Deep-learning-based design of active fault-tolerant control for automated manufacturing systems subjected to faulty sensors

Abstract

This research paper proposes a new implementation of a long short-term memory network for active fault-tolerant control subjected to single and multiple undiagnosable sensor faults. There are two networks used: the first performs as a diagnosis for automated manufacturing systems and can identify faulty sensors, while the second acts as an inverse model of these systems and is used to determine the reconfigured control action to take when sensors are not functioning as expected. The Factory I/O simulator is interfaced with MATLAB to simulate and verify the proposed approach for automated material handling case study with faultless and multiple fault sensors. Four sensors of the case study out of six are tolerated.