Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
APPARATUS AND METHOD FOR FLUID ANALYSIS
Document Type and Number:
WIPO Patent Application WO/2019/202124
Kind Code:
A1
Abstract:
A method for analysing particulates in a liquid is disclosed. The method comprises receiving a liquid for analysis via a flow cell (110); illuminating a first surface (112) of the flow cell using a first bandwidth; illuminating a second surface (114) of the flow cell using a second bandwidth; and capturing one or more images of the liquid during illumination by the first and second bandwidths.

Inventors:
MERRITT PAUL (GB)
SCOTT LINDA (GB)
TALMAGE DANIEL (GB)
Application Number:
PCT/EP2019/060190
Publication Date:
October 24, 2019
Filing Date:
April 18, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
CASTROL LTD (GB)
International Classes:
G01N15/10; G01N15/14; G01N21/94; G01N33/28; G01N15/00
Domestic Patent References:
WO1994014049A11994-06-23
Foreign References:
CN103983543B2017-01-04
CN106568695A2017-04-19
CN101943565A2011-01-12
US9891160B12018-02-13
CN201051080Y2008-04-23
Other References:
T. H. WU ET AL: "A New On-Line Visual Ferrograph", TRIBOLOGY TRANSACTIONS, vol. 52, no. 5, 8 September 2009 (2009-09-08), US, pages 623 - 631, XP055603739, ISSN: 1040-2004, DOI: 10.1080/10402000902825762
Attorney, Agent or Firm:
TOWLSON, SAMANTHA JAYNE (GB)
Download PDF:
Claims:
CLAIMS

1. A method for analysing particulates in a liquid, comprising: receiving a liquid for analysis via a flow cell; illuminating a first surface of the flow cell using a first bandwidth; illuminating a second surface of the flow cell using a second bandwidth; and capturing one or more images of the liquid during illumination by the first and second bandwidths.

2. The method of claim 1, comprising receiving the liquid whilst the liquid whilst the liquid is in motion.

3. The method of any preceding claim, further comprising applying an optical filter to the first bandwidth.

4. The method of any preceding claim wherein the first bandwidth corresponds to a colour absent from one or more of particulates in the liquid. 5. The method of any preceding claim, comprising illuminating the second surface using a plurality of light sources.

6. The method of any preceding claim, comprising illuminating the second surface at an obtuse angle with respect to a long axis of the flow cell.

7. The method of any preceding claim, comprising illuminating the second surface co- axially along an optical axis of image capture.

8. The method of any preceding claim, comprising illuminating the first and second surfaces of the flow cell simultaneously.

9. The method of any preceding claim, wherein the second bandwidth comprises wavelengths within the visible, infrared, or ultraviolet bands.

10. The method of any preceding claim, comprising illuminating the first and/or second surface of the flow cell using a light emitting diode (LED).

11. The method of any preceding claim, wherein the one or more images are captured using a microscope objective and a light sensor; optionally the light sensor is a camera. 12. The method of any preceding claim, comprising storing the one or more images in a storage means.

13. The method of any preceding claim, further comprising: receiving the one or more images and determining one or more of a physical or chemical property of a particulate present in the received images. 14. The method of claim 14, wherein determining the physical property comprises determining one or more of a size, two-dimensional shape, three-dimensional shape, surface texture, surface colour, magnetic susceptibility, contour, area, centroid, bounding box, aspect ratio, angle, convexity, circularity, colour, brightness, and surface topology.

Description:
APPARATUS AND METHOD FOR FLUID ANALYSIS

This invention relates to an apparatus and method for analysing particulates in a fluid.

BACKGROUND

Analytical ferrography is a technique that can be used to monitor the wear and tear of machinery components by identifying and analysing particulates and contaminants present in a sample fluid (for example a lubricating oil) associated with the machinery. Information regarding the lifespan, environment, and usage of the machinery may be inferred from characteristics associated with the identified particulates. As such, ferrography may be used to predict and diagnose potential faults and errors occurring or about to occur on the machinery.

It is generally desired to have a fast and precise particulate analysis procedure, wherein faults in the machinery can be deduced accurately from a fluid sample.

It is an aim of the present invention to address one or more disadvantages associated with the prior art.

SUMMARY OF THE INVENTION

An aspect of the present invention provides a method analysing particulates in a liquid, comprising: receiving a liquid for analysis via a flow cell; illuminating a first surface of the flow cell using a first bandwidth; illuminating a second surface of the flow cell using a second bandwidth; and capturing one or more images of the liquid during illumination by the first and second bandwidths.

The method preferably comprises receiving the liquid whilst the liquid whilst the liquid is in motion.

The method may further comprise applying an optical filter to the first bandwidth.

Preferably, the first bandwidth corresponds to a colour absent from one or more of particulates in the liquid. Preferably, the method comprises illuminating the second surface using a plurality of light sources.

Preferably, the method comprises illuminating the second surface at an obtuse angle with respect to a long axis of the flow cell.

Alternatively or additionally, the method preferably comprises illuminating the second surface co-axially along an optical axis of image capture.

Preferably, the method comprises illuminating the first and second surfaces of the flow cell simultaneously.

Preferably, the second bandwidth comprises wavelengths within the visible, infrared, or ultraviolet bands.

The method of any preceding claim, comprising illuminating the first and/or second surface of the flow cell using a light emitting diode (LED).

Preferably, the one or more images are captured using a microscope objective and a light sensor; optionally the light sensor is a camera.

The method may comprise storing the one or more images in a storage means.

The method may further comprise receiving the one or more images and determining one or more of a physical or chemical property of a particulate present in the received images.

