Rishee K. Jain

Big Data + Big Cities: Graph Signals of Urban Air Pollution [Exploratory SP]

In this article, we apply signal processing and data science methodologies to study the environmental impact of burning different types of heating oil in New York City, where currently the burning of heavy fuel oil in buildings produces more annual black carbon, a key component of PM2.5, emissions, than all cars and trucks combined. The data utilized in this article are collected through New York Citys Local Law 84 (LL84) energy disclosure mandate. The mandate requires annual energy consumption reporting for large buildings (i.e., approximately greater than 50,000 gross feet) of all use types. This analysis utilizes actual heating oil consumption data for calendar year 2012. The LL84 data set was merged with land use and geographic data at the tax lot level from the Primary Land Use Tax Lot Output (PLUTO) data set from the New York City Department of City Planning. The PLUTO data set provides building and tax lot characteristics, as well as their geographic location.

Web-based eco-feedback visualization of building occupant energy consumption in support of quantifying social network effects on conservation

Exposing building occupants to information about their energy use and the energy use of others in their social network through eco-feedback systems has been shown to significantly impact occupant energy consumption. In this paper, we describe the design and development of a web-based visualization system that exposes building occupants to real-time information about their individual energy utilization, the utilization of their peers in the building, and the average energy utilization of all monitored individuals in the building. The system also monitors and records user interaction with the system and other users of the system providing relevant usage data for conducting analysis to quantify network effects on user energy consumption. A description of the system’s physical and virtual architecture and how its design enables meaningful analysis, visualization and effective user monitoring is provided. We conclude by presenting the methods and challenges related to how our visualization system was developed to facilitate and monitor social interaction around energy conservation, and how it enables research into the underlying mechanisms that drive these actions.

A data-driven design framework for urban slum housing: Case of Mumbai: Poster Abstract

The current building assessment tools are limited to building performance analysis with respect to box models derived from the urban morphology of developed countries. Complex sociotechnical issues associated with rapidly urbanizing cities like Mumbai are often missed. Here, we forward a conceptual framework for designing slum habitation adheres to norms of energy, health and environmental sustainability. This can enable in designing slum rehabilitation projects such that they are not only energy efficient, but are also acceptable to the occupants. This conceptual framework attempts to bridge the missing link currently existing in the early design stages of slum rehabilitation projects. The proof of the concept is a work currently in progress, hence, here, we only elaborate on the conceptual framework exemplified through three cases of slum rehabilitation houses in Mumbai.

OESPG: Computational Framework for Multidimensional Analysis of Occupant Energy Use Data in Commercial Buildings

AbstractCommercial buildings account for much of the energy use both in the United States and globally. The role of occupant behavior within the physical building has been found to be an important ...

Data-driven benchmarking of building energy performance at the city scale

Cities across the country (20 to date) are rapidly passing laws to mandate the collection and disclosure of energy usage data with the hopes that such data could be utilized to benchmark building energy performance and provide a basis for designing and deploying efficiency measures. However, numerous municipalities are struggling to translate such data into actionable insights on which buildings are the best candidates for energy efficiency interventions. Current studies are limited in their ability to compare energy efficiency of buildings at the city scale and maintain interpretability necessary to result in effective decision making that can facilitate municipal policy and program design. In this paper, we present an integrated data-driven method based on recursive partitioning and stochastic frontier analysis to benchmark energy performance of building stock across an entire city. We implement the proposed method on a dataset of over 10,000 buildings in New York City. The preliminary results begin to quantify the potential for building energy efficiency and give further support to city officials for examining effects of possible interventions. We aim to establish a robust methodology that provides a standard way to benchmark building energy performance at the city scale, which can be easily applied to other cities across the country.

Development of a Novel BIM-Energy Use Ontology

The built environment provides a significant opportunity to reduce energy consumption and the associated environmental impacts. Researchers have focused on how Building Information Models (BIM) can be effectively used during the design phase to model energy efficient retrofits. However, research on the use of BIM during the operational phase to aid in energy conservation has been limited. The research presented in this paper focuses on the integration of real-time energy consumption data of building occupants into the spatial context through BIM. Real-time energy consumption data was collected and a standardized Industry Foundation Class (IFC) BIM with geometric, material and construction information was created for a test-bed building in New York City. To link data in the IFC with energy consumption data we developed a novel ontology that merges the existing IFC class hierarchy with newly defined energy use classes using relationship classes. The process entails the identification of the specific geometric information in a BIM that corresponds to energy usage.

