Trevor York

Status of electrical tomography in industrial applications

The paper surveys the status of electrical tomography for industrial applications. For the present purposes this is considered to include “low” excitation frequencies, up to about 1 MHz, comprising Resistance (ERT), Capacitance (ECT), Inductance (EMT) and Impedance (EIT) modalities. Introductory background material is followed by comparisons between the instruments that have emerged. Data processing is considered with emphasis on the trade-offs that are necessary in implementing algorithms for reconstructing images. Recent applications are tabulated and case studies are presented for six contrasting areas that illustrate significant progress towards industrial benefit. Some comparison is made with applications to medical tomography and a number of issues are identified for future research.

Design and application of a multi-modal process tomography system

This paper presents a design and application study of an integrated multi-modal system designed to support a range of common modalities: electrical resistance, electrical capacitance and ultrasonic tomography. Such a system is designed for use with complex processes that exhibit behaviour changes over time and space, and thus demand equally diverse sensing modalities. A multi-modal process tomography system able to exploit multiple sensor modes must permit the integration of their data, probably centred upon a composite process model. The paper presents an overview of this approach followed by an overview of the systems engineering and integrated design constraints. These include a range of hardware oriented challenges: the complexity and specificity of the front-end electronics for each modality; the need for front-end data pre-processing and packing; the need to integrate the data to facilitate data fusion; and finally the features to enable successful fusion and interpretation. A range of software aspects are also reviewed: the need to support differing front-end sensors for each modality in a generic fashion; the need to communicate with front-end data pre-processing and packing systems; the need to integrate the data to allow data fusion; and finally to enable successful interpretation. The review of the system concepts is illustrated with an application to the study of a complex multi-component process.

Microelectronic capacitance transducer for particle detection

This paper describes the CMOS circuit design of a sensor for detecting changes of capacitance due, for instance, to the incidence of particles or bubbles on the electrodes. The circuit is based on a simple design originating at the University of California, Berkeley, for measuring crosstalk on integrated circuits. The basic front-end sensor circuit comprises eight MOSFETs and has a sensitivity of 40 mV/fF. A differential amplifier receives the outputs from two sensor circuits each having 20-/spl mu/m square inter-digitated electrodes. The resulting sensitivity of the fabricated sensor is 1 V/fF with a noise level equivalent to 10 aF. Monte Carlo circuit simulations have been used to identify transistor dimensions to yield acceptable yield, and prototype custom silicon chips have been fabricated using a 0.8-/spl mu/m CMOS process. Static and dynamic tests, using polyamide particles as small as 10-/spl mu/m diameter, verify correct operation of the sensors. The sensor is now being developed for application in miniature electrical tomography systems.

Wireless sensor networks for industrial processes

The paper describes our work to explore the use of wireless sensor networks for industrial processes. Long term challenges including communication in a hostile environment, ad hoc networking, computing platforms, process imaging, sensing, miniaturization, compliance, micro-electromechanical systems and power harvesting are introduced. The issues are generic for applications to industrial process but the present work is targeted at monitoring grain storage. One requirement is to provide estimates of local position in a vessel and the paper describes progress towards using RF signal strength in the network for this purpose.

Agile Sensing Systems for Tomography

The concept of Agile Tomography is introduced and exemplified by reviewing the progress in tomography sensors and systems which can be deployed in situ. Agile tomography capabilities are examined across a number of electromagnetic and electrical modalities, ranging from gamma-rays to low-frequency electrical measurements. The recent achievements in already established areas are highlighted, as well as emerging technology and new modalities.

