Conference Paper

PEAT, how much am i burning?

Authors:
If you want to read the PDF, try requesting it from the authors.

Abstract

Depletion of fossil fuel and the ever-increasing need for energy in residential and commercial buildings have triggered in-depth research on many energy saving and energy monitoring mechanisms. Currently, users are only aware of their overall energy consumption and its cost in a shared space. Due to the lack of information on individual energy consumption, users are not being able to fine-tune their energy usage. Further, even-splitting of energy cost in shared spaces does not help in creating awareness. With the advent of the Internet of Things (IoT) and wearable devices, apportioning of the total energy consumption of a household to individual occupants can be achieved to create awareness and consequently promoting sustainable energy usage. However, providing personalized energy consumption information in real-time is a challenging task due to the need for collection of fine-grained information at various levels. Particularly, identifying the user(s) utilizing an appliance in a shared space is a hard problem. The reason being, there are no comprehensive means of collecting accurate personalized energy consumption information. In this paper we present the Personalized Energy Apportioning Toolkit (PEAT) to accurately apportion total energy consumption to individual occupants in shared spaces. Apart from performing energy disaggregation, PEAT combines data from IoT devices such as smartphones and smartwatches of occupants to obtain fine-grained information, such as their location and activities. PEAT estimates energy footprint of individuals by modeling the association between the appliances and occupants in the household. We propose several accuracy metrics to study the performance of our toolkit. PEAT was exhaustively evaluated and validated in two multi-occupant households. PEAT achieves 90% energy apportioning accuracy using only the location information of the occupants. Furthermore, the energy apportioning accuracy is around 95% when both location and activity information is available.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Working systems for direct real-time energy feedback, also called energy footprinting, have been demonstrated in [6] in commercial buildings and recently in [5] for the home environment. However, these systems require infrastructure such as energy monitoring devices for HVAC, lighting, and electric appliances, or localization beacons. ...
Conference Paper
Energy footprinting has the potential to raise awareness of energy consumption and lead to energy saving behavior. However, current methods for estimating energy consumption cannot provide fine enough temporal or spatial granularity for a reasonable personal energy footprint estimate. In this work, we present a data-driven system design for estimating personal energy footprint in the city in real-time, even in built environments that do not have existing or accessible energy data or population data.
Article
Full-text available
Personal mobile devices such as smartwatches, smart jewelry, and smart clothes have launched a new trend in the Internet of Things (IoT) era, namely the Internet of Wearable Things (IoWT). These wearables are small IoT devices capable of sensing, storing, processing, and exchanging data to assist users by improving their everyday life tasks through various applications. However, the IoWT has also brought new challenges for the research community to address such as increasing demand for enhanced computational power, better communication capabilities, improved security and privacy features, reduced form factor, minimal weight, and better comfort. Most wearables are battery-powered devices that need to be recharged-therefore, the limited battery life remains the bottleneck leading to the need to enhance the energy efficiency of wearables, thus, becoming an active research area. This paper presents a survey of energy-efficient solutions proposed for diverse IoWT applications by following the systematic literature review method. The available techniques published from 2010 to 2020 are scrutinized, and the taxonomy of the available solutions is presented based on the targeted application area. Moreover, a comprehensive qualitative analysis compares the proposed studies in each application area in terms of their advantages, disadvantages, and main contributions. Furthermore, a list of the most significant performance parameters is provided. A more in-depth discussion of the main techniques to enhance wearables' energy efficiency is presented by highlighting the trade-offs involved. Finally, some potential future research directions are highlighted. INDEX TERMS Wearables, Internet of Wearable Things, energy consumption, wearable applications, energy efficiency, computing, systematic literature review.