Preferably, determining the physical property comprises determining one or more of a size, two-dimensional shape, three-dimensional shape, surface texture, surface colour, magnetic susceptibility, contour, area, centroid, bounding box, aspect ratio, angle, convexity, circularity, colour, brightness, and surface topology.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described with reference to the accompanying drawings, in which:

FIG. 1 A shows a block diagram of an apparatus according to embodiments of the invention;

FIG. IB shows a block diagram of an alternative apparatus according to embodiments of the invention;

FIG. 2 shows a block diagram of a section of the apparatus;

FIGs. 3A-3C show an example of the apparatus in operation; FIGs. 4A-4F show an example of the apparatus in operation; and

FIG. 5 shows a method in accordance with embodiments of the invention.

DETAILED DESCRIPTION

Typical ferrography procedures comprise obtaining a sample of lubricating oil from machinery and separating out the particulates present in the oil sample for analysis. The sample may first be chemically diluted to improve particulate precipitation and adhesion, and then arranged to travel down a glass slide to deposit particulates on the slide. A magnet may be used such that ferromagnetic particulates present in the oil are attracted and separated at distances along the slide corresponding to their magnetic properties. Nonmagnetic particulates are randomly deposited along the length of the slide. Once this process is complete, the slide is washed clear of lubricant and analysed under a

microscope. By studying the parameters of the particulates obtained from the sample, information relating to the wear of the machinery from which the sample was taken may be inferred. However, soft particulates such as bubbles, or information about the actual lubricant itself may not be detected since they are washed of the slide with the lubricant during sample preparation. The ferrographic process, although accurate and informative, is labour-intensive, with the analysis of the particulates requiring considerable expertise and a wide-ranging knowledge of such particulates to be able to derive meaningful information regarding wear. Thus existing techniques of analytical ferrography require a sizeable analytical laboratory, are costly, time-consuming, and hence have limited application. Whilst in some industries these issues are less important, for some, the inability of the technique makes the analysis impractical. Current ferrographic techniques are therefore not economically viable for relatively simple applications (such as in the consumer vehicle industry) or in areas where it is not feasible to locate an analytical laboratory (a wind turbine) or to send samples to an analytical laboratory without risking sample degradation (remote land or sea locations). A further drawback of traditional ferrographic techniques is that the used lubricant and particulates are separated and the lubricant discarded, so it is not possible to view or analyse the particulates in-situ in the lubricant. Due to this, only static images of the particulates can be provided.

Embodiments of the present invention offer alternatives to these traditional ferrographic techniques. FIG. 1 A shows a block diagram of an apparatus 100 in accordance with an embodiment of the present invention for performing wear analysis on fluid samples. The apparatus 100 is arranged to image objects within a fluid, in particular but not limited to particulates in lubricant, as they move through a flow cell. The term “particulates” as used herein comprises metallic bodies (including but not limited to shards, fragments and flakes), non-metallic bodies and other inclusions (such as liquid and/or gaseous bubbles). Therefore the apparatus 100 may also be used to analyse both ‘hard’ and‘soft’ particulates, such as fluid-in-fluid or gas-in-fluid particulates, bubbles, or molecules of lubricant additives, such as anti -foams and surfactants. In addition, the technique in accordance with embodiments of the invention is dynamic and nondestructive, in contrast with the traditional ferrographic analysis techniques described above. Furthermore, embodiments of the present invention have the further advantages that they are portable, and not lab-bound; may be used in-line and continuously rather than in a batch process as in traditional techniques; and have a small footprint (less than lm 2 ). Measurement of the particulates in-situ in the lubricant offers a sizeable advantage over traditional ferrographic techniques.

The apparatus 100 comprises at least a flow cell 110 (also known as a microfluidic cell), a first light source 120, a second light source 130, and an image capture means 140. The apparatus 100 is connected to an analysis means 150 and a data storage means 160 for analysing and processing data from the apparatus 100. The first and second light sources 120, 130 are configured to illuminate a fluid within the flow cell, and image capture means 140 is configured to capture images of particulates within the fluid in the flow cell 110. In an embodiment the flow cell 110 is arranged to receive the fluid whilst the fluid is in motion. This enables dynamic images of a moving particulate to be seen.

The components of the apparatus 100 are chosen firstly, to enable a sample to follow a predictable route through the apparatus 100, free from anything that may impede the fluid flow so as to block it completely and secondly, to optimise the optical quality of the images captured.

A fluid source 170 is fluidly connected to the flow cell 110, which is arranged to receive fluid for analysis. In the specific example of FIG. 1A the fluid source comprises a pump 174 coupled to a fluid reservoir 172. The fluid reservoir 172 may be a sample bottle having a sealed lid through which tubing or pipework carrying the fluid passes. The pump 174 is chosen to minimise any damage to the particulates under analysis, with a peristaltic pump being particularly preferred. A flow sensor 176 may be coupled to the ingress of the flow cell 110 and arranged to determine a flow rate of fluid travelling into the flow cell 110. Alternatively, the flow sensor 176 may be coupled to the exit of the flow cell 110. A cleaning apparatus 178 may be arranged to run clean fluid through the apparatus 100 prior to the passage of the fluid under analysis to the flow cell 110. As will be appreciated any appropriate means may be utilised to supply fluid to the flow cell 110, such as an automatic pump, a manual pump, or a gravity feed. The components of the apparatus 100 are in fluid connection with one another via a series of tubes or pipes, and as such, create an overall flow direction from the fluid source 172 to the waste 180 via the flow cell 110 creating a fluid flow path. It is very important that the flow of the particulates is not blocked throughout the fluid flow path. Embodiments that have smaller lengths of pipework, such as those including a gravity feed system, have the additional advantage of reduced servicing time and costs, and reduced costs in terms of consumables (pipes/tubes).

