Search

The data presented here were collected independently for 6 real buildings by researchers of different institutions and gathered in the context of the IEA EBC Annex 81 Data-driven Smart Buildings, as a joint effort to compile a diverse range of datasets suitable for advanced control applications of indoor climate and energy use in buildings. The data were acquired by energy meters, both consumption and PV generation, and sensors of technical installation and indoor climate variables, such as temperature, flow rate, relative humidity, CO2 level, illuminance. Weather variables were either acquired by local sensors or obtained from a close by meteorological station. The data were collected either during normal operation of the building, with observation periods between 2 weeks and 2 months, or during experiments designed to excite the thermal mass of the building, with observation periods of approximately one week. The data have a time resolution varying between 1 min and 15 min; in some case the highest resolution data are also averaged at larger intervals, up to 30 min.

In recent years, large pretrained models have been used in dialogue systems to improve successful task completion rates. However, lack of reasoning capabilities of dialogue platforms make it difficult to provide relevant and fluent responses, unless the designers of a conversational experience spend a considerable amount of time implementing these capabilities in external rule based modules. In this work, we propose a novel method to fine-tune pretrained transformer models such as Roberta and T5. to reason over a set of facts in a given dialogue context. Our method includes a synthetic data generation mechanism which helps the model learn logical relations, such as comparison between list of numerical values, inverse relations (and negation), inclusion and exclusion for categorical attributes, and application of a combination of attributes over both numerical and categorical values, and spoken form for numerical values, without need for additional training dataset. We show that the transformer based model can perform logical reasoning to answer questions when the dialogue context contains all the required information, otherwise it is able to extract appropriate constraints to pass to downstream components (e.g. a knowledge base) when partial information is available. We observe that transformer based models such as UnifiedQA-T5 can be fine-tuned to perform logical reasoning (such as numerical and categorical attributes' comparison) over attributes that been seen in training time (e.g., accuracy of 90\%+ for comparison of smaller than $k_{\max}$=5 values over heldout test dataset).

Author(s): Wetter, M; Ehrlich, P; Gautier, A; Grahovac, M; Haves, P; Hu, J; Prakash, A; Robin, D; Zhang, K | Abstract: The current process for specifying, installing and commissioning building control sequences is largely manual and based on ambiguous natural language specifications. It lacks a formal end-to-end quality control and it has been shown not to deliver high performance sequences at scale. While high-performance HVAC control sequences enable significant reductions in energy consumption, errors in implementing the control logic are common even for less advanced sequences. To improve this situation, we present a digitized building control delivery workflow with formal end-to-end verification, a Control Description Language for the digital specification of building control sequences within this workflow, and software tools that enable digitization of this process. Using the process and tools introduced here, mechanical designers can customize, test and improve these sequences within annual energy simulation, store them in a library for use in other projects, and export them for bidding. Control providers can implement the sequences on existing control product lines through code generation. Commissioning providers can formally verify whether as-installed sequences conform to the digital design specification that was exported by the mechanical designer. Moreover, control product development teams can use the reference implementations of these libraries within their product testing to ensure that their products reproduce the behavior of the reference implementations. This paper presents this process, the language and the supporting software, together with examples of all of the above steps. The presented work has given rise to a new proposed standard, ASHRAE 231P, that will allow digitizing the building control delivery process through the standardization of a control-vendor independent format for exchanging control logic that we pioneered through the here presented work.

Semantic metadata standards pave the way for interoperability by providing building operators and application developers with common schemes to describe building resources. Applications can query building metadata models to retrieve the set of entities and relationships they need to operate, instead of hard-coding references to specific points and objects from the underlying data sources. Currently, querying such models requires the developer to be very specific when formulating queries in order to obtain meaningful answers (or any answer). The developer is inevitably expected to be familiar with the systems and components of the buildings being queried, as well as the schema used to represent them. The variety of buildings - both in the composition of their subsystems and in how they happen to be modeled - means that the developer will need to use multiple queries in order to retrieve necessary results. This is complex, time-consuming and error-prone. To address this limitation, we investigate query relaxation as a technique to facilitate discovery of meaningful building resources in a collection of ontology-based buildings data. We evaluate our query relaxation approach over a set of Brick models and demonstrate its use in the context of real-world building applications., BuildSys '21: Proceedings of the 8th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation, ISBN:978-1-4503-9114-6