Article
Full-text available
Many energy optimizations require fine-grained, load-level energy data collected in real time, most typically by a plug-level energy meter. Online load tracking is the problem of monitoring an individual electrical load’s energy usage in software by analyzing the building’s aggregate smart meter data. Load tracking differs from the well-studied problem of load disaggregation in that it emphasizes per-load accuracy and efficient, online operation rather than accurate disaggregation of every building load via offline analysis. In essence, tracking a particular load creates a virtual power meter for it, which mimics having a networked-connected power meter attached to the load, but notably does not require tracking every other load as well. We propose PowerPlay, a model-driven system for performing accurate, high-performance online load tracking. Our results from applying the system to real-world energy data demonstrate that PowerPlay (i) enables efficient online tracking on low-power embedded platforms, (ii) scales to thousands of loads (across many buildings) on server platforms, and (iii) improves per-load accuracy by more than a factor of two compared to a state-of-the-art load disaggregation algorithm. Our results point to the potential of replacing physical energy meters by “virtual” power meters using a system like PowerPlay.
Article
Full-text available
Sensor-based activity recognition seeks the profound high-level knowledge about human activity from multitudes of low-level sensor readings. Conventional pattern recognition approaches have made tremendous progress in the past years. However, most of those approaches heavily rely on heuristic hand-crafted feature extraction methods, which dramatically hinder their generalization performance. Additionally, those methods often produce unsatisfactory results for unsupervised and incremental learning tasks. Meanwhile, the recent advancement of deep learning makes it possible to perform automatic high-level feature extraction thus achieves promising performance in many areas. Since then, deep learning based methods have been widely adopted for the sensor-based activity recognition tasks. In this paper, we survey and highlight the recent advancement of deep learning approaches for sensor-based activity recognition. Specifically, we summarize existing literatures from three aspects: sensor modality, deep model and application. We also present a detailed discussion and propose grand challenges for future direction.
Conference Paper
Full-text available
Online load tracking is the problem of monitoring an individual electrical load's energy usage by analyzing a building's smart meter data. The problem is important, since many energy optimizations require fine-grained, per-load energy data in real time; it also differs from the well-studied problem of load disaggregation in that it emphasizes efficient, online operation and per-load accuracy, rather than accurate disaggregation of every building load via offline analysis. In essence, tracking a particular load creates a virtual power meter for it, which mimics having a networked-connected power meter attached to it. To enable high performance, we take a model-driven approach that focuses on efficiently detecting a small number of identifiable load features in smart meter data. Our results demonstrate that our system, called PowerPlay, i) enables efficient online tracking on low-power embedded platforms, ii) scales to thousands of loads (across many buildings) on server platforms, and iii) improves per-load accuracy by more than a factor of two compared to a state-of-the-art load disaggregation algorithm.
Article
Full-text available
Multi-label classification methods are increasingly required by modern applications, such as protein function classification, music categorization, and semantic scene classification. This article introduces the task of multi-label classification, organizes the sparse related literature into a structured presentation and performs comparative experimental results of certain multilabel classification methods. It also contributes the definition of concepts for the quantification of the multi-label nature of a data set.
Article
Full-text available
A typical large building contains thousands of sensors, monitoring the HVAC system, lighting, and other operational sub-systems. With the increased push for operational efficiency, operators are relying more on historical data processing to uncover opportunities for energy-savings. However, they are overwhelmed with the deluge of data and seek more efficient ways to identify potential problems. In this paper, we present a new approach called the Strip, Bind and Search (SBS); a method for uncovering abnormal equipment behavior and in-concert usage patterns. SBS uncovers relationships between devices and constructs a model for their usage pattern relative to other devices. It then flags deviations from the model. We run SBS on a set of building sensor traces; each containing hundred sensors reporting data flows over 18 weeks from two separate buildings with fundamentally different infrastructures. We demonstrate that, in many cases, SBS uncovers misbehavior corresponding to inefficient device usage that leads to energy waste. The average waste uncovered is as high as 2500~kWh per device.
Article
Full-text available
Buildings consume a significant portion of the total delivered energy. While the community has been working on monitoring the building energy usage, we argue that an accurate accounting of individual occupants' energy expenditure in real-time is still the missing piece. And, this missing piece makes incentivizing energy reduction a challenge. We take a systematic approach and identify the lack of real-time association between human actions and observed energy consumption as one roadblock. To this end, we first present a model that enables real-time accounting of appliances with delay characteristics. Building on top of this model, we introduce a complete system that fairly attributes energy usages of shared resources, and reveals per-person energy footprint information. And, our system has several ways to analyze and visualize individuals' energy footprint. Finally, we present the deployment experience at an office building.
