Search

Commercial buildings during operation are dynamic environments where changes to control strategies and space usage regularly occur. As a result of these and other issues, a performance gap between design intent and actual building performance emerges. This paper seeks to address the operational performance gap and enhance operational building performance through statistically-based fault detection. Additionally, this paper seeks to remedy the knowledge gap building managers face in the identification of key building faults based on minimal quantities and streams of time-series building data. A new methodology is presented that incorporates simulation and breakout detection to address these issues. Residual based exponentially weighted moving average (EWMA) charts and Shewhart charts are compared against a breakout detection algorithm to identify shifts or faults in building performance data. Artificial faults are introduced into the measured time-series data to test the validity of the chosen statistical techniques. Statistical metric sensitivity and precision are calculated to quantify the performance of the new methodology. A summary of results demonstrate that the breakout detection algorithm was the most effective method in detecting meaningful faults in building performance data, followed by residual based EWMA and Shewhart models. [ABSTRACT FROM AUTHOR]

A pronounced gap often exists between expected and actual building performance. The multi-faceted and cross lifecycle causes of this performance gap are found in design assumptions, construction issues and commissioning and operational compromises. Some important factors are firmly rooted in the lack of interoperability around building information. New solutions to the interoperability challenge offer the potential to leverage and reuse available heterogeneous data in a manner that can significantly assist building performance assessment. Linked data provides an open, modular and extensible solution for the challenge. However, in the buildings domain, the integration of rule-based performance metrics and contextual information has yet to be formally established. This paper describes an approach to the provision of in-depth building performance assessment through the integration of OpenMath and linked data. An ontology describing performance metrics in RDF is presented, together with an automated metric evaluation solution using multi-silo queries and computer algebra systems, providing a flexible, automated and extensible mechanism for the assessment of building performance. Building managers and engineers can simultaneously analyse time-series building performance at a range of levels, without burdensome manual intervention such as is the case with traditional solutions. A test implementation on a large university building highlights the potential of this solution. [ABSTRACT FROM AUTHOR]

• Evaluation of existing approaches or projects of building archetype. • Developing a generalized multi-scale building archetype methodology. • Implementing the data-driven approaches for the characterization of building archetypes. • Comparison of the modeling results of multi-scale building archetype at BEM and UBEM levels. Globally the building sector accounts for a significant portion of the overall energy demand and greenhouse gas emissions of any country. The most common approach for the collection of modeling and benchmarking data that can be used for predictions of energy performance at a national or urban scale is through classification of the building stock into representative archetypes. Developing such building archetypes is a complex task due to the difficulties associated with gathering detailed geometric and non-geometric data at an urban scale. Although existing databases and projects provide a valuable overview of a building stock, the information about buildings' physical descriptions are not regularly updated. Moreover, these databases cover only the national top-level archetypes and lack crucial information related to city or district scale building stocks. The use of national scale archetypes requires many assumptions that may not hold true for energy modeling at urban or district scale. This paper proposes a multi-scale (national, city, county and district) archetype development methodology using different data-driven approaches. The methodology consists of following five steps: 1) data collection, 2) segmentation, 3) characterization, 4) quantification, and 5) modeling results. We developed a test case based on the available building stock data of Ireland. The test case used previously developed archetype geometries coupled with the parameters determined by the characterization process to calculate annual energy use (kWh) of buildings at a multiple-scales. The resulting archetypes at national, city, county and district scale are analyzed and compared against one another. The results indicate that significant differences occur in terms of energy modeling results when national scale archetypes are used to simulate the energy performance of buildings at the local scale. These multi-scale building archetypes will aid local authorities and city planners when analyzing energy efficiency and consequently, help to improve sustainable energy policy decisions. [ABSTRACT FROM AUTHOR]