Many different digital representations of a building are produced over the course of its lifecycle. While these representations individually contain metadata required to support different stages of the building's lifecycle, they are largely not interoperable due to differences in structure, syntax and semantics. This impedes the development and deployment of data-driven applications providing fault detection and diagnosis, virtual metering or optimal control. [4] introduces a new platform for the continuous curation of a unified, Brick [1]-based metadata model that can be maintained throughout the building lifecycle. In this demonstration, we present a live proof-of-concept implementation of the platform with support for several metadata representations in the context of a simulated building lifecycle.

Many different digital representations of a building are produced over the course of its lifecycle. These representations contain the metadata required to support different stages of the building, from initial planning and design, to construction and commissioning, through operations, audits, retrofits and maintenance. However, because of differences in the semantics, structure and syntax of these representations, the metadata they contain is not interoperable. We present a novel method for leveraging these representations to create a unified, authoritative Brick metadata model for a building that can be continually maintained over the course of the building lifecycle. A simple synchronization protocol relays inferred Brick metadata from existing metadata sources such as gbXML, BuildingSync, Project Haystack and Modelica to a central integration server, which merges the metadata into a valid Brick model.

As deep reinforcement learning (DRL) continues to gain interest in the smart building research community, there is a transition from simulation-based evaluations to deploying DRL control strategies in actual buildings. While the efficacy of a solution could depend on a particular implementation, there are common obstacles that developers have to overcome to deliver an effective controller. Additionally, a deployment in a physical building can invalidate some of the assumptions made during the controller development. Assumptions on the sensor placement or on the equipment behavior can quickly come undone. This paper presents some of the significant assumptions made during the development of DRL based controllers that could affect their operations in a physical building. Furthermore, a preliminary evaluation revealed that controllers developed with some of these assumptions can incur twice the expected costs when they are deployed in a building.

With the falling costs of solar arrays and battery storage and reduced reliability of the grid due to natural disasters, small-scale local generation and storage resources are beginning to proliferate. However, very few software options exist for integrated control of building loads, batteries and other distributed energy resources. The available software solutions on the market can force customers to adopt one particular ecosystem of products, thus limiting consumer choice, and are often incapable of operating independently of the grid during blackouts. In this paper, we present the &ldquo, Solar+ Optimizer&rdquo, (SPO), a control platform that provides demand flexibility, resiliency and reduced utility bills, built using open-source software. SPO employs Model Predictive Control (MPC) to produce real time optimal control strategies for the building loads and the distributed energy resources on site. SPO is designed to be vendor-agnostic, protocol-independent and resilient to loss of wide-area network connectivity. The software was evaluated in a real convenience store in northern California with on-site solar generation, battery storage and control of HVAC and commercial refrigeration loads. Preliminary tests showed price responsiveness of the building and cost savings of more than 10% in energy costs alone.

There is growing interest in city and community scale energy efficiency goals across the US and around the world [1]. Forward thinking developers are expressing interest in developing methods to track the actual operations of new urban developments and campuses. Local governments and developers are exploring how to ensure that the built infrastructure, including buildings, transportation, energy generation and storage assets, is not only designed for energy efficiency, but operates efficiently as well. There is a need for city and district scale tools to support energy analysis from design through operations. Smart city data analytics is a growing area of interest in the smart city technology space. However, while there is enthusiasm for the concept of smart city energy analytics, there is a lack of common semantic data models, interoperable systems, and methods to collect data and support decision-making for operating multi-building city- and district-scale energy systems. Urban developers are asking broad questions about how such a system might be designed, built, valued, maintained, and optimized. This paper describes ongoing research to explore and define a City or District Energy Operations System (CityEOS).

