Data Integrity Checks for Building Automation and Control Systems


  • Markus Gwerder Smart infrastructure | Siemens Switzerland Ltd. | Switzerland
  • Reto Marek Center for integrated building technology | Lucerne school of engineering and architecture | Switzerland
  • Andreas Melillo Institute of mechanical engineering and energy technology | Lucerne school of engineering and architecture | Switzerland
  • Maria Husmann Smart infrastructure | Siemens Switzerland Ltd. | Switzerland



Building automation, analytics, data integrity, data plausibility, semantic modeling


Data from building automation and control systems are becoming more and more important, since they are used in a growing number of (novel and established) analytics applications such as fault detection & diagnostics (FDD), smart maintenance and optimization. However, the quality of such data is often poor due to erroneous installation, commissioning, data recording or meta-information. In addition, building automation engineering and service departments usually focus on implementing and maintaining basic control functionality – data acquisition, tagging quality, and analytics do often not take priority. Due to these data quality issues, a first important step in any data analytics operation is to ensure data integrity. One main goal of data integrity checks is to increase data reliability. The paper presents such checks for building automation applications, in particular three different types of plausibility checks for time series data: single signal tests, similarity tests, and reaction tests. Examples using data recorded from real building automation project are presented for each of the three check types, demonstrating the usefulness of these checks. Data integrity checks are set up and configured using the available metadata which – in our case – comes in the form of semantic models that are automatically generated from building automation engineering data. Many data integrity checks have been identified that are potentially of great benefit in practice – both as a stand-alone application or as first part in a data analytics process. The major prerequisite for successful data integrity checking is that the checks can be set up with minimal effort and executed periodically. To achieve a high degree of automation, semantic data is of great importance, because it is through them that the recorded time series are provided with context and meaning. The automatically generated semantic models from building automation engineering proved to be already rich in automation information and are sufficient for many of the checks investigated.




How to Cite

Gwerder, M., Marek, R., Melillo, A., & Husmann, M. (2022). Data Integrity Checks for Building Automation and Control Systems. CLIMA 2022 Conference.

Conference Proceedings Volume