This is further illustrated in an alternative embodiment of the present invention as illustrated in FIG. 1B. One advantage of using a fluid source 170 that is not located below the flow cell 110 and therefore removes the need to pump fluid upwards is that the issue of Stokes settling is reduced. This is where, for fluids pumped against gravity, at low fluid speeds larger particulates may move under the force of gravity, effectively in reverse when compared to the direction of fluid flow. In addition, the Stokes settling of larger particulates may be greater than the fluid velocity, which would prevent such particulates from moving in the direction of the fluid flow. Consequently, for flow fluid flow going with gravity, the Stokes settling works in the direction of fluid flow, and as such is minimised and far less likely to occur. The apparatus 100 comprises at least a flow cell 110 (also known as a microfluidic cell), a first light source 120, a second light source 130, and an image capture means 140. The apparatus 100 is connected to an analysis means 150 and a data storage means 160 for analysing and processing data from the apparatus 100. The first and second light sources 120, 130 are configured to illuminate a fluid within the flow cell, and image capture means 140 is configured to capture images of particulates within the fluid in the flow cell 110. In an embodiment the flow cell 110 is arranged to receive the fluid whilst the fluid is in motion. This enables dynamic images of a moving particulate to be seen. The components of the apparatus 100 are chosen firstly, to enable a sample to follow a predictable route through the apparatus 100, free from anything that may impede the fluid flow so as to block it completely and secondly, to optimise the optical quality of the images captured.

A fluid source 170 is fluidly connected to the flow cell 110, which is arranged to receive fluid for analysis. Preferably, this provides the fluid into the flow cell 110 using a constant displacement method. In the specific example of FIG. 1B the fluid source comprises a syringe l72a that contains a fluid for analysis. The syringe 172 a may be used to fill the flow cell 110 directly with fluid, either using a syringe driver or syringe pump. The syringe may comprise a plunger having a plate and plunger mechanism, where pushing the plunger causes the plate to push fluid out of the syringe, or only comprise a plate, such that an external plunger connects with the plate to force it through the syringe to expel fluid. A linear drive system is desirable to enable the constant displacement of the fluid into the flow cell 110. This, as stated above, helps to minimise issues of any particulates moving against the flow of the fluid during use of the apparatus 100.

Optionally a pressure sensor l72b may be used to monitor the injection of the fluid into the flow cell 110.

In both FIG. 1 A and FIG. 1B the first 120 and second 130 light sources are arranged to illuminate a first 112 and second 114 surface of the flow cell 110 respectively. The flow cell is preferably formed from an optically clear material (clear in the visible, UV and near IR spectra at least) that does not degrade in the presence of lubricants or other non-aqueous fluids. In the present embodiment quartz is used. The optical properties throughout the walls of the flow cell 110 are substantially constant. In the present embodiment a fused flow cell 110 is used, although it may also be possible to use fused flow cells in certain embodiments. The second surface 114 of the flow cell 110 may be positioned in registration or parallel with the first surface 112 of the flow cell 110, and as such, the second light source 130 may be positioned opposite the first light source 120. The flow cell 110 is shaped such that it comprises a long axis 116 parallel to the direction of fluid flow through the flow cell 110, and a short axis 118 perpendicular to the direction of fluid flow through the flow cell 110, where the length of the long axis 116 is greater than the length of the short axis 118. The flow cell 110 is generally rectangular in shape, such that the first 112 and second 114 surfaces are generally rectangular and contain the long axis 116, and separated by a short distance along the short axis 118. The long axis 116 of the flow cell 110 may also be defined as the axis along which fluid flows from the ingress of the flow cell 110 to the egress. Within the flow cell 110 a channel (preferably having a depth and width in the order of ±l0pm of the length of the short axis 118) is defined, running along the long axis 116, through which the fluid flows. The channel has dimensions where its depth and width are significantly shorter than its length. However by providing a long, narrow channel it is possible to ensure that particulates have a volume of fluid to travel along throughout the illumination and image capture area, thus enabling particular behaviours to be observed. Preferably the length of the long axis 116 of the flow cell is in the range of 10 to 50mm, and the short axis has a length in the range 50pm to l50pm, and is preferably around lOOpm as in the embodiment of the invention described here. The short axis is greatly exaggerated in Fig. 1 A and Fig. 1B to highlight the optical arrangement. As fluid flows through the flow cell 100 it is subject to a velocity gradient along the short axis 118, since the surface tension between the fluid and the flow cell 100 walls slows down the fluid flow. Consequently, fluid at the centre of the channel flows faster than adjacent the walls of the flow channel 100.

As will be explained later, the first light source 120 is arranged to illuminate the first surface 112 of the flow cell 110 and the second light source 130 is arranged to illuminate the second surface 114 of the flow cell 110. The first light source 120 may use a first bandwidth of optical frequencies for illumination. Preferably the first bandwidth of optical frequencies is a narrow bandwidth, centred on a particular frequency or group of frequencies. This may be done by using either a polychromatic light source, such as a white LED (light emitting diode) source employing an optical filter, such as a bandpass filter or an interference filter, to create a narrow bandwidth of optical frequencies, or a monochromatic or substantially monochromatic and coherent light source, such as a laser. The optical filter may also be combined with a neutral filter and/or a condenser lens to optimise the required optical properties. The second light source 130 may use a second bandwidth of optical frequencies for illumination. Preferably the second bandwidth of optical frequencies is a wide bandwidth, such as a polychromatic or white light source. Advantageously, illumination by the front 120 and second 130 light sources allows for the distinction of gas/vapour bubbles from fluid particulates from solid particulates, and even metallic particulates from non-metallic via their reflective properties. The first light source 120 provides backlit illumination of the flow cell 110, along the short axis 118 of the flow cell 110, enabling the outline of any particulates within the fluid to be seen. The second light source 130 provides front lit illumination of the flow cell 110, along the short axis 118 of the flow cell 110, enabling further refinement of the particulate images as detailed below. It may also be advantageous to include a calibration step in the method described below in order to ensure that images are obtained as accurately as possible. A white target containing markings of known size and separation distance may be employed to calibrate the image capture means 140, thus creating images in which any movement, whether rotational or translational, may be estimated accurately.