Occupancy detection, tracking, and estimation has a wide range of applications including improving building energy efficiency, safety, and security of the occupants. As depth sensors are getting cheaper, they offer a viable solution to estimate occupancy accurately in a non-privacy invasive manner. Even though there are publicly available depth datasets, they do not consider placing the sensor in the ceiling looking downwards to estimate occupancy. We deployed four Kinect for XBOX One in four CMU classrooms and conference rooms for a period of four weeks in 2017 and collected over 6 TB of depth data. We annotate this huge dataset by labelling bounding boxes around occupants and release the annotated dataset.

Behind-the-meter distributed energy resources (DERs), including building solar photovoltaic (PV) technology and electric battery storage, are increasingly being considered as solutions to support carbon reduction goals and increase grid reliability and resiliency. However, dynamic control of these resources in concert with traditional building loads, to effect efficiency and demand flexibility, is not yet commonplace in commercial control products. Traditional rule-based control algorithms do not offer integrated closed-loop control to optimize across systems, and most often, PV and battery systems are operated for energy arbitrage and demand charge management, and not for the provision of grid services. More advanced control approaches, such as MPC control have not been widely adopted in industry because they require significant expertise to develop and deploy. Recent advances in deep reinforcement learning (DRL) offer a promising option to optimize the operation of DER systems and building loads with reduced setup effort. However, there are limited studies that evaluate the efficacy of these methods to control multiple building subsystems simultaneously. Additionally, most of the research has been conducted in simulated environments as opposed to real buildings. This paper proposes a DRL approach that uses a deep deterministic policy gradient algorithm for integrated control of HVAC and electric battery storage systems in the presence of on-site PV generation. The DRL algorithm, trained on synthetic data, was deployed in a physical test building and evaluated against a baseline that uses the current best-in-class rule-based control strategies. Performance in delivering energy efficiency, load shift, and load shed was tested using price-based signals. The results showed that the DRL-based controller can produce cost savings of up to 39.6% as compared to the baseline controller, while maintaining similar thermal comfort in the building. The project team has also integrated the simulation components developed during this work as an OpenAIGym environment and made it publicly available so that prospective DRL researchers can leverage this environment to evaluate alternate DRL algorithms.

Endovascular treatment of arterial diseases has become first-line in most cases due to improved technology. However, until recently, excessive atherosclerotic calcification has been a major limiting factor in the endovascular management of peripheral arterial disease, as well as vascular access for endovascular aneurysm repair (EVAR) and transcatheter aortic valve replacement (TAVR). The Peripheral Intravascular Lithotripsy (IVL) System (Shockwave Medical, Inc., Fremont California) applies pulsatile mechanical energy under fluoroscopic guidance to disrupt calcified lesions. The purpose of this paper is to introduce IVL in the treatment of calcific access vessels in preparation for EVAR and TAVR, as well as peripheral arterial disease applications to enhance luminal gain. Using the IVL System, angioplasty can be performed with lower pressures, which may minimize arterial dissection. Further, the lithotripsy effect on calcium will enhance vessel compliance. We describe several cases where IVL was applied successfully and present additional cases that may have benefitted from the use of this technology.

Additional file 1: Table S1. Characteristics of peripheral artery disease in study participants. Table S2. Results of ABI, TBI and LBF in study participants.

As convolutional neural networks (CNNs) enable state-of-the-art computer vision applications, their high energy consumption has emerged as a key impediment to their deployment on embedded and mobile devices. Towards efficient image classification under hardware constraints, prior work has proposed adaptive CNNs, i.e., systems of networks with different accuracy and computation characteristics, where a selection scheme adaptively selects the network to be evaluated for each input image. While previous efforts have investigated different network selection schemes, we find that they do not necessarily result in energy savings when deployed on mobile systems. The key limitation of existing methods is that they learn only how data should be processed among the CNNs and not the network architectures, with each network being treated as a blackbox. To address this limitation, we pursue a more powerful design paradigm where the architecture settings of the CNNs are treated as hyper-parameters to be globally optimized. We cast the design of adaptive CNNs as a hyper-parameter optimization problem with respect to energy, accuracy, and communication constraints imposed by the mobile device. To efficiently solve this problem, we adapt Bayesian optimization to the properties of the design space, reaching near-optimal configurations in few tens of function evaluations. Our method reduces the energy consumed for image classification on a mobile device by up to 6x, compared to the best previously published work that uses CNNs as blackboxes. Finally, we evaluate two image classification practices, i.e., classifying all images locally versus over the cloud under energy and communication constraints., Comment: This conference paper will appear in the proceedings of ICCAD 2018

