Optical Measuring Method

This invention relates to distance measuring methods and particularly to distance measuring methods employing electro-optical equipment.

In the manufacture of steel, furnaces are often lined with refractory material. This refractory material wears as a result of the process. After the refractory material has worn down to a predetermined level, it must be replaced, otherwise a dangerous situation may arise wherein the molten steel will leak from the furnace. Steel making equipment is quite expensive and the economic efficiency of such equipment is dependent upon the length of time a furnace can be employed without shutting down for providing a new refractory lining. Therefore, it is of substantial economic significance to employ a refractory lining as long as possible, yet it is also important not to allow the lining to be used when it has worn below a safe minimum thickness.

Presently, several methods are employed for monitoring the thickness of refractory linings in steel making equipment.

One of the methods that has been employed is stereophotography which utilizes changes in parallax between a series of photographs to determine the dimension and changes of dimension in a furnace, mine or other object. A careful and time consuming comparison of photographic images makes the stereo method a cumbersome and expensive process control. In addition, an accuracy of only between 30 to 40 millimeters is achieved.

Another prior art method comprises the embedding into the furnace lining precise deposits of radioisotopes which have predetermined concentration and composition characteristics. The radioisotopes characteristics vary as a function of the distance from an original surface, which during erosion recedes releasing a predetermined concentration of radioisotopes. Appropriate monitoring equipment detect the erosion of the surface. The disadvantages of the isotope method include limited applicability to a specific area or areas. Hopefully, these areas are representative of the erosion of the entire surface to be monitored. In order to alleviate the abovedescribed disadvantages, a great number of unique isotope characteristics must be employed on a great number of different portions of a surface which may provide a variety of indications which due to their complexity may even have a tendency to mask the true erosion of the entire surface. However, if the area is small only a single or limited measurement sample is obtained which usually is not representative of the entire surface. In addition, the measuring time is prohibitively long and expensive since elaborate isotope measuring and evaluation techniques are required. Therefore, the isotope method would only be useful as an aid in checking and calibrating other methods. Sonic methods usually are precluded because of thermal gradients and high refractions in the atmosphere.

Distance measuring equipment such as electro-optical type would be ideally suited for measuring thickness in a furnace since one can do that from a distance. However, furnaces are moved during their usage and it would be impossible to maintain a constant relationship between the electro-optical distance measuring equipment and the furnace and, further, since the interior of the furnace is at an extremely high temperature level, it would be difficult, if not impossible, to site the electro-optical distance measuring equipment on a particular point for separate readings.

In order to overcome the problems of the prior art, the present invention provides a method of making a series of measurements of the intersection of a line transverse to a surface, the surface being subject to wear and the surface as it wears in which the object may move with respect to a measuring instrument between measurements which includes the steps of measuring the location of first, second and third non-collinear reference points on the object with reference to the measuring instrument with the object in a first location; measuring the location of a first intersection of the line and the surface with respect to the measuring instrument with the object in the first location, measuring the location of the first, second and third non-collinear reference points on the object with respect to said measuring instrument with said object in a second location, locating an intersection of the line and the surface with respect to the measuring instrument with the object in the second location, and measuring the distance thereof from the measuring instrument.

As one embodiment of the invention, the intersection in the second location is determined by determining the relationship of the measuring instrument to the three non-collinear reference points adding the information relating the original intersection to the three non-collinear reference points and adding a factor based upon estimated wear.

In accordance with the further aspect of this embodiment of the invention, the new intersection is determined by an iterative process of estimating where the intersection should occur and measuring to see if a surface exists at that point.

FIG. 1 is a perspective drawing of a distance and angle measuring instrument making measurements on a furnace in accordance with the teachings of the present invention.

FIG. 2 is a graphical showing of a pair of coordinate systems each of which has the electro-optical measuring system as its origin in accordance with the teachings of the present invention.

Referring now to FIG. 1, we see an electro-optical distance measuring instrument 10 such as an AGA GEODIMETER, Model 700 made by AGA Aktiebolag, Lidingo, Sweden, modified in accordance with the teachings of co-pending U.S. application filed simultaneously herewith entitled "APPARATUS FOR MEASURING THE DISTANCE TO A POINT ON THE INNER WALL OF A HOT FURNACE" invented by Ragnar Scholdstrom et al, which measures distances and horizontal and vertical angles from the instrument. The instrument provides signals on the leads 20, 21 and 22 indicative of the various angles and distances measured. These signals are applied to a computing unit 23 which has a display unit 24 associated therewith.

