Industrial Process Tomography - Platform II grant funded by EPSRC

FACULTY OF ENGINEERING

 

Colloid Vibration Potential Imaging

Infratography – wide area sensing

Industrial tomography applications typically present the challenge of gaining spatial distribution data from observations outside the process boundary. In contrast, in many large scale and wide area applications the major challenge is gaining effective coverage of the enclosed space using practicable resources at a reasonable cost. Distributed infrastructures such as mining operations and open processes require a variety of sensing and actuation duties and typically feature an intrinsic spatial organization. Wireless sensor network (WSN) technology now offers a viable platform for distributing a large number of sensing (and actuation) elements within a wide area. Although their communication capabilities are powerful, WSN’s offer only crude facilities for process monitoring in principle, through the periodic transmission of data from each node to a base node.

The term Infratronics describes heterogeneous wired and wireless networks, augmented by smart cooperating agents (Fig. 1) [1]. These enable a variety of distributed monitoring and control functions, such as high level interpretations related to the overall role or mission of the system:

Is the system operating normally?
Is there a fault or exception, and if so what are its extents and states?
What is the distribution of a spatial variable?

Where nodes have knowledge of their location their spatial distribution is an emergent network property which can be exploited to deliver smart operation in which multiple tasks can be consigned to a group of nodes. Specific spatial awareness properties can be realised through simple cooperation mechanisms within the network, both for fixed and mobile nodes, realising an Infratronic network [2].

Fast EIT System
Figure 1: Spatial agent cooperation in heterogeneous wired and wireless networks

These highly efficient local cooperation mechanisms enable distributed tasks that can realise holistic application focussed operations. For example, to automatically detect the cardinal (highest or lowest) value in a wide area distribution, and periodically send a single directed alert message to a monitoring point. A simple message can notify the network of an interest in a threshold level, perhaps a gas concentration value, and again nodes having this level can alert this status (Fig. 2). This may be configured as the tracking of a moving isocline in which the set of points is contained in an alert message (Fig.3) [3]. These facilities offer major benefits in the autonomous and distributed monitoring of wide area infrastructures – the new capability of Infratography.

Fast EIT System
Figure 2: Cardinal and Threshold operations – (CardThreshold.jpg)


Fast EIT System

Figure 3: Isocline tracking – (isoclinetrk.jpg)

Author Information: Brian Hoyle, Institute of Particle Science and Engineering, School of Process, Environmental and Materials Engineering, Email: b.s.hoyle@leeds.ac.uk

References:

B. S. Hoyle, Cooperating sensor nodes in spatially critical networks for smart measurement and monitoring systems, Proc Smart Object Systems Workshop, UbiComp 2005, Tokyo, pp77-82., September 2005

B. S. Hoyle, Location Aware Distributed Agents in Intelligent Environments, 5th International Conference on Intelligent Environments, Barcelona, Spain, doi: 10.3233/978-1-60750-034-6-161, pp161-168,  July 2009.

J. A. Baloch and B. S. Hoyle, Smart Spatially-Aware Sensing and Actuation System, IEEE Sensors 2010 Conference, November 2010, to be published.