The second light source 130 may be positioned co-axially with and/or opposite the first light source 120. In a preferred embodiment the second light source 130 comprises a first illumination source 131 and a second illumination source 132, arranged about a common axis. The light sources 131, 132 preferably comprise white LED (light emitting diode) sources. A number of such sources may be combined with a field lens to form an extended source in order to increase the intensity of light available for the image capture means 140 to capture images for an increased frame rate for multiple image capture. In one embodiment, the first 131 and second 132 illumination sources are positioned on either side of the image capture means 140 in the plane of the flow cell 110. Images of the fluid in the flow cell 110 under illumination are captured via the image capture means 140. The second light source 130 may be arranged to illuminate the second surface simultaneously while the first light source 120 illuminates the first surface.

Alternatively, the second light source 130 may be arranged to alternate illumination with the first light source 120. The second light source 130 may comprise at least one source of illumination, such as the two light sources 131, 132. The sources of illumination 131, 132 may be arranged to illuminate the second surface at respective angles so as to provide uniform illumination of the particulate. For example, the sources of illumination 131, 132 may be positioned at identical angles with respect to the line of sight of the image capture means 140. Preferably the first 131 and second 132 illumination sources are positioned at a 45° angle to the short axis 118 of the flow cell 110. This glancing illumination creates a shadowing effect on the surfaces of the particulates being imaged, enabling further details of the shape and size of these particulates to be determined. The second light source 130, or the sources of illumination 131, 132 comprising thereof, may be arranged to illuminate the second surface at obtuse angles with respect to the long axis 116 of the flow cell 110. The second light source 130 may be arranged to illuminate the second surface co-axially along the optical axis of the image capture means 140. The sources of illumination 131 and 132 may be activated simultaneously, either

simultaneously with the activation of the first light source 120 or independent of the activation of the first light source 120. For example, the time for which the sources of illumination are activated may be overlap, exceed or be reduced compared with the time for which the first light source 120 is activated, and/or be continuous or discontinuous, for example, pulsed, compared with the time for which the first light source 120 is activated. Alternatively or additionally, the first 131 and second 132 sources of illumination may be activated independently of one another, such that the time for which the first source of illumination 131 is activated may overlap, exceed or be reduced compared with the time for which the second source of illumination 132 is activated, and/or be continuous or discontinuous, for example, pulsed, compared with the time for which the second source of illumination 132 is activated.

In one embodiment the image capture means comprises a microscope optically coupled to an objective lens and a camera arranged to image particulates in the fluid. The image capture means 140 is arranged to capture one or more images of the particulates within the fluid present in the flow cell 110 under illumination by the first and second light sources 120, 130. The microscope objective (or other appropriate optical arrangement) of the image capture means 140 is arranged to provide suitable magnification of the particulates for individual imaging and analysis. The objective lens may be either an achromatic lens (where two wavelengths of light, typically red and blue, are brought into focus on a single plane) or an apochromatic lens (where three wavelengths of light, typically red, green and blue, are brought into focus on a single plane). Preferably an apochromatic lens is used as this has a lower amount of chromatic aberration and enables the imaging of small (<l0pm) particulates with greater accuracy in terms of both focusing and colour resolution.

The flow path defined by the flow cell 110 may be arranged to have a depth in the line of sight of the image capture means 140 of no less than the depth-of-field available to the camera, in order to ensure suitable focus. Alternatively, the flow path defined by the flow cell 110 may be arranged to constrain particulate flow to within the depth-of-field available to the camera. The image capture means 140 is positioned such that the focal point of the image capture means 140 is positioned to be able to image the particulates within the fluid (such that the focal plane falls on one of the first surface 112 or second surface 114 of the flow cell 110, the fluid or at any point between the first surface 112 or the second surface 114 of the flow cell 110 as desired). Preferably, the focal point of the camera is arranged to be at the centre of the flow path of the flow cell 110. Preferably the image capture means 140 has a line of sight parallel to the short axis 118 of the flow cell 110, and positioned centrally between the first 131 and second 132 illumination sources. The image capture means 140 may further be coupled to the analysis means 150 for performing automatic analysis of the images captured, and a storage means 160 for storing image data. The camera may be a Complementary Metal-Oxide-Semiconductor (CMOS) camera or Charge-Coupled Device (CCD) camera, or any other suitable imaging device, having the capability of capturing both static and dynamic images. In embodiments of the present invention a CCD camera is preferred.

Preferably a sample bottle, such as that used for the fluid source 172, is employed to collect analysed fluid. Alternatively, any form of suitable container may be used, depending on the method chosen for disposing of the analysed fluid. This may be, for example, to waste, to recycling, to a new container or to the container used for the fluid before analysis, with or without a filtration system.

The apparatus 100 may further comprise a magnet 220 arranged to provide a magnetic field 210 to the flow cell 110, as shown in FIG. 2 and as described below. The magnetic field 210 may be parallel or perpendicular to the long axis 116 of the flow cell 110, and causes deflection of any particulate within the fluid having a magnetic

susceptibility. All other particulates remain unaffected by the presence of the magnetic field. The field direction is chosen to maximise the amount of deflection of the particulates affected across the direction of the fluid flow within the flow cell. The magnet 220 may comprise any suitable electromagnet or permanent magnet. In embodiments of the invention a neodymium magnet has been found to be particularly effective.

