Fault Injection with Multiple Fault Patterns for Experimental Evaluation of Demand-Controlled Ventilation and Heating Systems

• Published in: Sensors (Basel, Switzerland) Vol. 22; no. 21; p. 8180
• Main Authors: Kiamanesh, Bahareh, Behravan, Ali, Obermaisser, Roman
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 25.10.2022; MDPI
• Subjects: Air quality; Airports; Analysis; Building management systems; Buildings; Climate; Energy consumption; Energy efficiency; Environmental conditions; fault injection framework; fault model; fault occurrence probability; Heating; HVAC; HVAC equipment; HVAC system; Hypotheses; multiple faults; scenario generation; Sensors; Simulation; Software; Ventilation; United States
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s22218180

Heating, ventilation, and air-conditioning (HVAC) systems are large-scale distributed systems that can be subject to multiple faults affecting the electronics, sensors, and actuators, potentially causing high energy consumption, occupant discomfort, degraded indoor air quality and risk to critical infrastructure. Fault injection (FI) is an effective experimental method for the validation and dependability evaluation of such HVAC systems. Today’s FI frameworks for HVAC systems are still based on a single fault hypothesis and do not provide insights into dependability in the case of multiple faults. Therefore, this paper presents modeling patterns of numerous faults in HVAC systems based on data from field failure rates and maintenance records. The extended FI framework supports the injection of multiple faults with exact control of the timing, locality, and values in fault-injection vectors. A multi-dimensional fault model is defined, including the probability of the occurrence of different sensor and actuator faults. Comprehensive experimental results provide insights into the system’s behavior for concrete example scenarios using patterns of multiple faults. The experimental results serve as a quantitative evaluation of key performance indicators (KPI) such as energy efficiency, air quality, and thermal comfort. For example, combining a CO2 sensor fault with a heater actuator fault increased energy consumption by more than 70%.