Conference Paper
Full-text available
Inferring human activity is of interest for various ubiquitous computing applications, particularly if it can be done using ambient information that can be collected non intrusively. In this paper, we explore human activity inference, in the context of energy consumption within a home, where we define an “activity” as the usage of an electrical appliance, its usage duration and its location. We also explore the dimension of identifying the occupant who performed the activity. Our goal is to answer questions such as “Who is watching TV in the Dining Room and during what times?”. This information is particularly important for scenarios such as the apportionment of energy use to individuals in shared settings for better understanding of occupant's energy consumption behavioral patterns. Unfortunately, accurate activity inference in realistic settings is challenging, especially when considering ease of deployment. One of the key differences between our work and prior research in this space is that we seek to combine readily available sensor data (i.e. home level electricity meters and sensors on smartphones carried by the occupants) and metadata information (e.g. appliance power ratings and their location) for activity inference. Our proposed EnergyLens system intelligently fuses electricity meter data with sensors on commodity smartphones -- the Wifi radio and the microphone -- to infer, with high accuracy, which appliance is being used, when its being used, where its being used in the home, and who is using it. EnergyLens exploits easily available metadata to further improve the detection accuracy. Real world experiments show that EnergyLens significantly improves the inference of energy usage activities (average precision= 75.2%, average recall= 77.8%) as compared to traditional approaches that use the meter data only (average precision = 28.4%, average recall = 22.3%).
Conference Paper
Full-text available
Non-intrusive load monitoring, or energy disaggregation, aims to separate household energy consumption data collected from a single point of measurement into appliance-level consumption data. In recent years, the field has rapidly expanded due to increased interest as national deployments of smart meters have begun in many countries. However, empirically comparing disaggregation algorithms is currently virtually impossible. This is due to the different data sets used, the lack of reference implementations of these algorithms and the variety of accuracy metrics employed. To address this challenge, we present the Non-intrusive Load Monitoring Toolkit (NILMTK); an open source toolkit designed specifically to enable the comparison of energy disaggregation algorithms in a reproducible manner. This work is the first research to compare multiple disaggregation approaches across multiple publicly available data sets. Our toolkit includes parsers for a range of existing data sets, a collection of preprocessing algorithms, a set of statistics for describing data sets, two reference benchmark disaggregation algorithms and a suite of accuracy metrics. We demonstrate the range of reproducible analyses which are made possible by our toolkit, including the analysis of six publicly available data sets and the evaluation of both benchmark disaggregation algorithms across such data sets.
Conference Paper
Full-text available
The continuous growth of energy needs and the fact that unpredictable energy demand is mostly served by unsustainable (i.e. fossil-fuel) power generators have given rise to the development of Demand Response (DR) mechanisms for flattening energy demand. Building effective DR mechanisms and user awareness on power consumption can significantly benefit from fine-grained monitoring of user consumption at the appliance level. However, installing and maintaining such a monitoring infrastructure in residential settings can be quite expensive. In this paper, we study the problem of fine-grained appliance power-consumption monitoring based on one house-level meter and few plug-level meters. We explore the trade-off between monitoring accuracy and cost, and exhaustively find the minimum subset of plug-level meters that maximize accuracy. As exhaustive search is time- and resource-consuming, we define a heuristic approach that finds the optimal set of plug-level meters without utilizing any other sets of plug-level meters. Based on experiments with real data, we found that few plug-level meters - when appropriately placed - can very accurately disaggregate the total real power consumption of a residential setting and verified the effectiveness of our heuristic approach.
Article
Full-text available
Apportioning the total energy consumption of a building or organisation to individual users may provide incentives to make reductions. We explore how sensor systems installed in many buildings today can be used to apportion energy consumption between users. We investigate the differences between a number of possible policies to evaluate the case for apportionment based on energy and usage data collected over the course of a year. We also study the additional possibilities offered by more fine-grained data with reference to case studies for specific shared resources, and discuss the potential and challenges for future sensor systems in this area.
