Process imaging is the art of visualizing events inside closed industrial processes. Image processing is the art of mathematically manipulating digitized images to extract quantitative information about such processes. Ongoing advances in camera and computer technology have made it feasible to apply these abilities to measurement needs in the chemical industry. To illustrate the point, this paper describes several applications developed at DuPont, where a variety of measurements are based on in-line, at-line, and off-line imaging. Application areas include compounding, melt extrusion, crystallization, granulation, media milling, and particle characterization. Polymer compounded with glass fiber is evaluated by a patented radioscopic (real-time X-ray imaging) technique to measure concentration and dispersion uniformity of the glass. Contamination detection in molten polymer (important for extruder operations) is provided by both proprietary and commercial on-line systems. Crystallization in production reactors is monitored using in-line probes and flow cells. Granulation is controlled by at-line measurements of granule size obtained from image processing. Tomographic imaging provides feedback for improved operation of media mills. Finally, particle characterization is provided by a robotic system that measures individual size and shape for thousands of particles without human supervision. Most of these measurements could not be accomplished with other (non-imaging) techniques.

Surface defects on metallic and non-metallic components can have serious effects on the reliability of the complete system. Typical examples are pistons with surface defects that cause scratching of the cylinder surface. For this reason it is of utmost importance to detect such defects as early as possible in the production process. So far, no real time solutions exist that fully satisfy industrial requirements in terms of speed, accuracy and reliability. Hence it is common to use visual inspection by humans which is error prone and causes additional personal costs. In this paper a new approach to obtain real time behaviour of industrial image processing systems by using multivariate techniques is presented. This methodology is originally used in chemometrics for statistical evaluation of measurement data and is now applied to image processing to take advantage of the high numerical efficiency of the underlying mathematics. Multivariate techniques can be applied to both the problem of automatic identification and classification of surface defects with digital images. The key to the envisaged real time ability is the high numerical efficiency of the proposed multivariate method. It manages defect detection with vector/matrix multiplication only - no calculation of powers or exponential functions is required. This enables efficient real time implementations on DSP platforms which are profiled for this type of calculations.

Information from on-line imaging sensors has great potential for the monitoring and control of quality in spatially distributed systems. The major difficulty lies in the efficient extraction of information from the images, information such as the frequencies of occurrence of specific and often subtle features, and their locations in the product or process space. This paper presents an overview of multivariate image analysis methods based on Principal Component Analysis and Partial Least Squares for decomposing the highly correlated data present in multi-spectral images. The frequencies of occurrence of certain features in the image, regardless of their spatial locations, can be easily monitored in the space of the principal components. The spatial locations of these features can then be obtained by transposing highlighted pixels from the PC score space into the original image space. In this manner it is possible to easily detect and locate even very subtle features from online imaging sensors for the purpose of statistical process control or feedback control of spatial processes. The concepts and potential of the approach are illustrated using a sequence of LANDSAT satellite multispectral images, depicting a pass over a certain region of the earth’s surface. Potential applications in industrial process monitoring using these methods will be discussed from a variety of areas such as pulp and paper sheet products, lumber and polymer films.

Extracting texture/roughness information from grayscale or multispectral images for off-line quality control, or on-line feedback control is a difficult problem. Several statistical, structural & spectral texture analysis approaches for grayscale images (using various pre-defined filters etc.) have been suggested in the literature1, 2 In this paper we propose a new approach based on Multivariate Image Analysis techniques using multi-way Principal Component Analysis. Prior to analysis the grayscale images are transformed into three-dimensional pixel intensity arrays through spatial shifting of the image in several directions followed by stacking the shifted images on top of each other. The resulting three -dimensional image data is a multivariate image where the third (i.e. variable) dimension is the spatial shifting index. Multi-way PCA is then used to extract features (PC scores), which contain the greatest amount of variation. Plots of the observed values of these scores against one another define a score space. Certain regions of this score space contain the texture information of the grayscale image. By masking these regions and tracking the number of pixels having features that fall in these regions, or by comparing the score spaces with template exemplars, one is able to monitor changes in the image surface textural properties. The approach is illustrated using a set of grayscale images of the surface of steel sheet. Based on the textural features extracted from the surface images a simple classification scheme is devised in which each sample image is assigned into one of two classes representing good or bad surface characteristics.