As mentioned, the flow cell 110 is arranged to receive a sample of fluid for analysis, for example lubricant previously used in machine operation. Such lubricant will contain various particulates and other machine wear debris that can provide an indication of the status of the machine and/or the state of the lubricant. The flow cell 110 is arranged to provide a directed flow path of the received sample fluid. For example, the flow cell 110 may define a flow path of the received fluid along the long axis 116 such that a light- permeable volume of the fluid may be observed for analysis. The flow cell 110 may direct the fluid such that the lumen is 100 microns (perpendicular to the direction of fluid flow), although other lumen may be envisaged. The apparatus 100 may be arranged such that the fluid is continuously flowing through the flow cell 110 and particulates present within the fluid may be observed directly in flow, i.e. dynamically. As mentioned, the flow cell 110 may be vertically oriented to prevent settling of the particulates within the flow cell due to gravity, or a slight positive pressure may be employed to prevent the particulates forming accumulations in various parts of the apparatus 100, since this would impede the flow of the fluid through the apparatus 100.

A magnetic field 210 may be applied to the flow cell 110 in order to influence the movement of ferrous particulates present within the fluid. FIG. 2 shows a side view of a section of another embodiment of the invention. The magnetic field 210 may be provided by the suitable electromagnet or permanent magnet 220. It will be appreciated that the magnet 220 may be substantially smaller or larger than the flow cell 110 depending on the desired magnetic effect and any restrictions on size of the apparatus 100. Here the magnetic field 210 is preferably applied such that it is parallel or perpendicular to the long axis 116 of the flow cell 110, or at an acute or obtuse angle to either the long 116 or the short 118 axis of the flow cell 110. Preferably, the magnetic field is applied such that it is parallel with the long axis 116 of the flow cell 110, that is, perpendicular to the line of sight of the image capture means 140. Alternatively to using a single magnet, the magnet 220 may be made up of at least two magnets positioned relative to each other to provide a desired magnetic effect. By applying a magnetic field 210 to the flow cell 110, the trajectory of moving magnetic particulates in the flow cell may be altered such that they are separated from other non-magnetic particulates, allowing for individual imaging and separate analysis of their properties. For example, diamagnetic, paramagnetic, and ferromagnetic particulates can be distinguished by their movement trajectories in response to the application of the magnetic field 210. Diamagnetic particulates will have an opposite trajectory to paramagnetic particulates, and the scale of the trajectory may also be used to distinguish between paramagnetic and ferromagnetic particulates, or to give an idea of the magnetic susceptibility of the particulates. The magnetic field 210 may further only magnetise a portion of the flow cell 110, in order to allow analysis of any changes in particulate motion between non-magnetised and magnetised sections that can yield information such as the proportion of magnetic vs non-magnetic particulates and the extent of the magnetic response of the magnetic particulates.

The magnet 220 may be positioned so as to produce a magnetic field perpendicular to the line of sight of the camera for maximum visible deflection of magnetic particulates in the field. Advantageously, this allows for improved analysis of the trajectory of the particulate.

As described above, the magnet 220 may also be moveable. For example, the magnet 220 may be rotatable around and with respect to the axis of the flow cell 110 in order to allow variation of the direction of the magnetic field 210. Where the magnet 220 comprises a permanent magnet, the distance between the magnet 220 and the flow cell 110 may also be variable to allow activation/deactivation and/or adjustment of the magnetic field strength. Alternatively, where the magnet 220 comprises an electromagnet, the field strength may be varied by adjusting the electric current supplied to the electromagnet. The magnetic field 210 may comprise one or more of a spatially uniform magnetic field, a temporally uniform magnetic field, a spatially varying magnetic field, a temporally varying magnetic field, or a magnetic field gradient.

As also described above, the first light source 120 is arranged to provide illumination for imaging by the image capture means 140. Illumination of the first surface of the flow cell 110 by the first light source 120 is arranged to produce shadows or silhouettes, e.g. a shadowgraph, of particulates present within the fluid for imaging by the image capture means 140. Information regarding the size and structure of the particulates may be derived from the recorded shadows or silhouettes during analysis. For example, the outline of the shadows of the particulates produced by the first light source 120 may provide particulate shape information. For example, the first bandwidth may comprise wavelengths within the visible, infrared, or ultraviolet bands, although other bandwidths may be envisaged such that dynamic image capture at multiple wavelengths may provide an improvement in data accuracy. Such images may be obtained simultaneously, so that the image obtained is a composite image. Illumination by the first light source 120 also provides a background field for images captured by the image capture means 140, thus allowing for digital segmentation of the captured images. This is achieved using a process of chroma-keying, where a frequency of light absent from the fluid or particulates under examination is used to create a backlit image. For example, by using a uniform bandwidth to illuminate the particulates via the first light source 120, chroma-keying may be applied to images captured under the first light source 120 in order to allow for distinguishing imaged particulates automatically during analysis. In this instance, frequencies corresponding to the first bandwidth by the first light source 120 that have not interacted with the particulate may be digitally removed from the resulting images taken by the image capture means 140, leaving only segments of the image corresponding to illumination that has interacted with the particulate, such as the remaining shadowgraphs.