Chapter
Full-text available
The development of data-mining applications such as classification and clustering has shown the need for machine learning algorithms to be applied to large scale data. In this paper we present the comparison of different classification techniques using Waikato Environment for Knowledge Analysis or in short, WEKA. WEKA is an open source software which consists of a collection of machine learning algorithms for data mining tasks. The aim of this paper is to investigate the performance of different classification or clustering methods for a set of large data. The algorithm or methods tested are Bayes Network, Radial Basis Function, Pruned Tree, Single Conjunctive Rule Learner and Nearest Neighbors Algorithm. A fundamental review on the selected technique is presented for introduction purposes. The data breast cancer data with a total data of 6291 and a dimension of 699 rows and 9 columns will be used to test and justify the differences between the classification methods or algorithms. Subsequently, the classification technique that has the potential to significantly improve the common or conventional methods will be suggested for use in large scale data, bioinformatics or other general applications.
Article
Full-text available
Reducing domestic energy consumption has a long, varied history, embracing pervasive computing, technology, sociology, and economics. The authors provide an overview and key references to help readers explore this rich field's background more deeply and broadly.
Conference Paper
Full-text available
Sensing, monitoring and actuating systems are expected to play a key role in reducing buildings overall energy consumption. Leveraging sensor systems to support energy efficiency in buildings poses novel research challenges in monitoring space usage, controlling devices, interfacing with smart energy meters and communicating with the energy grid. In the attempt of reducing electricity consumption in buildings, identifying individual sources of energy consumption is key to generate energy awareness and improve efficiency of available energy resources usage. Previous work studied several non-intrusive load monitoring techniques to classify appliances; however, the literature lacks of an comprehensive system that can be easily installed in existing buildings to empower users profiling, benchmarking and recognizing loads in real-time. This has been a major reason holding back the practice adoption of load monitoring techniques. In this paper we present RECAP: RECognition of electrical Appliances and Profiling in real-time. RECAP uses a single wireless energy monitoring sensor easily clipped to the main electrical unit. The energy monitoring unit transmits energy data wirelessly to a local machine for data processing and storage. The RECAP system consists of three parts: (1) Guiding the user for profiling electrical appliances within premises and generating a database of unique appliance signatures; (2) Using those signatures to train an artificial neural network that is then employed to recognize appliance activities (3) Providing a Load descriptor to allow peer appliance benchmarking. RECAP addresses the need of an integrated and intuitive tool to empower building owners with energy awareness. Enabling real-time appliance recognition is a stepping-stone towards reducing energy consumption and allowing a number of major applications including load-shifting techniques, energy expenditure breakdown per appliance, detection of power hungry and faulty appliances, and recogn- - ition of occupant activity. This paper describes the system design and performance evaluation in domestic environment.
Conference Paper
Full-text available
To promote energy-saving behavior, disaggregating electricity usage is critical for increasing consumer awareness of energy usage behavior. This study proposes HeatProbe, a thermal-based power meter system that uses thermal imaging to track disaggregated appliance usage. We have designed, prototyped, and tested the HeatProbe system. Results show that HeatProbe successfully senses individual appliance operating durations with an average error of 125.03 seconds, achieving 80.2% appliance power accounting accuracy in different appliance usage scenarios.
Conference Paper
Full-text available
A growing number of mobile computing applications are cen- tered around the user's location. The notion of location is broad, ranging from physical coordinates (latitude/longitude) to logical labels (like Starbucks, McDonalds). While extensive research has been performed in physical localization, there have been few attempts in recognizing logical locations. This paper argues that the increasing number of sensors on mo- bile phones presents new opportunities for logical localiza- tion. We postulate that ambient sound, light, and color in a place convey a photo-acoustic signature that can be sensed by the phone's camera and microphone. In-built accelerometers in some phones may also be useful in inferring broad classes of user-motion, often dictated by the nature of the place. By combining these optical, acoustic, and motion attributes, it may be feasible to construct an identifiable fingerprint for logical localization. Hence, users in adjacent stores can be separated logically, even when their physical positions are ex- tremely close. We propose SurroundSense, a mobile phone based system that explores logical localization via ambience fingerprinting. Evaluation results from 51 different stores show that SurroundSense can achieve an average accuracy of 87% when all sensing modalities are employed. We be- lieve this is an encouraging result, opening new possibilities in indoor localization.