This paper considers the processing of data collected by scanning gauges from industrial web processes such as metal rolling, coating, paper making and plastic film extrusion. It describes a method based upon the generalised sampling theorem for reconstructing 2-dimensional variations in material properties from scanned data. The technique is applied to methods of scanning and sampling that are commonly used in industry. The practical difficulties associated with using the technique are discussed.

The diameter setpoint of a growing crystal in the silicon Czochralski process is a key cost parameter, whose optimal choice depends in part upon the diameter calibration accuracy. Measurement of the crystal diameter during solidification is made remotely, due to high temperatures and vacuum vessel design. Vision systems for diameter control detect the diameter of the bright ring reflection from the silicon melt surface at the crystal meniscus, rather than the actual crystal diameter. Distortion due to the bright ring measurement would result in a destabilizing nonlinear diameter measurement even if the crystal diameter response were linear. Using a published model of the meniscus shape, two and three-dimensional modeling of the bright ring is performed, and simple approximations are made to predict the bright ring bias as a function of diameter slope. Tracking of a diameter maximum during vertical translation could provide a calibration measure, given accurate translation data. The use of deformable templates or snakes is suggested for tracking the diameter maximum, and is bench-tested to provide estimates of on-line calibration accuracy, a key parameter for selection of the optimum diameter setpoint. Implementation of the modified calibration strategy requires corrections for camera distortion, crystal thermal expansion and window diffraction.

The performance of 2-dimensional control systems is discussed in terms of spatial and dynamic bandwidths and is illustrated by examples from the control of cross-directional variations on a plastic film extrusion line. By relating these bandwidths to the specification of the required quality of the finished film and typical disturbances that enter the process, specifications are developed for the design of the actuators, the sensing system and the control algorithm. It is shown that the current generation of control systems is not suitable for controlling variations over small areas of sheet, but a system which regulates over much smaller areas could be developed using recent developments in sensors and 2-dimensional control algorithms.

Process tomography consists of tomographic imaging of systems, such as process pipes in industry. One typical feature for the industrial processes is that the state of the system changes fast. If the changes are very fast in comparison to data acquisition rate, the ordinary computational methods in tomography can not provide feasible reconstructions. We use state estimation in process tomography and take into account the time dependence of the object. Especially, we consider the case of the electric imaging of the moving fluid. We use the convection-diffusion equation in modeling time dependence of the target. The Kalman smoother algorithm is used for estimating the state of the object. We have previously shown that the state estimation works well in process tomography in the cases in which the fluid dynamics of the system is modeled correctly. However, in the real case the velocity field can not usually be determined accurately. This may be caused e.g. by complex nature of the flow, the turbulence, discretization, etc. In this paper we consider how the inaccuracies in the fluid dynamical model affect the state estimates in process tomography.

Level monitoring instrumentation is an essential part of hydrocarbon processing facilities, and has together with separator technology been widely addressed over the last decade. Key issues are production capacity, product enhancement and well- flow control. The reliability and accuracy of the level instrumentation, and its ability to monitor the thickness of the foam and the oil-water emulsion, are particularly important when considering the level instrumentation as the main sensing element in the automatic control of the separation vessel. Lately industry focus has been placed on optimal automatic control to improve the quality of the production output, and to minimize the use of expensive and environmentally undesirable separation enhancing chemicals. Recent developments in hydrocarbon production includes subsea separation stations, where the constraints placed on the reliability and accuracy of the level instrumentation are especially severe. This paper discuss the most common existing level monitoring technologies, and present some recent level monitoring developments for three-phase separators. In order to clarify the issue of cross sectional metering the notion tomometry is introduced in this paper. Tomometry denotes multipoint cross sectional metering aiming to acquire cross sectional information on the distribution of the substances in the process vessel for control purposes, not mainly to create a cross sectional reconstructed image of the process in question.

The subject of pneumatic conveying of solids is a complex one. The flow regime present in a conveying system is dependent upon: the size and shape of the particles to be conveyed, the geometry and orientation of the conveying pipe, the relative densities of the solid and the conveying air. The variable parameters present are the velocity of the conveying air and the solids mass flow rate. The variation of these two factors dictates the presence of either dilute or dense phase flow. At Manchester Metropolitan University a pneumatic conveying system transporting polyethylene nibs, was used to investigate the implementation of a Proportional and Integral control system using a tomographic imaging system in the feedback loop. The aim of the investigative work was to achieve control of the air velocity and solids loading factor for the conveying system to maintain dilute phase flow at a prescribed level. The solids material conveyed was sensed using a PC based electrical tomographic imaging system and this was used to control the air velocity in the conveying system.

