Fault detection and compensation for linear systems over networks with random delays and clock asynchronism

This paper proposes a fault detection and compensation scheme based on likelihood ratios (LRs) for networked predictive control systems with random network-induced time delays and clock asynchronism. The compensator is applied to compensate for the network-induced time delays. The measured outputs are sent back to the local node with random delays and the observer updates based on the time schedule of a remote node clock to avoid the deficiency of asynchronism. Two schemes are proposed in this paper to update the LRs of fault. One of them is to set up a buffer in the local node to save the measured outputs out of sequence, and the observer processes the measured outputs one by one in their original sequence. The other scheme is to discard the measured outputs that are out of sequence; thus, the observer has to estimate the state and update the LRs with intermittent observations. The convergence analysis of a generalized LR test with intermittent observations is proposed as well. A numerical simulation is also given to validate the proposed method.