A flexible alarm prediction system for smart manufacturing scenarios following a forecaster–analyzer approach

Abstract

The user has requested enhancement of the downloaded file.

Predictive Maintenance: A Novel Framework for a Data-Driven, Semi-Supervised, and Partially Online Prognostic Health Management Application in Industries

“…Finally, novelty detection and anomaly detection are not sensitive to point anomalies. An important aspect in PHM industrial applications is represented by false alarms [85]. The integration of anomaly detection and clustering makes it possible to generate an alarm only when the current observation is assigned to a distinct cluster of the previous point instead of when an anomaly is detected.…”

Predictive Maintenance and Intelligent Sensors in Smart Factory: Review

“…Villalobos et al [118] Alarms can allow the operators in the plant to conduct proactive management of the different controls in the machine for predictive maintenance of the equipment. CbM…”

“…Method Data Source Al-Jlibawi et al [107] simulation DCS (distributed control systems), PLC, or SCADA in refinery Barbieri et al [99] case study Alternating current (AC) motor (machinery), Pronistia dataset Bekar et al [108] case study Machine motor Farooq et al [38] case study SCADA in spinning factory, spinning frame JWF1562 Goodall et al [92] simulation RFID in remanufacturing facility Chien and Chen [109] case study health status of plasma enhanced chemical vapor deposition (PECVD) chamber tool in TFT(thin film transistor) and LCD (liquid crystal display) company Kiangala and Wang [94] experiment SCADA, conveyor motors Kozlowski et al [110] case study CNC cutter machine sensors for milling of thin-walled aircraft engine components Kumar et al [102] case study CNC machine sensors Lao et al [111] simulation chemical product concentration and temperature profiles Li et al [96] experiment test data from IoT devices and detectors Lin et al [103] experiment test data from IoT in smart factory Musselman and Djurdjanovic [104] experiment automated storage/retrieval systems (belt-driven material handling device) in semiconductor industry Park et al [112] experiment servo motor testing data in smart factory Peng et al [113] experiment NI-PXI (PCI extensions for instrumentation) and NI-Compact data acquisition from production lines in China Steel Corporation Peng and Tsan [98] experiment production line machines Sadiki et al [105] case study industrial machine behaviour Shan et al [114] simulation welding robot in automotive production line Tarashioon et al [115] experiment LED (light-emitting diode) lighting system technologies Tsao et al [116] simulation production system and production lines Uhlmann et al [117] experiment ball and screw monitoring of machine tools Villalobos et al [118] case study melting and extruder machines in plastic bottles production plant (Capital Equipment Manufacturer) Vlasov et al [119] case study the supporting bearing of electric machines (AC motors)…”

“…Sensor Type Sensor Description Al-Jlibawi et al [107] virtual software sensor Barbieri et al [99] vibration B&R X20CM4800X, Beckhoff EL3632 Bekar et al [108] temperature, vibration not available Farooq et al [38] speed, vibration not available Goodall et al [92] position tracking RFID for traceability [114] movement not available Tarashioon et al [115] light, temperature LED sensors architecture Tsao et al [116] not available not available Uhlmann et al [117] vibration, temperature Micro-Electro-Mechanical System (MEMS) of vibration sensors (LIS3DH) Villalobos et al [118] temperature, pressure, speed not available Vlasov et al [119] vibration vibration sensors network Yan et al [120] multiple (vibration, optical, acoustical, power)…”

“…Tmote sky sensors network and Z1 mote operate through the edge router and enhance the maintenance simulation's purpose. Villalobos et al [118] introduced a flexible forecaster analyser system for anticipation of alarms' activation based on the temperature sensor data. The deep learning techniques based on the short-term memory recurrent neural network contribute to the predictive maintenance approach.…”