A preliminary study of using electrical resistance tomographic imaging for bubble column measurement and control has been conducted. A number of methods for identifying flow regimes (e.g. bubbly and churn-turbulent flows) have been developed. An optimisation method is used to automatically search for the flow rate corresponding to the maximum interfacial area and determine a set-point for a PID controller. A simple flow-parameter control, based on bubble fluctuation as represented by derivatives of dynamic imaging, is reported. This work is believed to be the first published case of realtime on-line process control using electrical resistance tomography.

The computer automated radioactive particle tracking (CARPT) is a non-invasive flow monitoring technique used! for measuring mean and fluctuating velocity fields of a traced phase in a multiphase flow system. The method involves accurately monitoring the instantaneous position of a radioactive tracer particle using an array of strategically positioned scintillation detectors. A limitation to the accuracy of CARPT lies in the error associated with the reconstruction of the tracer particle position which affects the space-resolution of the technique. It is of interest, therefore, to minimize this error by choosing wisely the best hardware and an optimal configuration of CARPT detectors’ array. Such choices are currently based on experience, without firm scientific basis. In this paper, through theoretical modeling and simulation, we describe how the accuracy of a radioactive particle tracking setup may be assessed a priori. Through an example of a proposed implementation of CARPT on a gas- solids riser, we demonstrate how this knowledge can be used for choosing the hardware required for the experiment. Finally, we show how the optimal arrangement of detectors can be effected for maximum accuracy for a given amount of monetary investment for the experiment.

Methods for optical access to combustion chambers of internal combustion engines are shown. The optical access is either maximised to allow application of complex optical diagnostic techniques whilst maintaining minimum operability of the engine or engine components, or full engine operability is maintained whilst optical access and diagnostic techniques are tailored to the diagnostic demand and the restraints of engine operation. As full optical access allows a wide range of diagnostics to be applied it puts the emphasis on organising and extending realistic engine operation conditions. Full engine operability, however, challenges the optical access and the design of diagnostic systems. The paper gives examples for diagnostic applications for both of these situations.

The spatial distribution of chemical species can be a critical determinant of the performance of chemical reactors. One such reactor is the combustion chamber of the Internal Combustion engine, in which the spatial variation of air-fiiel ratio has a significant influence on both fuel efficiency and emissions performance. We report the development of a fibre-based Near Infra-Red Absorption Tomography system, in order to measure the distribution of hydrocarbons in-cylinder. The technique exploits the specific (but weak) hydrocarbon absorption of 1.7 µm radiation, which wavelength has only recently become accessible for the present application by the availability of solid-state all-optoelectronic components. A custom-specified InGaAsP/InP laser diode has been supplied, delivering 3mW at 1.700µm, with about lnm tunability. A standard telecommunications laser diode is used to provide a reference wavelength at 1.55 µm, which is not absorbed by any species in the combustion environment. Along each of 32 absorption paths through the subject, both wavelengths are launched simultaneously via a single-mode optical fibre and GRIN lens. The transmitted light is collected by a large-core fibre and measured by an extended-sensitivity InGaAs photodiode. The attenuation at each individual wavelength is measured by modulating the intensity of the laser sources in a frequency-division multiplexed scheme. The logarithm of the ratio of the two measurements yields the path integral of the hydrocarbon absorption, and hence, of concentration. Single-channel characterisation shows that the technique is readily calibrated for temperature and pressure effects, over the region 70- 150°C and 1- 10bar. Tomographic reconstruction of different gaseous hydrocarbon flows has been achieved. Design considerations will be discussed concerning the deployment of the technique to a running engine, to achieve image rates over 10,000 per second.

