Data-driven intermittent connection fault diagnosis for complex topology DeviceNet based on Bayesian inference

Abstract

As the topology of DeviceNet in industrial automation systems grows more complex and the reliability requirement for industrial equipment and processes becomes more stringent, the importance of network troubleshooting is increasingly evident. Intermittent connection (IC) faults frequently occur in DeviceNet systems, impairing production performance and even operational safety. However, existing IC troubleshooting methods for DeviceNet, especially those with complex topologies, cannot directly handle multi-fault scenarios, which require human intervention for a full diagnosis. In this paper, a novel data-driven IC fault diagnosis method based on Bayesian inference is proposed for DeviceNet with complex topologies, which can accurately and efficiently localize all IC faults in the network without interrupting the normal system operation. First, the observation symptoms are defined by analyzing the data frames interrupted by IC faults, and the suspected IC faults are derived by integrating the observation symptoms and the network topology information. Second, a Bayesian inference-based estimation approach for the posterior probability of each suspected fault occurring in the network is proposed using the quantity of observation symptoms and their causal relationships regarding the suspected faults. Finally, a maximum likelihood-based fast diagnosis algorithm is developed to rapidly identify the IC fault locations in various complex scenarios. A laboratory testbed is constructed and case studies are conducted under various topologies and fault scenarios to demonstrate the effectiveness and advantages of the proposed method. Experimental results show that the IC fault locations diagnosed by the proposed method agree well with the experimental setup.

1. Introduction

DeviceNet, a fieldbus network using the controller area network (CAN) as the physical and data link layer protocol (Open DeviceNet Vendor Association, 2021), is widely used for transmitting important data in industrial automation systems, such as automotive manufacturing systems and distributed control systems (Gessner et al., 2014; Cheng et al., 2024a, 2024b), which require excellent real-time performance and network reliability. However, in real-world environments, unavoidable factors such as ambient interference, vibration, and loose connectors can lead to frequent intermittent connection (IC) faults in the network. Manifested as intermittent and random disconnections of network cables in short time intervals, IC fault is a common but hard-to-diagnose problem.

Such faults can cause transmission delays or even loss of critical messages due to random interruptions in network communications, which may cause system shutdowns and security-related problems. It is tremendously challenging to diagnose IC faults in complex topologies, that is, the hierarchical structures that evolve from bus structures by combining multi-port connectors, because the limited available open-style ports of DeviceNet in practice hinder the diagnosability of the subnets inside the complex topology network (Wang LK et al., 2023). Therefore, timely detection and complete diagnosis of IC faults before they cause system-level failures are of great importance for the reliability of complex topological systems. In the literature, some preliminary studies have been conducted on IC fault diagnosis methods for CAN networks. Zhao and Lei (2012) and Lei et al. (2014b) applied generalized zero inflated Poisson models to describe the IC-induced errors, and used a ranked probability control chart to monitor the error anomalies; however, the localization of IC faults was not addressed. Lei et al. (2014a, 2015) defined two error events corresponding to different IC fault scenarios in CAN networks and estimated the confidence intervals of the parameters for the error events to diagnose IC faults; however, their approach depends on analog signal analysis at the physical layer, which is not robust. Based on the data link layer information for the IC-induced errors, Zhang et al. (2017) proposed a context-free-grammar-based IC fault localization method, and Zhang et al. (2019) also proposed a tree-based IC fault diagnosis method. However, the diagnostic accuracy of these qualitative diagnosis methods is unsatisfactory in multi-fault cases, because the ability to discriminate between faulty and non-faulty cables is insufficient due to an inability to precisely quantify the possibility of an IC fault.