Image analysis – Applications
Reexamination Certificate
1999-02-26
2001-12-11
Johns, Andrew W. (Department: 2621)
Image analysis
Applications
Reexamination Certificate
active
06330343
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to an improved method of analyzing petroleum products in particle form. More particularly, the present invention relates to an improved method for analyzing the quality of petroleum by-products in particle form, especially coke, and relates to improved methods for refining petroleum to obtain desirable coke by-products.
2. Prior Art
As will be appreciated by those skilled in the art, petroleum products are analyzed for many purposes by using various tests. The test results are used for many purposes, sometimes even including controlling the refining of these products in order to achieve desired results but more often for appropriately classifying the products for subsequent sale.
For example, coke is a resulting by-product from petroleum refining that must be appropriately classified prior to sale. Cokes are typically classified by measuring their coefficient of thermal expansion (CTE) which provides the rate of expansion (or contraction) of a substance with temperature change.
In one use, coke is used in manufacture of large graphite electrodes for electric arc furnaces employed in the steel industry. It is known that cokes with lower CTE (e.g. FCTE=0.0-2.0×10
−7
/° C.) result in better performance of graphite electrodes under electric arc furnace (EAF) operations than those with higher CTE (e.g. FCTE>4.0×10
−7°
C.). Therefore, lower CTE cokes are more desirable than higher CTE cokes.
It is believed that, in typical representative coke samples, the lower CTE materials consist of a larger number (percent) of highly needle-like particles than the higher CTE materials. Needle-like particles are those whose structure has a preferential orientation. They are generally elongated as a result; hence the term “needle”. For a particular premium coke grade, the needle-like structure is not constant throughout the coke particles of that grade. Instead, there is a distribution of highly needle-like to highly non-needle-like particles.
One known method of classifying coke involves testing or examining these calcined cokes to determine the coke CTE. This procedure involves extrusion of a mixture of calcined coke particles and a binder pitch followed by baking, graphitizing and measurement of CTE of the resulting artifacts. Typically, the artifact is heated and measurements are taken at various temperatures to determine the CTE for a particular batch of coke. As will be appreciated by those skilled in the art, this procedure is fairly involved and time-consuming, on the magnitude of three to four days for a CTE measurement.
The resulting CTE determination has been previously used to segregate cokes into quality grades. However, the known methods for measuring coke CTE are complex and time-consuming. Thus, an improved method for classifying cokes is desirable.
It has been observed that, to even an untrained eye, higher quality cokes are shinier than lower quality ones. Subsequent research has shown (KOA Oil Company, 1997 Carbon Conference, PSU) that there is a correlation between CTE measurements and the “shininess” of a field of coke particles.
Another method of classifying coke is to use the varying reflectivities of the coke. The lustre method of the present invention quantifies this visual perception by digitally measuring the reflection of light from a pan of coke particles. This measurement is the coke's lustre (hereinafter, the phrase “coke lustre” will be used to refer to the intensity of visible light reflected from the surface of a coke particle or a layer of coke particles). The measurement of coke lustre is approximately analogous to extracting and counting the particles with the most needle-like character. Thus, coke lustre reveals information that can be used for ranking cokes based on their needle-like structure that is similar to CTE measuring methods.
Measuring coke lustre is easier than other known methods of physically measuring coke CTE. Classification of cokes based on lustre measurements has been difficult because only small differences in lustre are seen, even when CTE differences are significant. In other words, it has been difficult to correlate the small deviations noticed between coke lustres with a meaningful correlation with CTE variances.
Therefore, a need exists in the art for an improved method of classifying coke. A particularly desirable improvement would be an improved method of classifying coke based on its reflectivity or lustre as opposed to existing methods of physically measuring coke CTE. An even more desirable improvement would be a method for increasing the accuracy and reliability of such a lustre classification method.
SUMMARY OF THE INVENTION
The present invention addresses the above referenced needs in the art. In an exemplary embodiment, the invention provides a method for measuring coke quality by digital quantification of high intensity reflectivity. The invention includes an improved test for qualifying petroleum products, especially coke. The test involves obtaining a sample from a petroleum product (i.e. calcined coke) produced under known operating parameters during petroleum refining. The test enables the user to quickly and efficiently classify the product.
The test generally includes several broad steps. The first step is to obtain a representative sample for the target product, in the exemplary process coke, although the test may work well for other types of petroleum products as well.
The representative sample is then prepared in accordance with standard laboratory protocols (i.e. sieving, de-oiling, etc.) until the sample is properly suited for further testing. The properly prepared sample is then appropriately placed beneath an illuminating device in a manner that promotes the formation of a substantially smooth upper surface. This may require physical leveling of the sample or the like. The critical consideration is that the upper surface of samples being tested are prepared in an uniform manner to remove or at least curtail errant deviations from the test procedure.
The sample is then exposed to illumination. Preferably, a ring light or other lighting mechanism is deployed to shine on the sample with visible light from a desired direction (i.e. directly over the top surface of the sample). The illumination produces a visible light pattern from the light reflected from the sample's upper surface. As discussed previously, the reflectivity or lustre of a particular sample due to the illumination varies in correlation to the sample's CTE.
The pattern resulting from the illumination is then acquired to capture a digital image of the reflection of the sample. In a preferred embodiment, the digital image was acquired through use of a photography camera. In some circumstances, it may be desirable to slightly magnify the sample for acquisition, but this is not always necessary. Of course, the magnification can be accomplished with lenses or in other conventional manners.
Preferably, the pattern is acquired multiple times to develop an average image to reduce noise. In practice, the pattern was acquired 16 times consecutively although the multitude of acquisitions can be increased or diminished as desired.
Once a representative image has been obtained, it is processed digitally. The digital image is then quantified by measuring the gray levels present in the image. The highest gray levels are extracted and measured to produce a representative lustre measurement for the sample. During this extraction and measurement, a computer is preferably used to select the highest gray levels although specialized equipment having pattern recognition digital processing functions capable of producing and defining the video image as having a low or high CTE may be alternatively employed.
Ideally, the preceding process is repeated several times for each sample and the resultant lustre measurements for each iteration are totalled and averaged to avoid potential errors or deviations. In this manner, a more accurate re
Johnson Paige Lea
Jozavi Hooshang
Conoco Inc.
Head Johnson & Kachigian
Johns Andrew W.
Nakhjavan Shervin
LandOfFree
Method for measuring coke quality by digital quantification... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for measuring coke quality by digital quantification..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for measuring coke quality by digital quantification... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2567407