The basic motivation of this work comes from the fact that most of the numerical reconstruction algorithms were developed for hard sensing fields, i.e. they have the underlying assumption that the sensing field is two-dimensional and parallel. When using soft sensing fields, such as in electrical impedance tomography, these conditions are barely satisfied, unless in the situation of a low contrast between the electrical properties of the flowing phases, for which the ill-conditioned nature of the problem will be much more critical anyway. The development of a less restrictive method can be achieved by refining the qualitative images of a direct imaging probe, through the minimization of a conveniently defined error functional reflecting the difference between experimental and numerical values taken at the boundary of the sensing volume. However, in order to do this, these boundary values must be sufficiently sensitive to changes in the distribution of the electrical properties of the flow. This paper addresses this problem and presents a sensitivity analysis of different excitation strategies and their applicability in a reconstruction algorithm such as described above. Classical Dirac excitation is compared to an alternative excitation profile with regards to the capability of producing significant changes in boundary measurements when the internal organization of the sensed media is altered. Results confirm that classical strategies suffer from a major lack of sensitivity and that new ones must be developed, possibly based on the optimization of the excitation profiles or on multisensing strategies.

The selection of ballistic photons from a radiation, passed through a biological medium, is necessary for obtaining qualitative images in optical tomography. Within the framework of a non-stationary two-flux model of radiation transport in high scattering medium there is found the fraction of ballistic photons as function on macroscopic characteristics of a medium. With the purpose of separation of photons the application of bisphthalocianine dyes of rare-earth and transitional elements is considered. The experiments on deriving contrasted images in margarine and water solution of milk are carried out.

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.

This paper presents an overview of an integrated multi-modal system designed to support a range of common modalities: electrical resistance, electrical capacitance and ultrasonic tomography. Many complex processes exhibit behaviour that changes over time and space. Such properties demand equally diverse sensing modalities. A 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 reviews 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.

Tomographic techniques are most frequently used as a research tool but they are now being implemented in industrial environments. A development of sensor design and manufacturing procedures for electrical tomography is described. Issues such as sensor multi-modality, robustness and maintenance are addressed and illustrated by examples of sensor design for the oil and chemical industries. These sensors were constructed for imaging and controlling multi-phase and multicomponent processes at elevated temperatures and pressures in an environment where an acute deposition may occur. The paper discusses two important and diverse issues namely multi-modality and a development of on-line and in-situ electrode cleaning methods.

In the production of steel strip, the temperature distribution and cooling rates along the mill run-out table have a significant effect on the steel microstructure and hence on final material properties, such as yield strength, tensile strength and ductility. Consequently, demands for improved product consistency and a greater diversity of sophisticated steel grades have increased the requirements for tighter control of these process conditions. Non-contacting optical temperature sensors are typically used to implement feedback control of cooling. Unfortunately water spray variations and surface emissivity irregularities can adversely affect these sensors. In addition, temperature is only used as an assumed indicator of microstructure and only the surface of the steel is measured. Ideally, the control of cooling path should take account of the progress of dynamic transformation at required points rather than the strip temperature alone. There are a number of reports describing the use of magnetic sensors to monitor transformation. These sensors exploit the change in the electromagnetic properties as the steel progresses through transformation, for example the austenitic phase is paramagnetic and the ferritic phase is ferromagnetic below the Curie point. Previous work has concentrated on the operation and design of individual transformation sensors. This paper now describes the novel use of an array of electromagnetic sensors to image the progression of transformation along a sample steel block on a pilot scale industrial mill. The paper will describe the underlying physical principles, the design of the system and present images showing the progress of transformation along one surface of the sample.

It is highly desirable to be able to interrogate in 3-D the whole fluid inside a typical stirred vessel, especially when used as a chemical reactor. Electrical resistance tomography (ERT) can scan the fluid with a spatial discrimination of the order of O(103) voxels with video frame rate acquisition speeds. Some new augmented-reality results are presented for a 2.3m3 pilot scale vessel detecting miscible fluid mixing of a brine tracer pulse and solid-liquid mixing. The images are created by solid- body iso-surfaces between which are variable opacity layers. Extra augmentation can be achieved by the use of colour, giving a 5-D representation of the mixing process. For miscible mixing, the geometry and spatial patterns clearly highlight the effect of injection position on mixing rates. For the solid-fluid mixing, it is possible to visualise in 5-D the way solid resides on the vessel base behind baffles as well as indicate the height to which solids are effectively suspended. The dynamics of mixing after a charge of solids can also be captured and visualised using augmented-reality. The facility to envisage mixing information in up to 5-D offers new possibilities for control action linked to spatial distributions of phases and components.