The arrangements described hereinbefore enable certain characteristics of

particulates within the fluid in the flow cell 110 to be captured by the image capture means 140 under the second light source 120, such as particulate colour, two-dimensional shape, three-dimensional shape, size, and surface texture. For example, illumination provided by the second light source 130 causes light to be reflected off particulates present within the fluid, which is imaged by the image capture means 140. Specular and diffuse reflections may also be seen from any particulates present within the fluid of the flow cell 110, and geometric analysis of reflections given by the particulates under the second light source 130 may provide information regarding the structural composition of the particulates during analysis. For example, imaged specular highlights that change in size and/or shape during travel of the particulate through the flow cell 110 may indicate a change in surface form. Constant specular highlights may indicate a bubble in the fluid. . In this way, reflections given by the second light source 130 are analysed to determine one or more of particulate colour, two-dimensional shape, three-dimensional shape, size, structure, and surface texture.

Images of the flow cell 110 and fluid/particulates within may be taken singly, either as static images, or as a group of images taken at pre-determined time periods. Unlike traditional ferrographic techniques, the static images taken using embodiments of the present invention are static images of the particulates in situ in the lubricant. Preferably at least three images are taken as a particulate moves through the flow cell 110, creating dynamic image capture. Multiple images may be taken of an individual particulate to determine its size and shape more accurately, with three images being the minimum number required to give a reliable estimate of particulate size and shape. The greater the number of images taken, the more reliable any information regarding any particular particulate will be. Preferably the image capture means 140 is able to capture a series of images resulting in a video image of the particulate, and therefore a dynamic rendition of particulate information. The ability of embodiments of the invention to capture and analyse the dynamic behaviour of particulates is very advantageous. Analysis of dynamic behaviour may also include observation of relative movement of particulates as the fluid containing them flows through the flow cell 110. For example, particulates may be observed to rotate or tumble across the field of view, highlighting 3-D topography in a manner that cannot be repeated using traditional ferrographic techniques. Particulates travelling through the flow cell 110 may be of asymmetrical shape, and since they are subject to a velocity gradient of the fluid in the flow cell 110, and thus rotate during travel through the flow cell 110. As will be explained, capturing multiple images allows for imaging at various angles of the particulate as they rotate, thus allowing for a more accurate and thorough analysis each particulate. In addition, it may be desirable to use more than one image capture means 140 to ensure that a number of high-quality images are obtained. The captured image stream may be stored in a storage means electronically coupled to the image capture means 140 in any suitable digital image format.

Advantageously, once stored the captured images may be transferred and/or analysed at a later date.

The analysis means 150 may comprise one or more processing devices arranged to receive captured images from the storage means and perform suitable analysis of any imaged particulates to determine one or more of a physical or chemical property of the particulates. The analysis means 150 is also responsible for activating the first 120 and second 130 light sources, and for the rate of flow of the fluid through the apparatus 100. The processing means may be arranged to receive the captured one or images, and perform analysis of the imaged particulates using techniques such as segmentation, feature extraction, classification, or other forms of computational analysis. The processing means may be arranged to perform segmentation of the received images to distinguish imaged particulates, such as via the chroma-key methods described above or any other suitable digital segmentation algorithm. For example, the processing means may be arranged to perform object recognition and detection of a shape corresponding to an imaged particulate in each of the received images. The processing means may be arranged to identify and track the movement of an individual particulate through the flow cell 110 in the received images using any suitable object tracking method, such as using kernel -based tracking or contour tracking

algorithms. The processing means may be arranged to determine individual characteristics of each imaged particulate, such as size, two-dimensional shape, three-dimensional shape, colour, topography, magnetic susceptibility, surface texture, and/or structure. Other attributes that may be derived include one or more of contour, area, centroid, bounding box, aspect ratio, angle, convexity, brightness, and circularity of the particulate.

Attributes corresponding to the contour may comprise the location and position of the boundaries or edge of the imaged shape of the particulate.

Attributes corresponding to the area may comprise a measurement of the section of space inside the boundary of the imaged particulate.

Attributes corresponding to the centroid may comprise the location of the geometric centre of the imaged shape of the particulate.

Attributes corresponding to the bounding box may comprise the location, size, and position of the minimum enclosing box containing the imaged shape of the particulate.

Attributes corresponding to the aspect ratio may comprise a measurement of the ratio of the width to the height of the imaged particulate.

Attributes corresponding to the angle may comprise a measurement of the angle of the particulate with respect to particular axis.

Attributes corresponding to the convexity of the particulate may comprise a measurement of the curvature of the imaged particulate, preferably the primary axis of the particulate.

Attributes corresponding to the brightness may comprise a brightness value of the imaged particulate.

Attributes corresponding to the circularity may comprise a measurement of the roundness of the imaged particulate.

Qualitative information regarding asymmetry and amorphous shape of the image particulate based on its changing profile during rotation may also be obtained.

Information regarding particulate formation may further be derived from the analysed attributes using standard ferrography analysis methods, such as identifying rolling contact fatigue, micro cracking, etc. The behaviour of a particulate as it moves through flow cell and/or through the fluid may be analysed via measuring one or more of its movement path, velocity, or trajectory on a series of images taken successively as the particulate moves through the flow cell. The rotation or spin of a particulate during flow may be measured by detecting a change in shape during movement, for example. As noted above, geometric analysis of reflections given by the particulates may also provide information regarding the structural composition of the particulates during analysis.

The processing means may be arranged to classify each imaged particulate based on the measured characteristics, attributes, and geometric analysis using any suitable classification method. In this way, information relating to the composition and origin of the imaged particulates may be gleaned. Information relating to the wear and tear and status of the machine from which the imaged particulates originated from may therefore be derived. Analysis results may be output in any suitable output means, such as a computer screen, or a memory device, such as a hard drive or USB memory stick or other physical media.