Article
Full-text available
In this paper we discuss how machine learning, and specifically how naive Bayes classifiers, can be used for user modeling tasks. We argue that in general, machine learning techniques should be used to improve a user modeling system's interactions with users. We further argue that a naive Bayes classifier is a reasonable approach to many user modeling problems, given its advantages of quick learning and low computational overhead. These are critical features for an online user modeling system. We discuss two such user modeling systems and how this technique can be applied to them. Finally, we propose a set of enhancements to naive Bayes classifiers to improve their predictive accuracy, and allow them to better adapt to the user's performance. Our eventual goal is to construct a learning system that requires no intervention from the designer other than a list of potentially useful features. Keywords: Bayes classifiers, machine learning, student modeling 1 Introduction Const...
Conference Paper
We introduce a mobile app for collecting in-the-wild data, including sensor measurements and self-reported labels describing people's behavioral context (e.g., driving, eating, in class, shower). Labeled data is necessary for developing context-recognition systems that serve health monitoring, aging care, and more. Acquiring labels without observers is challenging and previous solutions compromised ecological validity, range of behaviors, or amount of data. Our user interface combines past and near-future self-reporting of combinations of relevant context-labels. We deployed the app on the personal smartphones of 60 users and analyzed quantitative data collected in-the-wild and qualitative user-experience reports. The interface's flexibility was important to gain frequent, detailed labels, support diverse behavioral situations, and engage different users: most preferred reporting their past behavior through a daily journal, but some preferred reporting what they're about to do. We integrated insights from this work back into the app, which we make available to researchers for conducting in-the-wild studies.
Conference Paper
We propose a system that tracks each occupant's personal share of energy use, or "energy footprint", inside commercial building environments, and provides insights to occupants on the real-time energy impact of their actions. We propose a new space-centric policy for fair apportionment of energy in shared environments and demonstrate a method for automatically determining space-centric energy zones. We design and implement ePrints - a system for tracking personalized energy usage in real-time. ePrints supports different apportionment policies, with μs-level footprint computation time and graceful scaling with size of building, frequency of energy updates, and rate of occupant location changes. Finally, we present applications enabled by our system, such as mobile and wearable applications to provide users timely feedback on the energy impacts of their actions, as well as applications to provide energy saving suggestions and inform building-level policies.
Article
We demonstrate that a person's behavioral and environmental context can be automatically recognized by harnessing the sensors built into smartphones and smartwatches. We propose a generic system that can simultaneously recognize many contextual attributes from diverse behavioral domains. By fusing complementary information from different types of sensors our system successfully recognizes fine details of work and leisure activities, body movement, transportation, and more. Health monitoring, clinical intervention, aging care, personal assistance and many more applications will benefit from automatic, frequent and detailed context recognition.
Conference Paper
Current smartphones and smartwatches come equipped with a variety of sensors, from light sensor and inertial sensors to radio interfaces, enabling applications running on these devices to make sense of their surrounding environment. Rather than using sensors independently, combining their sensing capabilities facilitates more interesting and complex applications to emerge (e.g., user activity recognition). But differences between sensors ranging from sampling rate to data generation model (event triggered or continuous sampling) make integration of sensor streams challenging. Here we investigate the opportunity to use deep learning to perform this integration of sensor data from multiple sensors. The intuition is that neural networks can identify nonintuitive features largely from cross-sensor correlations which can result in a more accurate estimation. Initial results with a variant of a Restricted Boltzmann Machine (RBM), show better performance with this new approach compared to classic solutions.
Conference Paper
The use of deep learning for the activity recognition performed by wearables, such as smartwatches, is an understudied problem. To advance current understanding in this area, we perform a smartwatch-centric investigation of activity recognition under one of the most popular deep learning methods - Restricted Boltzmann Machines (RBM). This study includes a variety of typical behavior and context recognition tasks related to smartwatches (such as transportation mode, physical activities and indoor/outdoor detection) to which RBMs have previously never been applied. Our findings indicate that even a relatively simple RBM-based activity recognition pipeline is able to outperform a wide-range of common modeling alternatives for all tested activity classes. However, usage of deep models is also often accompanied by resource consumption that is unacceptably high for constrained devices like watches. Therefore, we complement this result with a study of the overhead of specifically RBM-based activity models on representative smartwatch hardware (the Snapdragon 400 SoC, present in many commercial smartwatches). These results show, contrary to expectation, RBM models for activity recognition have acceptable levels of resource use for smartwatch-class hardware already on the market. Collectively, these two experimental results make a strong case for more widespread adoption of deep learning techniques within smartwatch designs moving forward.
