FBMTP: An Automated Fault and Behavioral Anomaly Detection and Isolation Tool for PLC-Controlled Manufacturing Systems

Abstract

The fault and behavioral anomaly detection and isolation (FBADI) in programmable logic controller (PLC) controlled systems has been under an active study for several decades. In this paper, we present a tool, called the fault and behavior monitoring tool for PLC (FBMTP) that can solve the FBADI problem in PLC-controlled manufacturing systems effectively. FBMTP first creates a nominal deterministic finite-state automaton-based model of the PLC control process, and then utilizes that model to detect and isolate the faults and the behavioral anomalies. The key idea is to check whether the observed behavior is consistent with the modeled behavior or not. The control signal data stored in the PLC memory is used for this whole procedure. With growing size of the manufacturing systems, it becomes difficult to obtain all the control signal data accurately because of the limited data acquisition time in each PLC scan cycle. The proposed control process modeling and the FBADI methodology of FBMTP can cope with such data inaccuracy problems associated with large manufacturing systems efficiently. Our experiments show that FBMTP provides highly accurate FBADI results for both small and large manufacturing systems.