By determining a physical or chemical property of particulates within the fluid, the condition of the machinery component from which the fluid was sampled from may be determined. In addition, it may be possible to determine information regarding the performance, behaviour and/or any degradation of the fluid, which is of particular use in the situation where the fluid is a lubricant. The analysis means 150 may determine a chemical composition of the particulate, or whether the particulate is organic or inorganic.

As an alternative to using various processing means, it may also be desirable to analyse the images captured by eye, for example, to obtain qualitative information about the particulate.

FIGS 3 A-3C illustrate a technique for analysing the behaviour of a particulate as it flows through a flow cell, according to the invention. The technique may be used by the apparatus 100 as described above. According to the invention there is provided example movement paths at sequential points in time of a ferrous particulate 310 and a non-ferrous particulate 320 as they travel through a flow cell 300. As described previously, FIGS 3A- 3C may also correspond to images captured of the particulates 310, 320 in the flow cell 300 in operation. In this instance, a magnetic field has been applied to the flow cell 300 in order to influence the movement of the ferrous particulate 310. The ferrous particulate 310 follows movement trajectory 315 as it travels through the flow cell 300. Similarly, the non-ferrous particulate 320 follows movement trajectory 325 as it travels through the flow cell 300. As will be appreciated the presence of the magnetic field will causes the movement trajectory 315 of the ferrous particulate 310 to deflect, causing the movement trajectories 315, 325 to separate. In this way, ferrous particulates may be distinguished from non-ferrous particulates according to their movement paths. Images of the movement paths of particulates flowing through the flow cell 300 may be captured by an image processing means and thus analysed in order to ascertain their magnetic properties. The presence of the magnetic field will also cause the ferrous particulate to deflect at an angle 330 respective to the non-ferrous particulate dependent on its magnetic properties. For example, under identical magnetic fields ferromagnetic particulates will exhibit a larger angle of deflection 330 compared to paramagnetic particulates. Similarly, diamagnetic particulates may exhibit an opposite movement trajectory and angle of deflection with respect to the direction of the applied magnetic field when compared to ferromagnetic and paramagnetic particulates. Thus, the movement trajectories of a particulate under analysis may be compared with the magnetic responses of known materials, and the derived scale and angle of the movement trajectories may be used to ascertain further information about the magnetic properties of the particulate. It will also be appreciated that a velocity of the particulate under a magnetic field may be analysed in order to derive information about its properties. Analysis of the scale, direction, and angle of movement trajectories may be performed computationally by a processing means. The trajectory or degree of deflection of a particulate by the magnetic field will also indicate the magnetic susceptibility of a particulate, for example, an indication of which iron oxide is present.

FIGS. 4A-4C illustrate a technique for analysing the behaviour of a particulate as it flows through a flow cell 400, according to embodiments of the invention. The technique may be used by the apparatus 100 as described above. FIGS 4A-4C show the rotation of a particulate 410 at sequential points in time as it travels through a flow cell 400, according to the invention. FIGS. 4A-4C may also correspond to images captured of the particulate 410 in the flow cell 400 in operation.

As described previously, the particulate 410 is transported via fluid flowing through the flow cell 400. The particulate 410 will follow movement trajectory 420 as it travels through the flow cell 400. However, fluid will flow at different speeds between the edges of the flow cell 400 and the centre, due to a fluid velocity gradient. As will be appreciated, the presence of the velocity gradient causes the particulate 410 to rotate as it travels through the flow cell 400.

As the particulate 410 rotates during its travel through the flow cell 400, the shape of the particulate 410 as imaged by the image capture means will change according to side of the particulate facing the image capture means. In this way, multiple angles of the particulate 410 may be captured by image capture means, advantageously allowing a more complete imaging of the particulate 410. In addition, the presence of specular reflections 430 that emerge during particulate rotation may indicate a change in surface form. A more complete surface topology may thus be derived from images taken by the image capture means of the particulate 410 as it rotates.

In addition, the rate of rotation or spin of the particulate 410 during flow may be measured by detecting a rate of change in shape during movement, for example, allowing information to be derived regarding physical attributes of the particulate. Analysis of the velocity and surface topology may be performed computationally by a processing means.

FIGS. 4D - 4F show images of a single particulate taken using a fluid analysis apparatus in accordance with an embodiment of the present invention as shown in FIG. IB. In the first frame, FIG. 4D, a single particulate P is highlighted. In the second frame, FIG. 4E, taken immediately following the first image shown in FIG. 4D, the particulate P has rotated whilst moving through the fluid and appears to be absent from the image completely. In the third and final frame, FIG. 4F, taken immediately after that in FIG. 4E, the particulate P is visible once more. However, in the first frame FIG. 4D, the size of the particulate P is measured as 8 pm and in the third frame FIG. 4F, the size of the particulate P is measured as 32pm. This highlights the benefits of a multiple frame imaging approach.

It will be appreciated that the invention may utilise any combination of the techniques described above.

According to embodiments of the invention, there is provided a method 500 for imaging particulates present within a fluid. The method 500 may be performed using the apparatus 100 in order to assess the wear behaviour of an apparatus by examination of a fluid required for the apparatus to function. The method 500 comprises a first step 510 of causing such a fluid to flow into, through and out of a flow cell 110, as described above. The fluid may be directed in the flow cell so as to provide a light-permeable sample of the fluid for analysis. A second method step 520 comprises illuminating the fluid within the flow cell 110 using a first light source 120. The illumination by the first light source 120 is arranged to produce a simple backlit image of the particulates present within the fluid for imaging. Illumination by the first light source 120 may be provided using a first bandwidth. The first bandwidth of optical frequencies is a narrow bandwidth, centred on a particular frequency or group of frequencies. This may be done by using either a polychromatic light source, such as a white LED (light emitting diode) source employing an optical filter, such as a bandpass filter or an interference filter, to create a narrow bandwidth of optical frequencies, or a monochromatic or substantially monochromatic and coherent light source, such as a laser. The first bandwidth may comprise frequencies within the visible, infrared, or ultraviolet bands, although other bandwidths may be envisaged. Preferably, the first bandwidth corresponds to a range of wavelengths associated with a colour that is absent from the fluid or particulates present within the fluid, such that chroma-keying, as described above, may be carried out.

