Data Preparation Process for the Buildings Performance Database

Walter, Travis ; Dunn, Laurel ; Mercado, Andrea ; Brown, Richard E. ; Mathew, Paul

关键词: Data processing;    data preparation;    energy information;    database performance;    commercial buildings;    residential buildings;    building benchmarking;   

DOI  :  10.2172/1163654
RP-ID  :  LBNL-6724E
PID  :  OSTI ID: 1163654

学科分类：能源（综合）

美国|英语

来源: SciTech Connect

【 摘 要 】

The Buildings Performance Database (BPD) includes empirically measured data from a variety of data sources with varying degrees of data quality and data availability. The purpose of the data preparation process is to maintain data quality within the database and to ensure that all database entries have sufficient data for meaningful analysis and for the database API. Data preparation is a systematic process of mapping data into the Building Energy Data Exchange Specification (BEDES), cleansing data using a set of criteria and rules of thumb, and deriving values such as energy totals and dominant asset types. The data preparation process takes the most amount of effort and time therefore most of the cleansing process has been automated. The process also needs to adapt as more data is contributed to the BPD and as building technologies over time. The data preparation process is an essential step between data contributed by providers and data published to the public in the BPD.