurate prediction method is expected to bring many benefits at different levels in power system (distributors, producers, traders, brokers and industrial end-users).
 E. Mocanu, P.H. Nguyen, M. Gibescu, W.L. Kling, Deep learning for estimating building energy consumption Sustainable Energy, Grids and Networks. 6, 91–99p., 2016.
 E. Mocanu, E. M. Larsen, P. H. Nguyen, P. Pinson, M. Gibescu, Demand forecasting at low aggregation levels using factored conditional restricted Boltzmann machine, Proceedings of the 19th Power Systems Computation Conference (PSCC), 20-24 June 2016, Genoa.
 Mynhoff, P. A., Mocanu, E., Gibescu, M., Statistical learning versus deep learning: performance comparison for building energy prediction methods, IEEE PES Innovative Smart Grid Technologies Conference Europe, 2018.
 D.C. Mocanu, E. Mocanu, P. Stone, P.H. Nguyen, M. Gibescu, A. Liotta: “Scalable Training of Artificial Neural Networks with Adaptive Sparse Connectivity inspired by Network Science”, Nature Communications, 2018, https://www.nature.com/articles/s41467-018-04316-3