The instrument 10 is mounted in front of a furnace 11 which is mounted for pivotal movement by a support structure not shown.

In accordance with this invention, three non-collinear points a, b and c are physically marked on a front surface 26 of the furnace, preferably on a circle perpendicular to an axis of the furnace, with the relationship of the points to the furnace not significantly changing with time.

In accordance with this invention, the thickness of a refractory lining 27 is monitored by making an initial measurement of the position of that lining with regard to a coordinate system associated with the furnace and a comparison with its drawing and then making further measurements of the position of the lining 27 relative thereto and comparing the same with the initial measurement. Since the furnace 11 may move relative to the instrument 10 between successive measurements, it is necessary to be able to relate the coordinate system of the furnace 11 with a coordinate system relative to the measuring instrument 10.

In accordance with this invention, the three non-collinear points a, b and c are employed in a coordinate conversion technique to relate the coordinates of the furnace 11 no matter what its orientation to the coordinates of the measuring instrument 10.

In one embodiment of this invention, the computing unit 23 is a small general purpose digital computation unit programmed to store signals from distance measuring instrument 10 while the furnace 11 is in a first position. An operator of the distance measuring instrument 10 measures the points a, b and c and the distance and angular relationship thereof is stored in the memory of the computing unit 23. A point P on the surface of the lining 27 is then measured and the distance and angular relationship thereof to the distance measuring instrument 10 is also stored in the computing unit 23. The point P is the intersection of the surface 27 with a line 30 transverse thereto.

After the furnace is used for steel processing, the lining 27 is worn and accordingly the location of the surface is changed. Additionally, the furnace is now in a different position relative to the measuring instrument.

In order to measure the new location of the surface of the lining 27, an operator of the measuring instrument 10 measures the location of the three collinear points a, b c with respect to the measuring instrument 10. Signals indicative of these measurements are provided by the measuring instrument 10 to the computing unit 23. The computing unit 23 by means of the program entered therein employs data from the furnace drawing or results from the first measurement to establish origin coordinates for the instrument 10 relative to the furnace 11. The computing unit 23 then displays aiming angles to be used to relocate the wanted position on the surface of the lining 27 in terms of the coordinate system of the furnace 11. The computing unit 23 next employs the second measurements of the three non-collinear points a, b and c to establish the coordinate system of the furnace 11 in terms of its new orientation to the measuring instrument 10. With this information, the computing unit next determines the coordinates of the point P in terms of the present location of the furnace 11 in relationship to the measuring instrument 10. The computing unit 23 also adds an estimated distance along line 30 from the location of the point P based upon the expected wear of the lining 27 to locate where the surface to be measured should be presently located if the estimated wear is correct. The operator of the measuring instrument 10 next points the measuring instrument 10 in the direction indicated by these calculations and measures to see if a surface is measured at the distance determined. If a surface, in fact, exists at the predetermined distance, then the estimated wear is, in fact, the actual wear. If the distance measured is not the predetermined distance but is greater or smaller than the predetermined distance then the estimated wear is changed accordingly with a new estimate, the computer 23 recalculates the angle at which measurement should be made and the distance to be measured and a new reading is taken with unit 10. If this action is repeated, iteratively, until the estimated wear chosen produces the predetermined distance an accurate reading is thus achieved of the wear of the lining.

The program employed in the calculating unit 23 utilizes coordinate conversion techniques based upon vector notation. The unit employs the three reference points a, and c measured by the instrument 10 in polar coordinates with the instrument as its origin to a cartesian coordinate system with the instrument 10 still as its origin. The coordinate system of the blast furnace 11 is derived by manipulation of vector values relating to the relative positions of the instrument 10 and the reference points a, b and c.

To begin the transformation, vector quantities are generated in the cartesian coordinate system with the instrument as its origin for each of the points a, b and c. Thus we have for the origin vector A(A.sub.x, A.sub.y, A.sub.z).

and equivalent expressions for B (B.sub.x, B.sub.y, B.sub.z) and C(C.sub.x, C.sub.y, C.sub.z). (See FIG. 2)

From these origin vectors two vectors in the plane of ABC are derived.

