Anomaly Detection and Smart Control of Buildings
Brief Description of Project:
We have been working with sensor data collected from buildings on the Vanderbilt campus to develop anomaly detection methods and reinforcement learning methods to improve overall energy consumption in buildings without sacrificing occupant comfort.
The undergraduate intern will work closely with the faculty mentor and other graduate students to develop and apply machine learning algorithms for fault detection and for optimal control of building energy consumption.
CS 3250, 3251;
CS 4260 preffered.
Nature of Supervision:
The student will work closely with the faculty supervisor in the various phases of the project as outlined in the Research Plan below. The student will also have the opportunity to work with graduate students pursuing similar interests.
A Brief Research Plan (period is for 10 weeks):
1. Understanding problem and conduct literature review on anomaly detection and reinforcement learning methods.
2. Acquiring and organizing data from multiple sources
3, Data cleaning and pre-processing
4.Study anomaly detection and reinforcement learning algorithms
5. Iterative process of running algorithms, interpreting results and iterating to derive better solutions.
6. Write report and generate poster.
Number of Open Slots: 1
Name: Gautam Biswas