Conference Paper
Providing detailed appliance level energy consumption information may lead consumers to understand their usage behavior and encourage them to optimize the energy usage. Non-intrusive load monitoring (NILM) or energy disaggregation aims to estimate appliance level energy consumption from the aggregate consumption data of households. NILM algorithms, proposed hitherto, are either centralized or do require high performance systems to derive appliance level data, owing to the computational complexity associated. This approach raises several issues related to scalability and privacy of consumer's data. In this paper, we present the Location-aware Energy Disaggregation Framework (LocED) that utilizes occupancy of users to derive accurate appliance level usage information. LocED framework limits the appliances considered for disaggregation based on the current location of occupants. Thus, LocED can provide real-time feedback on appliance level energy consumption and run on an embedded system locally at the household. We propose several accuracy metrics to study the performance of LocED. To test the robustness of LocED, we empirically evaluated it across multiple publicly available datasets. LocED has significantly high energy disaggregation accuracy while exponentially reducing the computational complexity. We also release our comprehensive dataset DRED (Dutch Residential Energy Dataset) for public use, which measures electricity, occupancy and ambient parameters of the household.
Conference Paper
We propose a novel mobile system, called EchoTag, that enables phones to tag and remember indoor locations without requiring any additional sensors or pre-installed infrastructure. The main idea behind EchoTag is to actively generate acoustic signatures by transmitting a sound signal with a phone's speakers and sensing its reflections with the phone's microphones. This active sensing provides finer-grained control of the collected signatures than the widely-used passive sensing. For example, because the sensing signal is controlled by EchoTag, it can be intentionally chosen to enrich the sensed signatures and remove noises from useless reflections. Extensive experiments show that EchoTag distinguishes 11 tags at 1cm resolution with 98% accuracy and maintains 90% accuracy even a week after its training. With this accurate location tagging, one can realize many interesting applications, such as automatically turning on the silent mode of a phone when it is placed at a pre-defined location/area near the bed or streaming favorite songs to speakers if it is placed near a home entertainment system. Most participants of our usability study agree on the usefulness of EchoTag's potential applications and the adequacy of its sensing accuracy for supporting these applications.
Article
Energy auditing and feedback is an effective and low cost technique that has the potential to save 20-50% energy in homes. Several new sensing technologies can now detect and disaggregate energy usage in homes at a fixture level, which is helpful for eco-feedback in homes. However, without disaggregating and assigning fixture energy usage to individuals (fixture assignment problem), it is hard for residents to discover individual energy saving actions. In this paper, we explore the hypothesis that fixture assignment can be performed based on coarse-grained room-level location tracking - even when a fixture is used and multiple people are in the same room. To test this hypothesis, we perform a study with 5 groups of 2 participants each, who lived together for 7-12 days in a test home. We find that fixture assignment can be performed with an average accuracy of 87% using room-level tracking. In comparison, fixture assignment has 12% accuracy with house-level tracking (who is home vs not home) and 97% accuracy with coordinate-level tracking (who is standing at the oven vs fridge).
Chapter
Near Field Communication (NFC), is an integration of Radio Frequency Identification (RFID) technology with mobile devices. NFC offers a quick and convenient method of interaction between humans and NFC enabled devices. Current research concerning NFC appears to mainly focus on development of NFC enabled applications and services. In this paper, we study the performance of NFC devices by considering metrics such as achieved data rates and received power for several distances. Knowledge of these metrics may be useful for application developers to build applications efficiently. We have developed various applications on NFC enabled devices for public transport systems. We also describe the design of 13.56 MHz antenna which was used for measurements of the received power.
