Current Projects

Day4_small

Performance Evaluation of Event-Based Non-Intrusive Load Monitoring Approaches

The ability of correctly evaluating the learned models is one of the most crucial aspects of machine learning. On the one hand, it is necessary to assess that one model is better than another, whereas on the other hand it is of vital importance to understand which metric should be used to evaluate a particular learned model. In this project,

Building DR

Estimating Electricity Demand Flexibility in Buildings for Energy Efficiency and Demand Response Applications

The continuous growth of electricity usage and number of electricity-powered devices results in exceeding the capacity of local grids. To lower the electricity usage during peak demand periods, utility providers encourage building owners to reduce electricity consumption by incentivizing their demand curtailment through demand response (DR) programs. One of the most essential components to utilize such DR opportunities is a

Data-Driven Optimization of Heating, Ventilation and Air Conditioning Systems

Heating, Ventilation and Air conditioning (HVAC) systems account for more than half of the energy consumption in buildings. The HVAC system provides thermal and ventilation services to support indoor thermal comfort and air quality requirements. Because of the large scale of energy consumed by the HVAC systems, enormous amount of energy could be wasted if the system is running inefficiently

Inferring Sensor Meta-Data to Create Portable Software Applications for Buildings

In modern buildings, thousands of sensors and instruments need to be installed and configured. However, manual work to record meta data information(e.g., location, functionality, sensor type and etc.) of those devices is time-consuming and error-prone. Yet such information is often required by software and hardware in building infrastructure system. The purpose of this project is to automatically infer such meta

Data-Driven Models of Infrastructure Interdependencies

Infrastructure systems (e.g. water distribution, energy, transportation, emergency services, telecommunication) are usually modeled as independent operating systems, including interactions only among components of a single system. However, most systems interact with others, and an integrated framework is crucial to accurately capture their performance. For instance, the functioning of the water distribution network relies on the availability of energy while, in

sensor_andrew

Sensor Andrew: Large-Scale Sensing and Actuation

Sensor Andrew is an experiment on the interoperability and semantic capabilities of cyber-physical systems: a virtual instrument for large-scale sensing and actuation, intended to help deliver sensor-driven decision support for our built environment.