Machine Learning, ANN

edit

For a broad review of ML methods applied to buildings, see the literature review by Foucquier et al., "State of the art in building modelling and energy performances prediction: A review".

For calibration focused work, see literature review done as part of ASHRAE Research Project 1051-RP "Procedures for Reconciling Computer-Calculated Results With Measured Energy Data" and the literature review by D. Coakley et al., "A review of methods to match building energy simulation models to measured data".

For M&V focused work, see ASHRAE Research Project 1040-RP "Measurement, Modeling, Analysis and Reporting Protocols for Short-term M&V of Whole Building Energy Performance". Berkeley Lab's Building Energy Information Systems research portfolio and recent publications has M&V method reviews, especially the 2016 paper "Accuracy of Automated Measurement and Verification (M&V) Techniques for Energy Savings in Commercial Buildings".

Clayton Miller's BUDS lab focuses on data-mining and ML for buildings. See his older website Data Driven Buildings.

Also look for the search terms "Non-Intrusive Load Monitoring", "Inverse Modeling", and "Disaggregation" for ML approaches in buildings.