Article
Many occupant-oriented smarthome applications such as automated lighting, heating and cooling, and activity recognition need room location information of residents within a building. Surveillance based tracking systems used to track people in commercial buildings, are privacy invasive in homes. In this paper, we present the RF Doormat - a RF threshold system that can accurately track people's room locations by monitoring their movement through the doorways in the home. We also present a set of guidelines and a visualization to easily and rapidly setup the RF-Doormat system on any doorway. To evaluate our system, we perform 580 doorway crossings across 11 different doorways in a home. Results indicate that our system can detect doorway crossings made by people with an average accuracy of 98%. To our knowledge, the RF Doormat is the first highly accurate room location tracking system that can be used for long time periods without the need for privacy invasive cameras.
Article
Buildings account for 40% of US primary energy consumption and 72% of electricity. Of this total, 50% of the energy consumed in buildings is used for Heating Ventilation and Air-Conditioning (HVAC) systems. Current HVAC systems only condition based on static schedules; rooms are conditioned regardless of occupancy. By conditioning rooms only when necessary, greater efficiency can be achieved. This paper describes POEM, a complete closed-loop system for optimally controlling HVAC systems in buildings based on actual occupancy levels. POEM is comprised of multiple parts. A wireless network of cameras called OPTNet is developed that functions as an optical turnstile to measure area/zone occupancies. Another wireless sensor network of passive infrared (PIR) sensors called BONet functions alongside OPTNet. This sensed occupancy data from both systems are then fused with an occupancy prediction model using a particle filter in order to determine the most accurate current occupancy in each zone in the building. Finally, the information from occupancy prediction models and current occupancy is combined in order to find the optimal conditioning strategy required to reach target temperatures and minimize ventilation requirements. Based on live tests of the system, we estimate ~30.0% energy saving can be achieved while still maintaining thermal comfort.
Article
Energy and sustainability issues raise a large number of problems that can be tackled using approaches from data mining and machine learning, but traction of such problems has been slow due to the lack of publicly available data. In this paper we present the Reference Energy Disaggregation Data Set (REDD), a freely available data set containing de-tailed power usage information from several homes, which is aimed at furthering research on energy disaggregation (the task of determining the component appliance contributions from an aggregated electricity signal). We discuss past ap-proaches to disaggregation and how they have influenced our design choices in collecting data, we describe the hardware and software setups for the data collection, and we present initial benchmark disaggregation results using a well-known Factorial Hidden Markov Model (FHMM) technique.
Conference Paper
The monitoring of electrical and water fixtures in the home is being applied for a variety of "smart home" applications, such as recognizing activities of daily living (ADLs) or conserving energy or water usage. Fixture monitoring techniques generally fall into two categories: fixture recognition and fixture disaggregation. However, existing techniques require users to explicitly identify each individual fixture, either by placing a sensor on it or by manually creating training data for it. In this paper, we present a new fixture discovery system that automatically infers the existence of electrical and water fixtures in the home. We call the system FixtureFinder. The basic idea is to use data fusion between the smart meters and other sensors or infrastructure already in the home, such as the home security or automation system, and to find repeating patterns in the fused data stream. To evaluate FixtureFinder, we deployed the system into 4 different homes for 7-10 days of data collection. Our results show that FixtureFinder is able to identify and differentiate major light and water fixtures in less than 10 days, including multiple copies of light bulbs and sinks that have identical power/water profiles.
Conference Paper
We explore the overlooked area of personal energy consumption in the context of a shared domestic household. We discuss the potential benefits of such an approach. We report the results of a lab study and field trial with four households using a personal energy monitoring system. We describe the results of the studies and discuss how such previously hidden information might raise awareness of individual energy consumption and the benefits and problems this entails.
Article
Buildings account for a relevant fraction of the energy consumed by a country, up to 20-40% of the yearly energy consumption. If only electricity is considered, the fraction is even bigger, reaching around 73% of the total electricity consumption, equally divided into residential and commercial dwellings. Building and Home Automation have a potential to profoundly impact current and future buildings’ energy efficiency by informing users about their current consumption patterns, by suggesting more efficient behaviors, and by pro-actively changing/modifying user actions for reducing the associated energy wastes. In this paper we investigate the capability of an automated home to automatically, and timely, inform users about energy consumption, by harvesting opinions of residential inhabitants on energy feedback interfaces. We report here the results of an on-line survey, involving nearly a thousand participants, about feedback mechanisms suggested by the research community, with the goal of understanding what feedback is felt by home inhabitants easier to understand, more likely to be used, and more effective in promoting behavior changes. Contextually, we also collect and distill users’ attitude towards in-home energy displays and their preferred locations, gathering useful insights on user-driven design of more effective in-home energy displays.