It is difficult to know the details of material distribution in a fluidised bed. Electrical capacitance tomography (ECT) technology has been developed for imaging industrial processes containing dielectric materials. This paper discusses the principles of ECT and presents the first experimental results of using ECT to measure gas/solids distribution in square circulating fluidised beds (CFB). A wide range of fluidisation conditions was tested, from bubbling to circulating fluidisation. It has been shown that over the range, ECT can provide solids concentration and voids distribution instantaneously. Different fluidisation regimes could also be identified from the reconstructed images.

Stray-immune circuits are available that will measure very small floating capacitances at high speed These circuits have facilitated the development of Electrical Capacitance Tomography (ECT), which uses fixed, robust metal plates as sensors, arranged non-intrusively around the periphery of the volume of interest. The phenomenon of flame ionization is well understood, with general agreement that the concentration of all the major ions maximize at the flame front. There is also strong experimental evidence to suggest that the presence of carbon particles (and their precursors) results in a large increase in the charge carrying capacity of the flame. This paper reports some initial results from using ECT to visualize the position, size, composition and movement in flames in open and closed combustion chambers.

The ultimate aim of the work described here is to explore miniaturised electrical tomography systems. The initial target, described in this paper, has been to implement a sensor, integrated on silicon, and able to detect particles as small as 10 pm in diameter from changes in capacitance. 3D finite element modelling has been used to determine the capacitance changes due to the presence of small particles and to explore electrode geometries to give optimum sensitivity. Interdigitated electrodes, measuring 25µm on a side, have been integrated alongside measurement circuitry using a 0.8µm CMOS process. The measurement circuitry is able to detect changes in capacitance down to tens of attofarads. Computational fluid dynamics has been used to determine the likely trajectories of particles in the vicinity of the sensor and confirm that they are likely to hit the surface. Static tests verify the sensitivity of the sensor to horizontal and vertical displacement of small particles. Dynamic tests show that the sensor is able to detect the presence of plastic particles, as small as 20µm diameter, conveyed in an air stream moving at up to 10 m/s. Time resolution is currently 2 µs. Recent work has yielded high speed video images of the particles as they hit the surface of the chip and these are to be used to verify correct operation of the sensors.

Microwave tomographic imaging techniques have mainly been studied for medical applications in the past two decades. In recent years, however, there has been increased interest in the application of microwave imaging techniques for industrial processes or multiphase flows. In the medical case, water has been used as the background material, with microwave antennas and the object immersed in water, and the contrast of the object is measured against the dielectric properties of water. For industrial application, it is more convenient to use air as the background medium. However, this leads to a large contrast problem if the material being imaged contains a large amount of water. Consequently, the image reconstruction algorithms need to be more adaptive to the level of contrast and uncertainty in the initial guessed values in the iterative reconstruction process. The electromagnetic noise in the open air environment would usually be higher than that in water as a result of the surrounding industrial noise, and the near field region is much larger in air than that in water at the same operating frequency. Therefore, the algorithms need to be less sensitive to the effect of noise. In this paper, two algorithms based on the Newton- Kantorovich and Conjugate Gradient error minimisation methods are investigated with a view to their applications in the imaging of industrial processes using air as the background medium. The results on the effect of noise and the images reconstructed using the algorithms are presented.

This paper describes a new method for quantitative image reconstruction with electromagnetic inductance tomography (EMT) using a parameterised finite-element (FE) based model to solve the forward problem. The algorithm has been applied to a 2D planar array to locate the position and size of circular steel bars. A commercial FE solver is used within the algorithm as a convenient and practical means of modelling the response of the sensor. The paper presents results from a number of object distributions and concludes with some suggestions on how this approach could be generalised to address a variety of EMT reconstruction problems.

Montell Carrington Limited produces polyethylene and polypropylene nibs, which are sold in bulk form to companies who use them to make products for the consumer market. The nibs are stored in 30 storage bunkers, each with a capacity of 500 tonnes. The external distribution of nibs is achieved using 40 tonne road tankers that are filled from each of the bunkers using gravity feed. Work has been undertaken at the Manchester Metropolitan University, in collaboration with Montell, to develop a Variable Density Flowmeter using Process Tomography that will enable the mass flow of nibs to be measured with an accuracy of ±2% and hence control the loading of the road tankers. The flowmeter (260mm diameter) was situated between the bunker discharge outlet valve and the tanker. Measurement of the density distribution across the pipe, using Process Tomography, enabled the mass flow into the road tankers to be determined. The Montell Process Tomography (MPT) system was a PC based system incorporating Texas Instruments C40 parallel processors and a 12 electrode capacitance measuring system with a driven axial shield. The capacitance detector was an AC bridge detection circuit working at 100kHz, a demodulator and a back projection algorithm were used to obtain the process images.