The vector N of the central axis, which is perpendicular to the plane ABC, can now be calculated by vectorial multiplication of U and V.

or in detail

For later simplification it is practical to use N in a normalized form with unit length. ##EQU1##

By establishing the coordinates of one point on the central axis an expression of this axis in parametric notation can be made. If two of the points AB, BC or AC can be chosen to constitute a symmetrical diameter, then their midpoint E is on the axis. If two of the A and C are on the diameter of a circle

or in detail

In practical application, however, it may not be possible to use two points constituting a diameter and in these usual cases it is necessary to use more complicated algebra. In either event point E on the axis is, since ABC is a circle, defined as the crossing of the midpoint perpendiculars S and T to the vectors U and V, which is contained in the common plane of U and V. Since S is perpendicular to both U and N we can find the direction of S by vectorial multiplication.

and equivalent for

the detailed expressions are found according to the rules given at (6), (7), (8) and (9).

The midpoint M.sub.1 of U and M.sub.2 of V are found according to rules given at (11), (12), (13) and (14). e1 ? ##STR1## The lines from M.sub.1 to E and from M.sub.2 to E can be expressed in parametric notation with p as the common parameter

and

at the crossing of these lines we have the center point E and here x of (19) is equal to x of (22) and equivalent for (20), (23) and (21), (24). From this we get for point E with parameters p.sub.1E and p.sub.2E at that point E.

By eliminating p.sub.2E we get ##EQU2##

By putting the value of p.sub.1E into the equations (19), (20), (21) we get the origin vector of E(E.sub.x, E.sub.y, E.sub.z). We can now write the expression of the central axis in parametric notation as

where t constitutes a depth from the plane ABC.

We can also establish a reference for angular positions by taking, for example, the vector A- E as a reference vector, which also preferably is taken in normalized format. ##EQU3## similar to (10).

We have now established the mathematical reference system for cylindrical coordinates of a point anywhere, its depth h by its projection on the line (28), (29), (30), its radial distance r by the perpendicular distance to the same line, and its angular position .alpha. by comparing its perpendicular radius to the line with the reference direction of a.

For the sake of completeness this is described in more detail.

After having aimed the beam at a point on the surface to be evaluated we get from the instrument the polar origin coordinates of the point P(P.sub.x, P.sub.y, P.sub.z). These coordinates are transformed to cartesian coordinates according to (1), (2) and (3). From this we calculate the difference vector from the point E to P.

by scalar multiplication of the F vector with the normalized n vector we get the depth h of P.

By definition h is identical to t in (28), (29), (30) for the point G where the plane perpendicular to n and containing P intersects the line through E with the direction n. We have therefore

The radius from the central axis to the point is calculated by

To resolve the angular position of P we make further multiplications. We define an additional directional vector in the plane ABC by vectorial multiplications as

We now normalize the vector P - G (which follows the line 30 of FIG. 1) to ##EQU4## With two scalar multiplications we get

From these values for cos .alpha. and sin .alpha. we can resolve .alpha..

If the cylindrical coordinates h, r and .alpha. are unpractical for the purpose, it is easy to transform into cartesian coordinates along n, a and b.

When it is desired to point the beam to the prescribed area on the new surface the equations, e.g. (41), (40), (37), (36), (35), (34), (30), (29), (28), (3), (2), (1) are run in a reversed fashion with the added estimate of wear to find an azimuth angle .beta. and a zenith angle .gamma. for the requested point.

This appendix is a program listing in direct machine language for running on a Hewlett-Packard calculator No. 9810 which may serve as the computing unit 23.

As is seen from the program, the steps thereof are numbered. In step 0, TBL 3 means that the system employs a 400.degree. measuring unit rather than the usual 360.degree.. This is because geodetic instruments normally use the 400.degree..

The following comments aid in an understanding of the above computer program:

An arrow .fwdarw. means "put into a memory cell" the number of which is following.

The following is a memory map for the above program which is helpful in debugging the program and also in transposing the program for use on other computing devices:

It should be understood that while the invention has been described with respect to a particular embodiment thereof, numerous others will become obvious to those with ordinary skill in the art in light thereof.