Search

Electricity load forecasting for buildings and campuses is becoming increasingly important as the penetration of distributed energy resources (DERs) grows. Efficient operation and dispatch of DERs require reasonably accurate predictions of future energy consumption in order to conduct near-real-time optimized dispatch of on-site generation and storage assets. Electric utilities have traditionally performed load forecasting for load pockets spanning large geographic areas, and therefore forecasting has not been a common practice by buildings and campus operators. Given the growing trends of research and prototyping in the grid-interactive efficient buildings domain, characteristics beyond simple algorithm forecast accuracy are important in determining true utility of the algorithm for smart buildings. Other characteristics include the overall design of the deployed architecture and the operational efficiency of the forecasting system. In this work, we present a deep-learning-based load forecasting system that predicts the building load at 1-hour intervals for 18 hours in the future. We also discuss challenges associated with the real-time deployment of such systems as well as the research opportunities presented by a fully functional forecasting system that has been developed within the National Renewable Energy Laboratory Intelligent Campus program., Comment: 13 pages, 4 figures

This article presents a Unified Architecture for automated point tagging of Building Automation System data, based on a combination of data-driven approaches. Advanced energy analytics applications-including fault detection and diagnostics and supervisory control-have emerged as a significant opportunity for improving the performance of our built environment. Effective application of these analytics depends on harnessing structured data from the various building control and monitoring systems, but typical Building Automation System implementations do not employ any standardized metadata schema. While standards such as Project Haystack and Brick Schema have been developed to address this issue, the process of structuring the data, i.e., tagging the points to apply a standard metadata schema, has, to date, been a manual process. This process is typically costly, labor-intensive, and error-prone. In this work we address this gap by proposing a UA that automates the process of point tagging by leveraging the data accessible through connection to the BAS, including time series data and the raw point names. The UA intertwines supervised classification and unsupervised clustering techniques from machine learning and leverages both their deterministic and probabilistic outputs to inform the point tagging process. Furthermore, we extend the UA to embed additional input and output data-processing modules that are designed to address the challenges associated with the real-time deployment of this automation solution. We test the UA on two datasets for real-life buildings: 1. commercial retail buildings and 2. office buildings from the National Renewable Energy Laboratory campus. The proposed methodology correctly applied 85-90 percent and 70-75 percent of the tags in each of these test scenarios, respectively., Comment: 19 pages, 9 figures, accepted for publication in Automation in Construction

The All-Sky Automated Survey for Supernovae (ASAS-SN) is the only project in existence to scan the entire sky in optical light every $\sim$day, reaching a depth of $g\sim18$ mag. Over the course of its first four years of transient alerts (2013-2016), ASAS-SN observed 53 events classified as likely M dwarf flares. We present follow-up photometry and spectroscopy of all 53 candidates, confirming flare events on 47 M dwarfs, one K dwarf, and one L dwarf. The remaining four objects include a previously identified TT Tauri star, a young star with outbursts, and two objects too faint to confirm. A detailed examination of the 49 flare star light curves revealed an additional six flares on five stars, resulting in a total of 55 flares on 49 objects ranging in $V$-band contrast from $\Delta V = -1$ to $-10.2$ mags. Using an empirical flare model to estimate the unobserved portions of the flare light curve, we obtain lower limits on the $V$-band energy emitted during each flare, spanning $\log(E_V/{\rm ergs})=32$ to $35$, which are among the most energetic flares detected on M dwarfs. The ASAS-SN M-dwarf flare stars show a higher fraction of H$\alpha$ emission as well as stronger H$\alpha$ emission compared to M dwarfs selected without reference to activity, consistent with belonging to a population of more magnetically active stars. We also examined the distribution of tangential velocities, finding that the ASAS-SN flaring M dwarfs are likely to be members of the thin disk and are neither particularly young nor old., Comment: 29 pages, 10 figures, 6 tables, submitted to AAS journals (ApJ)

K-12 education resources are often allocated non-strategically, with schools spending time and money on activities that have little relationship to student outcomes. Most of these decisions take place within districts, rooted in the processes of setting schedules, staffing levels, and assignments, and creating final budgets. Local Education Agencies (LEAs) could dramatically improve their resource allocation by better analyzing data on how people, time, and money are used through the district, and looking for ways to better match student needs. To do that, districts have to link specific student needs to specific school interventions, and that requires merging and reconciling a range of data sets. Most LEAs do not make this investment, but State Education Agencies (SEAs) can and do. States already use this kind of data to create public accountability reports or inform research or policy, but not typically to support local decisions. As such, they are sitting on a veritable gold mine of untapped material that could be used to support LEA resource decisions. By aiding local decision-makers to be more strategic, SEAs could help reallocate resources across the state without introducing a single mandate. In this paper, and an accompanying detailed appendix, we look at key decision-making processes that largely determine how people, time and money are used in LEAs. And for each process, we identify questions and metrics where the strategic use of state data would help school and district leaders make better resource decisions. The following is appended: Sample Questions and Decision-Support Metrics.