Article
Radio Frequency (RF) fingerprinting, based onWiFi or cellular signals, has been a popular approach to indoor localization. However, its adoption in the real world has been stymied by the need for sitespecific calibration, i.e., the creation of a training data set comprising WiFi measurements at known locations in the space of interest. While efforts have been made to reduce this calibration effort using modeling, the need for measurements from known locations still remains a bottleneck. In this paper, we present Zee -- a system that makes the calibration zero-effort, by enabling training data to be crowdsourced without any explicit effort on the part of users. Zee leverages the inertial sensors (e.g., accelerometer, compass, gyroscope) present in the mobile devices such as smartphones carried by users, to track them as they traverse an indoor environment, while simultaneously performing WiFi scans. Zee is designed to run in the background on a device without requiring any explicit user participation. The only site-specific input that Zee depends on is a map showing the pathways (e.g., hallways) and barriers (e.g., walls). A significant challenge that Zee surmounts is to track users without any a priori, user-specific knowledge such as the user's initial location, stride-length, or phone placement. Zee employs a suite of novel techniques to infer location over time: (a) placement-independent step counting and orientation estimation, (b) augmented particle filtering to simultaneously estimate location and user-specific walk characteristics such as the stride length,(c) back propagation to go back and improve the accuracy of ocalization in the past, and (d) WiFi-based particle initialization to enable faster convergence. We present an evaluation of Zee in a large office building.
Article
Nowadays, multi-label classification methods are increasingly required by modern applications, such as protein function classification, music categorization and semantic scene classification. This paper introduces the task of multi-label classification, organizes the sparse related literature into a structured presentation and performs comparative experimental results of certain multi-label classification methods. It also contributes the definition of concepts for the quantification of the multi-label nature of a data set.
Article
When agents are ascribed selfish motives, economic theory points to grave inefficiencies resulting from externalities. We study a restaurant setting in which groups of diners are faced with different ways of paying the bill. The two main manipulations are splitting the bill between the diners and having each pay individually. We find that subjects consume more when the cost is split, resulting in a substantial loss of efficiency. Diners prefer the individual pay to the inefficient split-bill method. When forced to play according to a less preferred set of rules, they minimise their individual losses by taking advantage of others. Copyright 2004 Royal Economic Society.
Article
A nonintrusive appliance load monitor that determines the energy consumption of individual appliances turning on and off in an electric load, based on detailed analysis of the current and voltage of the total load, as measured at the interface to the power source is described. The theory and current practice of nonintrusive appliance load monitoring are discussed, including goals, applications, load models, appliance signatures, algorithms, prototypes field-test results, current research directions, and the advantages and disadvantages of this approach relative to intrusive monitoring
Personalized Energy Services: A Data-Driven Methodology towards Sustainable
  • S N A U Nambi
S. N. A. U. Nambi, "Personalized Energy Services: A Data-Driven Methodology towards Sustainable, Smart Energy Systems", Delft University of Technology, 2017.
EchoTag: Accurate Infrastructure-Free Indoor Location Tagging with Smartphones
  • Y Tung
  • K Shin
Y. Tung and K. Shin, "EchoTag: Accurate Infrastructure-Free Indoor Location Tagging with Smartphones", In Proc. of ACM Mobicom, 2015.
Comparison of different classification techniques using WEKA for breast cancer
  • B Othman
  • M Fauzi
  • T Moh Shan
  • Yau
B. Othman, M. Fauzi, and T. Moh Shan Yau, "Comparison of different classification techniques using WEKA for breast cancer", In Proc. of 3rd Kuala Lumpur International Conference on Biomedical Engineering, Springer, Berlin, 2007.
Naive bayes classifiers " University of British Columbia 2006. K. Murphy "Naive bayes classifiers
  • K Murphy