Three-dimensional electrical impedance tomography applied to a metal-walled filtration test platform

The first true three-dimensional image reconstructions from a metal-walled vessel using electrical impedance tomography (EIT) are presented. Two image reconstruction techniques have been applied via relatively sophisticated FEM modelling of a bespoke laboratory test vessel from which data have been obtained using an EIT instrument designed to intrinsically safe requirements. A generalized Tikhonov regularization method is compared with the linear back-projection (LBP) technique. Subsequent image reconstructions strongly suggest that the LBP method when applied to a metal-walled vessel is highly sensitive to the level of detail within the FEM model. By comparison, the regularized technique is far less sensitive to the complexity of the modelled geometry. Additionally, unlike the LBP method, the regularization technique has been successful in accurately reconstructing multiple inhomogeneities within an aqueous system. A further experiment has shown similar sensitivity in a wetted powder-based system. It is concluded that EIT via a regularized difference imaging approach has significant potential for detecting 3D malformations and non-uniformities in industrial pressure filtration systems.

A miniature electrical capacitance tomograph

The paper describes a miniature electrical capacitance tomography system. This is based on a custom CMOS silicon integrated circuit comprising eight channels of signal conditioning electronics to source drive signals and measure voltages. Electrodes are deposited around a hole that is fabricated, using ultrasonic drilling, through a ceramic substrate and has an average diameter of 0.75 mm. The custom chip is interfaced to a host computer via a bespoke data acquisition system based on a microcontroller, field programmable logic device and wide shift register. This provides fast capture of up to 750 frames of data prior to uploading to the host computer. Data capture rates of about 6000 frames per second have been achieved for the eight-electrode sensor. This rate could be increased but at the expense of signal to noise. Captured data are uploaded to a PC, via a RS232 interface, for off-line imaging. Initial tests are reported for the static case involving 200 µm diameter rods that are placed in the sensor and for the dynamic case using the dose from an inhaler.

Imaging nylon polymerisation processes by applying electrical tomography

The paper presents an application of electrical tomography for imaging a nylon polymerisation process at elevated temperatures (up to 275°C) and pressures (up to 22 bar). The process was realised within a metal walled tank using a sensor constructed in the form of a ceramic sleeve with electrodes coated on the inner surface. The images obtained show the dynamic behaviour of the process and how the electrical properties of the material varied with time.

Towards process tomography for monitoring pressure filtration

This paper reports on progress towards the first continuous application of electrical impedance tomography to a production scale industrial process. It includes the design and implementation of the worlds first certified intrinsically safe electrical tomography system. Zener barrier (ZB) modules and intrinsically safe relays provide electrical isolation and the instrument is certified for operation in a Zone 0 environment. Two systems have been operating successfully on production pressure filters for two years. A sensor architecture has been implemented that is compliant with the process such that it is not detrimental to efficiency or the integrity of the associated vessel structure. The MATLAB-based EIDORS three-dimensional software toolkit has been employed to yield images from simulated data. To overcome problems that arise due to dramatic changes in the conductivity of materials in the vessel during the process, a new approach is introduced that proposes the use of dynamic sensitivity maps that evolve during the batch based on the predicted mean conductivity at each stage. Real data present challenges due to a number of factors, notably the ZB modules that are integral to the intrinsically safe instrumentation. However, a simple approach to data processing has yielded process relevant information using real tomographic data.

Particle detection using an integrated capacitance sensor

Capacitance sensors, for detecting small particles, have been integrated onto a custom CMOS VLSI chip. The sensors employ a charge measuring circuit and differential amplifier which depend strongly on close matching of components for correct operation. Finite element modelling has revealed the magnitude of anticipated capacitance changes which, for particles on the surface of the chip, may be of the order of a femtofarad. Simulations using computational fluid dynamics (CFD) suggest that, if the sensor is inclined to the flow, the particles can be expected to hit the surface. Consideration of the surface of the chips suggest that surface roughness is not significant. Static tests reveal output voltage changes of several hundred millivolts in the presence of small particles on the surface. This drops dramatically as particles are moved vertically away from the surface. Dynamic tests confirm that the sensor is able to detect plastic particles as small as 20 μm that are carried at velocities up to about 10 m/s with a time resolution of 2 μs. Recent work has involved the capture of high speed video images of particles passing the chip in order that we can verify correct operation of the sensor. Representative video images are presented.