Supplemental Material for Bivalirudin versus heparin in patients undergoing percutaneous peripheral interventions: A systematic review and meta-analysis by Wilman Olmedo, Pedro A Villablanca, Cristina Sanina, Jonathan Walker, Michael Weinreich, Jeannine Brevik, Ricardo Avendano, Claudio A Bravo, Jorge Romero, Harish Ramakrishna, Anvar Babaev, Michael Attubato, DF Hernandez-Suarez, P Cox-Alomar, Robert Pyo, Prakash Krishnan and Jose Wiley in Vascular

Occupancy estimation is very useful for a wide range of smart building applications including energy efficiency, safety, and security. In this demonstration, we present a novel solution called FORK, which uses a Kinect depth sensor for estimating occupancy in real-time. Unlike other camera-based solutions, FORK is much less privacy invasive (even if the sensor is compromised) and it does not require a powerful machine like a Microsoft XBOX or an Intel® Core™ i7 processor to process the depth data. Our system performs the entire depth data processing on a cheaper and lower-power ARM processor, in real-time. In order to do that, FORK uses a novel lightweight human model by leveraging anthropometric properties of human bodies for detecting individuals. We will show how FORK detects, tracks, and counts occupants accurately in real-time.

The use of smart-meters is proliferating, they are now being deployed without asking the obvious question: Do we really need each of them? Beyond the cost of smart-meters, there are overheads related to installation, wiring, etc. To formally tackle this question, we first define the notion of observability that one or more pieces of information (including that from smart-meters) enable. This notion allows us to compare two different deployments of sensors with respect to their information content and their usefulness. We then examine some commonly available information from which one can infer power consumption of devices in a given space. We show how we have applied this approach to systematically decide the optimal number and location of smart-meters to ensure observability of consumption by different parts of a building.

As buildings get smarter they need to be aware of their spaces and occupants to improve prediction and management of energy consumption and environment customization based on user preference. User identification is crucial to this. However, accuracy of identification, intrusiveness and cost are important factors that one considers before installing such a system. Accounting for these factors, we built a Smart-Door that incorporates fusion of not-so-smart sensors, soft information available and learning algorithms to build an economical and accurate user identification system that requires no user intervention to monitor the occupant count and identities in a shared office space that can be scaled up to a building. It provides real-time occupancy status for the area and it can also learn to identify new users. In addition to energy management, such a user identification system has significant applications including evacuation procedures and localizing malfunctioning appliances.

This study evaluated the in vitro and in vivo antidiarrheal activity, oral acute toxicological profile, and developed a chemical fingerprint of Berberis aristata Linn. (Berberidaceae).The ethanol (by maceration) and aqueous (by Soxhlet) extracts of Berberis aristata bark were used for the study. The study involved the antimicrobial (minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) by micro dilution) and antidiarrheal (castor oil induced diarrhea, charcoal motility) tests. The active principle, berberine was characterized by different spectroscopic and chromatographic techniques.The MIC and MBC (of both extracts) against all strains of Shigella were recorded between 125 to 500 µg/mL and 300 to 600 µg/mL, respectively. The MIC and MBC values of berberine are almost comparable to standard ciprofloxacin. UV and IR spectroscopy along with HPTLC and HPLC studies showed presence of berberine in the extracts. The onset of castor oil induced diarrhea was delayed and number of diarrheal episodes was reduced by both the extracts in a dose dependent manner. Similarly, the length of intestine traveled by the feed was also significantly reduced in the charcoal motility test by both the extracts. LD(50) of5000 mg/kg body weight was observed for both extracts in the acute oral toxicity studies with Swiss albino mice.The results validate in vivo and in vitro antidiarrheal activity of Berberis aristata extracts and provide its chemical fingerprint.