Data-Driven, Multi-metric, and Time-Varying (DMT) Building Energy Benchmarking Using Smart Meter Data

New and emerging data streams, from public databases to smart meter infrastructure, contain valuable information that presents an opportunity to develop more robust data-driven models for benchmarking energy use in buildings. In this paper, we propose a new Data-driven, Multi-metric, and Time-varying (DMT) energy benchmarking framework that utilizes these new data streams to benchmark building energy use across multiple metrics at the daily time scale. High fidelity data from smart meters enables the DMT benchmarking framework to produce daily benchmarking scores and use daily weather data to understand seasonally adjusted performance. Intra-day building efficiency is also investigated by benchmarking buildings across several metrics (e.g., total energy usage, operational energy usage, non-operational energy usage) thereby enabling deeper insights into building operations than traditional yearly benchmarking models. By using quantile regression modeling, the DMT framework can differentiate and understand the main drivers of energy consumption between low and high performing buildings and between building operational states. To illustrate the insights that can be gleaned from the proposed DMT framework, we apply the framework to understand building performance for over 500 schools throughout the state of California. The DMT framework provided insights into how various drivers impacted energy usage for both high and low performing buildings, and results indicated that schools had consistent drivers of energy usage. Overall the DMT framework was designed to be highly interpretable such that it could help bridge the gap between data science and engineering methods thus enabling better decision-making in respect to energy efficiency.

Understanding building occupant activities at scale: An integrated knowledge-based and data-driven approach

Abstract Buildings are our homes and our workplaces. They directly affect our well-being, and they impact the natural global environment primarily through the energy they consume. Understanding the behavior of occupants in buildings has vital implications for improving the energy efficiency of building systems and for providing knowledge to designers about how occupants will utilize the spaces they create. However, current methods for inferring building occupant activity patterns are limited in two primary areas: First, they lack adaptability to new spaces and scalability to larger spaces due to the time and cost intensity of collecting ground truth data for training the embedded algorithms. Second, they do not incorporate explicit knowledge about occupant dynamics in their implementation, limiting their ability to uncover deep insights about activity patterns in the data. In this paper, we develop a methodology for classifying occupant activity patterns from plug load sensor data at the desk level. Our method makes us of a common unsupervised learning algorithm—the Gaussian mixture model—and, in addition, it incorporates explicit knowledge about occupant presence and absence in order to preserve adaptability and effectiveness. We validate our method using a pilot study in an academic office building and demonstrate its potential for scalability through a case study of an open-office building in San Francisco, CA. Our method offers key insights into spatially and temporally granular occupancy states and space utilization that could not otherwise be obtained.

Financial Viability of Residential Photovoltaic and Battery Systems in Californias

Using a battery on a household level has become easier after the launch of Tesla’s Powerwall. Storing electricity during daytime’s PV overproduction or charging the battery during night with an attractive tariff are the most prominent applications. This paper explores the economic impact of the usage of residential battery storage combined with solar photovoltaics (PV) based on real load data from Northern California, USA. A data-driven, deterministic model to benchmark electricity cost savings for single households is presented and the financial viability of such systems is scrutinized for California. Our results indicate that under current capacity and price points, battery systems have limited financial viability and have a payback period exceeding 20 years in most cases. We deepen our analysis and compare the results of our deterministic model to that of a stochastic model to demonstrate that for an hourly time resolution the deterministic model provides an adequate benchmark for estimating cost (within 3%) savings with a short (1/60th) computation time.

A Data Integration Framework for Urban Systems Analysis Based on Geo-Relationship Learning

The world is rapidly urbanizing, and for the first time in history over 50% of the world’s population reside in urban areas. This rapid urbanization brings about tremendous challenges at the intersection of governance, infrastructure and the environment. Advanced sensing and data analytics techniques have been developed in the context of so called “smart cities” with the goal of providing insights on how urban systems could be designed and managed more effectively. However, the proliferation of data from heterogeneous sources makes interoperability and mining of such urban data streams difficult. Facilitating the extraction of insights that support data-informed policymaking and program recommendations will require frameworks to integrate such heterogeneous data streams. In this paper, we introduce a novel data integration framework that utilizes a RDF (Resource Description Framework) model to integrate disparate urban data streams based on geo-relationships that are iteratively learned from semantic information and the structure of relational databases. The development of our framework was driven by interviews and observations of city officials responsible for managing and integrating urban data and a review of the various types of disparate datasets generated from sources like departmental databases, sensors, and crowdsourcing. Finally, we apply our proposed framework to an urban data scenario in order to demonstrate the applicability and usefulness of the framework.