Abstract

An approach for intelligent monitoring of mobile cyberphysical systems is described, based on consensus among distributed self-organised agents. Its usefulness is experimentally demonstrated over a long-time case study in an example domain: a fleet of city buses. The proposed solution combines several techniques, allowing for life-long learning under computational and communication constraints. The presented work is a step towards autonomous knowledge discovery in a domain where data volumes are increasing, the complexity of systems is growing, and dedicating human experts to build fault detection and diagnostic models for all possible faults is not economically viable. The embedded, self-organised agents operate on-board the cyberphysical systems, modelling their states and communicating them wirelessly to a back-office application. Those models are subsequently compared against each other to find systems which deviate from the consensus. In this way the group (e.g., a fleet of vehicles) is used to provide a standard, or to describe normal behaviour, together with its expected variability under particular operating conditions. The intention is to detect faults without the need for human experts to anticipate them beforehand. This can be used to build up a knowledge base that accumulates over the life-time of the systems. The approach is demonstrated using data collected during regular operation of a city bus fleet over the period of almost 4 years.

Highlights

  • Current approaches for equipment monitoring, i.e., fault detection, fault isolation and diagnostics, are based on creating some form of a model, see e.g., Isermann (2006), Jardine et al (2006), Hines and Seibert (2006), Hines et al (2008a, b), Peng et al (2010), Ma and Jiang (2011) for reviews

  • This can be done by embedded software agents that search for interesting relationships among the signals available inside internal Electronic Control Units (ECUs)

  • This kind of situation was experienced in November 2011, when a front wheel speed sensor had been deviating for a month, the on-board diagnostic (OBD) system warned about an Electronic Braking System (EBS) problem and the driver complained about weak brakes

Read more

Summary

Introduction

Current approaches for equipment monitoring, i.e., fault detection, fault isolation and diagnostics, are based on creating some form of a model, see e.g., Isermann (2006), Jardine et al (2006), Hines and Seibert (2006), Hines et al (2008a, b), Peng et al (2010), Ma and Jiang (2011) for reviews. In the simplest case this model is a range that a signal should be within, but it can be a physics based reference model constructed prior to production and later compared to the actual operation of the system; or a pattern recognition model that is trained from collected data and later compared either to the real performance, or used directly to label the operation as normal or abnormal Building such models requires a significant amount of “manual” expert work, e.g., trying out different model structures, different feature sets, collecting data, formulating suitable residuals to monitor, etc. A third contribution is the idea of how one can build a knowledge base from co-occurrences of faults repaired (as described in genuine maintenance records) and disappearances of fault signals

Related work
The COSMO approach
Data models
Histograms
Autoencoders
Linear functions
Detecting deviations
Evolving novelty detection framework
Description of the data
Uptime and downtime for the bus fleet
Application of COSMO to the city bus fleet
Linear relations
Knowledge discovery
Findings
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call