Recently, there has been growing interest in adapting Return-on-Investment thinking to education--sometimes called educational productivity, or academic-ROI. Education leaders do not seek a monetary return on their spending, rather greater student learning, or other outcomes like student citizenship, higher graduation rates, or increased lifetime earnings and career options. Most ROI analysis misses a big opportunity; it is often used to evaluate programs and initiatives, such as comparing two professional development programs. This process leaves out some of the biggest cost drivers in the district's budget and never addresses some of the fundamental resource issues that may actually drive student success. Education leaders need a new approach--a System-Strategy ROI. This approach starts with a fundamental student need and asks not "Which program is better?" but "What resources will meet this need?" School system leaders can help their teams take a System-Strategy ROI approach by structuring the planning conversation around five key steps: (1) Identify the core need: What fundamental student performance need are we focusing on, and what's our theory of change for addressing it?; (2) Consider a broad range of investment options: What are the investments we currently make to address this need, and what else could we do?; (3) Define ROI metrics and gather data: What are the relative returns (costs weighed against benefits) to the set of current/potential options?; (4) Weigh investment options: What other factors do we need to consider, in order to select from among the options?; and (5) Make investment decisions: How can we free resources to do what we want to do? It is important to note that ROI--in any form--is not a magic formula, but the System-Strategy approach is a powerful tool for adding structure, rigor, and data-backed evidence to the difficult decisions a school system must make on behalf of its students. Ultimately, the fundamental question we must ask is, "How can we use all our limited resources strategically to improve student achievement and meet our goals?" This document guides districts toward the goal of aligning school system resources to strategic priorities so that every school succeeds for every student.

Across the country, school districts are struggling to improve student performance on flat or declining budgets. Many districts are understandably cautious about implementing large changes, such as redesigning the step-and-lane system that has existed for decades. New evaluation systems must be implemented and vetted before they are linked to compensation, and it is challenging to find common ground among administration, teachers, and unions on the best approach.This paper outlines a series of actions that districts can take to start moving toward a future vision of the teaching job. These options can be implemented one at a time, or can be combined to provide more learning impact and/or so that the net result is budget neutral.

Superintendent Andres A. Alonso arrived in Baltimore in 2007 with a vision for improving the city's struggling schools. His vision included empowering school leaders and creating accountability for student learning through a series of reforms that center around a new system for giving resources to schools, called Fair Student Funding (FSF). FSF attempts to create equity across schools and students by targeting resources according to student needs--academic and social-emotional--and build a system of school-based autonomy and flexibility focused on fulfilling those needs. Instead of awarding employees to schools based on staffing ratios, under the leadership of Alonso in 2008-2009, Baltimore City Public Schools (City Schools) adopted a per pupil funding formula. This gave dollars to schools with the amount weighted to vary based on estimates of student learning need and the schools' respective needs. Since the rapid implementation of FSF and related reforms, City Schools has moved off the federal "needs improvement" list. The district has celebrated a 50% decline in the drop-out rate (from 9% to 4%), a notable improvement in student performance, and the reversal of a decades-long enrollment decline. With the rapid implementation and encouraging results, FSF and related reforms continue to be a work-in-process, as Baltimore City Public Schools adjusts student weights, renegotiates contracts and builds capacity and support at the school level. At the request of Carnegie Corporation of New York and with the cooperation of City Schools, Education Resource Strategies (ERS) set out to examine Baltimore's implementation of FSF and the reforms that accompanied it. In this summary, ERS shares the extent to which City Schools has met the objectives of FSF and related reforms and summarizes key lessons learned. (Contains 1 footnote.)

ERS final report summarizing research on the Syracuse City School District (SCSD) and the participative community "work-out" process that was held in response to research results. The report shows how SCSD is in a strong position to improve and deliver on its promise for all children, but that to succeed, dramatic action is still necessary. There are six recommendations to help the district ensure highly effective teaching, more targeted individual attention for students, and strategic use of in-school time: (1) Restructure the teacher job to attract and retain the highest contributors and support effective teams; (2) Focus more on the quality of instruction than on the number of instructors; (3) Move away from one-size-fits-all class sizes to target individual attention based on student and subject needs; (4) Extend and organize student time strategically; (5) Redesign Special Education to shift resources to early and ongoing intervention in general education settings; and (6) Unbundle the provision of instructional and other services to take advantage of high-quality, lower-cost options. ERS Methodology Overview is appended. (Contains 8 footnotes.) [Funding for this paper was provided by Say Yes To Education.]

