@conference {31839, title = {Comparison of MPC Formulations for Building Control under Commercial Time-of-Use Tariffs}, booktitle = {IEEE PowerTech Milan 2019}, year = {2019}, month = {2019}, abstract = {

Most medium and large commercial buildings in\ the U.S. are subject to complex electricity tariffs that combine\ both Time-of-Use (TOU) energy and demand charges. This study\ analyses the performances of different economic Model Predictive\ Control (MPC) formulations, from the standpoints of monthly bill\ reduction, load shifting, and peak demand reduction. Simulations\ are performed on many simplified commercial building models,\ with multiple TOU demand charges, and under various summer\ conditions. Results show that compared to energy-only MPC, the\ traditional method for dealing with demand charges significantly
reduces peak demand and owner bill, however, highlight a lack\ of load shifting capability. A proposed incremental approach
is presented, which better balances the bill components in the\ objective function. In the case study presented, this method
can improve monthly bill savings and increase load shifting\ during demand response events, while keeping a similarly low
peak demand, compared to traditional MPC methods taking into\ account demand charges.

}, keywords = {commercial building, demand charge, Model predictive control (MPC), peak demand, time-of-use tarrif}, author = {Olivier Van Cutsem and Maher Kayal and David Blum and Marco Pritoni} } @article {31671, title = {Inferring occupant counts from Wi-Fi data in buildings through machine learning}, journal = {Building and Environment}, volume = {158}, year = {2019}, month = {05/2019}, pages = {281 - 294}, abstract = {

An important approach to curtail building energy consumption is to optimize building control based on occupancy information. Various studies proposed to estimate occupant counts through different approaches and sensors. However, high cost and privacy concerns remain as major barriers, restricting the practice of occupant count detection. In this study, we propose a novel method utilizing data from widely deployed Wi-Fi infrastructure to infer occupant counts through machine learning. Compared with the current indirect measurement methods, our method improves the performance of estimating people count: (1) we avoid privacy concerns by anonymizing and reshuffling the MAC addresses on a daily basis; (2) we adopted a heuristic feature engineer approach to cluster connected devices into different types based on their daily connection duration. We tested the method in an office building located in California. In an area with an average occupancy of 22{\textendash}27 people and a peak occupancy of 48{\textendash}74 people, the root square mean error on the test set is less than four people. The error is within two people counts for more than 70\% of estimations, and less than six counts for more than 90\% of estimations, indicating a relatively high accuracy. The major contribution of this study is proposing a novel and accurate approach to detect occupant counts in a non-intrusive way, i.e., utilizing existing Wi-Fi infrastructure in buildings without requiring the installation of extra hardware or sensors. The method we proposed is generic and could be applied to other commercial buildings to infer occupant counts for energy efficient building control.

}, keywords = {Building control, Machine learning, Occupancy estimation, Occupant count, Random forest, Wi-Fi data}, issn = {03601323}, doi = {10.1016/j.buildenv.2019.05.015}, url = {https://linkinghub.elsevier.com/retrieve/pii/S0360132319303336}, author = {Zhe Wang and Tianzhen Hong and Mary Ann Piette and Marco Pritoni} }