Sequential state logic for pneumatic valve monitoring using piezo film sensors

Intelligent Approaches for Anomaly Detection in Compressed Air Systems: A Systematic Review

“…DAMADICS is a benchmark developed by multiple universities for evaluation, contrast, and comparison regarding standard performance criteria based on industry requirements for an industrial pneumatic actuator within a Polish manufacturing site producing sugar [40]. Anomaly detection techniques based on the behaviour of a pneumatic control valve were investigated by Venkata et al [41], Yerra et al [42], and Subbaraj and Kannapiran [43], where although these research teams chose the DAMADICS component, they implemented a different anomaly detection approach. Additionally, Kovács et al [44,45] implemented a monitoring system for a pneumatic actuator and subsequently applied an ML technique.…”

“…Conversely, the studies performed by Venkata et al [41], Yerra et al [42], and Subbaraj and Kannapiran [43] focused on one pneumatic component, rather than the compressor system. Apart from this, all three publications featured a different approach on which data were collected.…”

“…Data collection was carried out directly from the real-time system interfaced with a personal computer and a c-DAQ card [43]. An interesting approach was used by Yerra et al [42], as the dataset comprised measurements from piezofilm sensors for the acquisition of valve acoustic signals through an unintrusive method. This differs from any other included publication.…”

“…Additionally, the three pneumatic valves within the experimental setup were controlled by a PLC program. The investigation was based on one of these valves only, which was a pneumatically operated Burkert 2/2-way valve (Type 2100) of angle-seat type [42].…”

“…A real-world dataset was used in [38], which was not generated by the research team, but obtained from the ECP braking system of a DK-2 locomotive from Shouhuang Railway. Within the manufacturing research field, Rajashekarappa et al [52] presented a real-world implementation, as opposed to the investigations carried out by [41][42][43]51,65]. Rajashekarappa et al [52] generated their dataset comprising airflow, temperature, and pressure sensors from a real-world pneumatic production system.…”