The paper deals with the goal of component fraction estimation in multi-component flows, a critical measurement in many process systems. Electrical Capacitance Tomography (ECT) is an attractive sensing technique for this task, due to its low- cost, non-intrusion and fast response. However, typical systems, which include practicable real-time reconstruction algorithms have shown to give inaccurate results and the existing approaches to direct component fraction measurement have a performance that is typically flow-regime dependent, and they fail to discriminate fractions in three-component flows. Such systems also depend upon an intermediate image that must be interpreted to yield useful plant data. In the investigation described, an artificial neural network approach has been used to directly estimate the component fractions in gas-oil, gas- water and gas-oil-water flows from ECT measurements. A two-dimensional finite-element electric field model of a 12- electrode ECT sensor has been used to simulate measurements in stratified, annular and bubble-flow conditions. The singular-value decomposition has been used to reduce the raw measurement data to a mutually independent set. Multi-Layer Feed-Forward Neural Networks (MLFFNNs) have been trained with sets of such reduced ECT data with their corresponding component fractions. The trained MLFFNNs have been tested with test patterns consisting of unlearned ECT data. The paper reviews results of the best-trained networks that give a mean absolute error of less than 1% for the estimation of various multi-component fractions. The MLFFNNs’ estimations are also compared with a direct ECT method proposed in one of the previous works. The direct ECT method gives larger mean absolute errors than the MLFFNNs, demonstrating that artificial neural systems provide more accurate component fraction estimations.

Control of molten steel delivery in continuous casting is critical to ensure stability of the meniscus and satisfactory mould flow patterns, which in turn are determinants of steel cleanness and surface quality. Considerable effort has been expended over the last 10 years in optimizing the design of the metal delivery system, particularly the pouring nozzle, to enable the consistent production of high quality steel at high throughput. This paper looks forward to possible systems that are capable of topographically imaging the distribution of molten steel flows in these applications. The paper will concentrate on the feasibility of using electromagnetic methods. The paper will present some initial results and an overview of the image reconstruction process used will also be included. The paper will conclude with a discussion of possible future developments, such as the use of a tomographic or multifrequency approach, future research on the reconstruction image procedures and the potential for visualisation and flow measurement. There is a need for further research in this area and some priority areas for future work will be suggested.

There are many principles for interface level detection in separation tanks based on capacitance, ultra sound, microwave, nuclear radiation etc. These principles work well in many situations, in others they fail. The high frequency magnetic field principle works in most of the situations that will occur in separation tanks for crude oils for detection of the clean water level, the layers of water continuous water/oil emulsion and the oil continuous oil/water emulsion, the oil level, the thickness of the foam layer and the gas. When a coil is dipped into a fluid its electrical impedance will be dependent on the characteristics of the fluid. If the material is electrical conductive the impedance of the coil will be reduced due to the eddy currents induced in the material, setting up a magnetic field directed against the field generated by the coil. The inductance will increase but still remain low also in the water continuous water/oil emulsion zone, but will rapidly increase in the oil continuous oil/water emulsion zone. In pure crude oil the inductance will be high and even higher in gas. The coil inductance is measured by connecting the coil to a LC-oscillator.

Using a conventional electrical capacitance tomography (ECT) sensor, it is difficult to image either small objects or objects in the central area. For the measurement of a thin layer of solids found near the walls of a circulating fluidised bed (CFB), a new type of sensor is proposed, having conventional external array of electrodes plus an additional internal array of electrodes. Sensitivity maps for this sensor were generated using a finite difference method (FDM) and its performance in image reconstruction evaluated using both a linear back-projection (LBP) and an iterative algorithm. Compared with a square external-electrode sensor, the new sensor has higher sensitivity in the central area of the sensor, and thus is able to produce improved images. This has significant advantages for imaging the well-known core-annulus structure of solids distribution in CFB combustors.