The method 500 further comprises a third step 530 of illuminating the fluid using a second light source 130. The second light source 130 may be positioned opposite the first light source 120, and coaxial with the first light source 120. The second light source 130 may provide illumination using a second bandwidth of optical frequencies. The second bandwidth is a wide bandwidth, for example, from a polychromatic or white light source, such as a light emitting diode or tungsten filament. The second light source 130 may comprise one or more sources of illumination, such as a first 131 and a second 132 source of illumination, and may be arranged to provide uniform illumination of the fluid. The second light source may also be arranged to illuminate the second surface co-axially along the optical axis of the image capture means 140. As described above, when two sources of illumination 131, 132 are used these are set at an angle, for example, 45° to the optical axis of the image capture means 140. The step 530 of illuminating the fluid using a second light source may be performed simultaneously with the step 520 of illuminating the fluid using the first light source. Alternatively, the step 520 of illuminating the fluid using the first light source may be performed separately in time to the step 530 of illuminating the fluid the second light source, as described above.

In an embodiment of the invention, the method 500 may optionally further comprises a fourth step 540 of applying a magnetic field to the fluid to influence the movement path of any ferrous particulates present within the fluid. The magnetic field may comprise a spatially uniform magnetic field, a temporally uniform magnetic field, a spatially varying magnetic field, a temporally varying magnetic field, or a magnetic field gradient.

The method 500 further comprises a fifth step 550 of capturing one or more images of the fluid during illumination by the first and second light sources. The images may be taken apart at pre-determined periods. Multiple images may be taken of an individual particulate. The captured image stream may be stored in a storage means linked to the image capture means 140 and used subsequently or simultaneously for analysis. The apparatus 100 is connected to an analysis means 150 that may comprise one or more processing devices arranged to receive captured images from the storage means and perform suitable analysis of any imaged particulates to determine one or more of a physical or chemical property of the particulates. The analysis means 150 is also responsible for activating the first 120 and second 130 light sources, and for the rate of flow of the fluid through the apparatus 100. The processing means may be arranged to receive the captured one or images, and perform analysis of the imaged particulates using techniques such as segmentation, feature extraction, classification, or other forms of computational analysis. The processing means may be arranged to perform segmentation of the received images to distinguish imaged particulates, such as via the chroma-key methods described above or any other suitable digital segmentation algorithm. For example, the processing means may be arranged to perform object recognition and detection of a shape corresponding to an imaged particulate in each of the received images.

The method 500 further comprises a sixth step 560 of determining a physical or chemical property of a particulate present in the captured images. Physical properties may include individual characteristics of each imaged particulate, such as size, two-dimensional shape, three-dimensional shape, colour, topography, magnetic susceptibility, surface texture, and/or structure. Other attributes that may be derived include one or more of contour, area, centroid, bounding box, aspect ratio, angle, convexity, brightness, and circularity of the particulate. Chemical attributes may include a chemical composition of the particulate, or whether the particulate is organic or inorganic.

Determining a physical property of the particulate may comprise using techniques described previously such as segmentation, feature extraction, classification, or other forms of computational analysis and image analysis.

Additional and/or alternative steps may be included within the method 500 as desired. For example, a seventh step of performing segmentation of the received images to distinguish imaged particulates, such as via chroma-key methods described above may be included. An eighth step of identifying and tracking the movement of an individual particulate through the flow cell 110 in the received images using any suitable object tracking method, such as using kernel-based tracking or contour tracking algorithms may be included. A ninth step of measuring one or more of a movement path, velocity, and trajectory of a particulate, and/or a step of classifying each imaged particulate based on the measured characteristics, attributes, and geometric analysis using any suitable

classification method may be included.

Embodiments of the invention are particularly useful in analysing non-aqueous fluids, such as lubricants, or aqueous fluids, such as hydraulic fluids, and various other industrial fluids used in a variety or work situations. For example, in the case of both hydraulic fluids and lubricants the purpose of the fluid is to prevent wear occurring between at least two surfaces in contact. The ability of the fluid to do this may be assessed using the amount and types of particulates, for example, whether these are‘hard’ or‘soft’, as discussed above, examined using embodiments of the invention. Additionally or alternatively, the condition of each of the at least two surfaces in contact may also be assessed by studying the particulates within the fluid. Embodiments of the invention therefore find use in predictive maintenance, servicing, regular maintenance, tool and lubricant development and understanding the interaction at a basic chemical and physical level of a surface and a lubricant/hydraulic fluid.

Although the present invention has been described in connection with some embodiments, it is not intended to be limited to the specific form set forth herein. Rather, the scope of the present invention is limited only by the accompanying claims.

Additionally, although a feature may appear to be described in connection with particular embodiments, one skilled in the art would recognize that various features of the described embodiments may be combined in accordance with the invention. In the claims, the term ‘comprising’ does not exclude the presence of other elements or steps. Furthermore, the order of features in the claims does not imply any specific order in which the features must be performed and in particular, the order of individual steps in a method claim does not imply that the steps must be performed in this order. Rather, the steps may be performed in any suitable order. In addition, singular references do not exclude a plurality. Thus, references to‘a’,‘an’,‘first’,‘second’, etc. do not preclude a plurality. In the claims, the term‘comprising’ or“including” does not exclude the presence of other elements.