There is no disputing that for high school students to be college and career ready, they need adequate time to learn, and individualized attention to meet each student's academic goals, learning styles, and social needs. Over the past decade, Education Resource Strategies (ERS) has partnered with urban schools and districts to improve their use of people, time, and money in addressing students' most pressing needs. Through this work, the author and his colleagues have studied many high-performing schools that are using every staff member, dollar, and minute in a way that is improving student learning. Their study of high schools in six urban school systems yielded the following findings regarding time and attention: (1) Student time in school varies by up to 30 percent across districts; (2) Time allocation is strikingly similar across districts because it is driven by traditional graduation requirements and rigid structure of school schedules; (3) General Education class sizes and teacher loads are not lower for core academic subjects, high needs students, or foundation grade levels; (4) Schools aren't using data to adjust time and attention; and (5) Struggling students get extra attention primarily through special education placement, which can drain resources from general education and instructional support. Appended are: (1) District Summary Characteristics; (2) Average Class Size, by Subject and Grade; and (3) ERS Analytic Methods. A bibliography is included. (Contains 14 figures and 4 footnotes.)

In this paper, a companion piece to "Strategic Designs: Lessons from Leading Edge Small Urban High Schools" (ED544382), the authors analyze small high school spending in three urban districts--Baltimore City, Boston, and Chicago--to understand whether it is higher than in larger schools; if it is, why it is different; and the policy considerations connected to small school spending. Specifically, they examined two key questions: (1) How much do districts spend to operate small high schools? and (2) How do patterns of resource use in small high schools differ from patterns of resource use in larger high schools? They detail the methodology they used to address these two questions and present a discussion of findings, policy considerations, and recommendations for how school and district leaders can approach small school resource use and funding. [For the full report, see ED544392.]

In this paper, a companion piece to "Strategic Designs: Lessons from Leading Edge Small Urban High Schools" (ED544382), the authors analyze small high school spending in three urban districts--Baltimore City, Boston, and Chicago--to understand whether it is higher than in larger schools; if it is, why it is different; and the policy considerations connected to small school spending. Specifically, they examined two key questions: (1) How much do districts spend to operate small high schools? and (2) How do patterns of resource use in small high schools differ from patterns of resource use in larger high schools? They detail the methodology they used to address these two questions and present a discussion of findings, policy considerations, and recommendations for how school and district leaders can approach small school resource use and funding. Two appendices present: (1) Data Request; and (2) Coding Scheme for Resource Use. (Contains 25 endnotes and 11 figures.) [For the executive summary, see ED544391.]

In the fall of 2006, Los Angeles Unified School District (LAUSD) and the Unified Teachers of Los Angeles (UTLA) signed a memorandum of understanding (MOU) that called for a third-party research firm to analyze resource use across the Los Angeles school system. Education Resource Strategies (ERS) was chosen to conduct this analysis. The study was organized into two phases. Phase I of the study called for a six-year historical review of spending in the district to examine how spending patterns had changed and specifically whether or not central expenses had grown while spending on teachers had declined. The findings of phase I were released publicly in November of 2006 and presented to the Board of Education of LAUSD. Phase II of the study explores spending and use of resources more deeply to identify opportunities for more strategic use based on research and comparison to other urban school districts. This management brief series presents Phase II key findings and recommendations. This document includes an executive summary that provides a concise summary of Phase II findings and describes the top 5 actions that flow from ERS's detailed recommendations. This summary is followed by three management briefs that supply detailed answers to key questions as well as additional recommendations and questions for further leadership discussion: (1) Brief #1--Distribution and Use of System Resources: Analyzes the overall funding level in LAUSD as well as the way time, people, and money flow to schools and students, focusing on questions of equity, strategy, flexibility, transparency, and efficiency, with comparisons to other urban districts; (2) Brief #2--Time, People and Money in Schools: Describes how LAUSD schools currently allocate their time, people, and money and compares findings with practices often found in high-performing urban schools; and (3) Brief # 3--Professional Development: Analyzes LAUSD's overall investment in professional development and compares it to patterns and practices found elsewhere, focusing on whether the overall mix of initiatives aligns with district needs. (Contains 7 footnotes.)

In the Fall of 2006, Education Resource Strategies (ERS), a nation-wide leader in helping school systems make the most of their resources, was invited by Saint Paul Public Schools (SPPS) to study the use of resources in the district and create priority recommendations to inform the district's strategic plan. Saint Paul is in a situation where overall enrollment is declining while key subgroups (students in poverty, English language learners, students with disabilities, and students of color) are increasing as a percentage of district enrollment. In this context, the analysis focused on three key questions: (1) How much money does SPPS spend and how are the resources distributed? (2) Does SPPS empower strategic use of school-level resources to improve student performance? and (3) Does SPPS use district resources in a way that supports schools in promoting student success? This report summarizes the key findings and priority recommendations regarding strategic resource use in SPPS. (Contains 13 footnotes.)