A Novel Simulation-Based Framework for Sensor Error Impact Analysis in Smart Building Systems: A Case Study for a Demand-Controlled Ventilation System: Article No. 114638

Yanfei Li, Xing Lu, Zheng O'Neill, Fuxin Nu

Research output: Contribution to journalArticlepeer-review

35 Scopus Citations

Abstract

Sensors are one of the fundamental components for sensor-rich controls in buildings but are prone to different errors. Existing studies show that sensor errors hold a place among top-priority faults in building systems. Before we take countermeasures to mitigate the sensor errors, it is vital to prioritize key sensors and quantify the collective impacts of concurrent sensor errors. In response to this, a simulation-based methodology is introduced to conduct a comprehensive sensor error impact analysis in building systems, which adds a stochastic sensor prioritization through a sensitivity analysis on top of a commonly used deterministic sensor error quantification. The synergies of these two parts help better interpret the sensor error impacts on building energy consumption, ventilation performance, thermal comfort, etc. A sensor-rich CO2-based Demand-Controlled Ventilation system is used as a case study to demonstrate the viability of the methodology as a proof-of-the-concept. The results show that the energy savings potential and ventilation performance are mostly influenced by the accuracy of the AHU outdoor airflow sensors. The accuracy of zone level airflow sensors has a negligible impact on both energy savings and ventilation performance. The accuracy of zone CO2 sensors has more influence on the ventilation performance compared with the accuracy of zone airflow sensors. Compared with the baseline case with zero errors, the largest deviation percentages could reach 16.90% and 94.32%, respectively, in terms of the Heating, Ventilation, and Air-Conditioning (HVAC) annual energy consumption and the Outdoor Air Ratio (OAR) when multiple key sensors suffer from normal error intensities simultaneously.
Original languageAmerican English
Number of pages17
JournalApplied Energy
Volume263
DOIs
StatePublished - 2020

NREL Publication Number

  • NREL/JA-5500-76212

Keywords

  • DCV
  • demand-controlled ventilation
  • error impact analysis
  • sensitivity analysis
  • sensors
  • simulation

Fingerprint

Dive into the research topics of 'A Novel Simulation-Based Framework for Sensor Error Impact Analysis in Smart Building Systems: A Case Study for a Demand-Controlled Ventilation System: Article No. 114638'. Together they form a unique fingerprint.

Cite this