WO1996013699A2

WO1996013699A2 - Pipeline inspection pig and method for using same

Info

Publication number: WO1996013699A2
Application number: PCT/US1995/013878
Authority: WO
Inventors: Ken Baldwin Nichols; Lowell Robert Milligan; Phillip Eldon Miller; Lennie Charles Reeh
Original assignee: Id Measurements Inc
Current assignee: Id Measurements Inc
Priority date: 1994-10-27
Filing date: 1995-10-26
Publication date: 1996-05-09
Anticipated expiration: 1997-04-27
Also published as: WO1996013699A3; AU4362796A

Abstract

An instrumented inspection pig is disclosed. The pig comprises a central mandrel; two supporting members affixed to the central mandrel; two pairs of diametrically opposed radial sensors outputting a signal proportional to the distance between the inner wall of the pipe and the pig centerline through the radial sensor; an orientational sensor measuring the rotational orientation of the pig relative to the gravity vector and generating data indicating the same; a processor receiving the signals from the radial sensors and the orientational sensor and storing the data therein indicating the distance between the inner wall and the pig's axial centerline; and a power supply providing power to the rotational sensors and the processor. The raw diameters measured by the respective pairs of radial sensors at any given point for which data is available are calculated from the distances measured by the radial sensors. Each raw diameter is corrected for skew and is then orthogonally corrected using the distance measurements from the radial sensors. Approximate bend radius for the specific point is also derivable from the measured distances. The corrected diameters are then scrutinized to determine whether a mash or ovality condition is present and, where desirable, the bend radius of the pipe.

Description

/13699 PCΪ7US95/13878

PIPELIKE INSPECTION PIG AND METHOD FOR USING SAME

BACKGROUND OF THE INVENTION

Field of the Invention The invention pertains to pipeline inspection pigs and methods for using the same; more particularly, to pigs and methods for detecting deformations and bends in pipelines.

Description of the Prior Art The pipeline industry has long developed and used tools known as "pipeline pigs" for many tasks associated with inspection and maintenance of pipelines. A pig is typically inserted into a pipeline at a predetermined point and passed through the pipeline to a second predetermined point where it is removed from the pipeline. Sophisticated pigs carry electronic instrumentation to obtain information regarding some characteristic of the pipeline such as the internal dimensions of the pipeline. Information from internal dimensional inspection typically includes internal diameters and their position relative to some point in the pipe. Internal dimensional information can also indicate the direction and radius of bends in the pipe and the location of various internal features such as welds, valves, couplings, and tees.

Calipering pigs measure a pipeline's internal dimensions only in diameters, which is significant because conditions are frequently encountered where raw diameters are not accurate measurements of actual diameters. We have discovered, however, that measuring radii as opposed to diameters is advantageous as the radii can not only be used to derive raw diameters, but also to correct the raw diameters. Further, measurement of radii rather than diameters provides greater freedom in pig configuration in some instances.

Summary of the Invention An instrumented inspection pig is disclosed. The pig comprises a central mandrel; two supporting members affixed to the central mandrel; two pairs of diametrically opposed radial sensors outputting a signal proportional to the distance between the inner wall of the pipe and the pig centerline through the radial sensor; an orientational sensor measuring the rotational orientation of the pig relative to the gravity vector and generating data indicating the same; a processor receiving the signals from the radial sensors and the orientational sensor and storing the data therein indicating the distance between the inner wall and the pig's axial centerline; and a power supply providing power to the rotational sensors and the processor.

The distances measured by the respective pairs of orthogonal radial sensors at any given point for which sampled data is available are used to calculate the raw diameters. Each raw diameter is corrected for skew and is then orthogonally corrected using the distance measurements from the radial sensors. Approximate bend radius for the specific point is also derivable from the measured distances. The corrected diameters are then analyzed to determine whether a mash or ovality condition is present. Brief Description of the Drawings So that the manner in which the above-recited features, advantages and objects of the invention, as well as others which will become apparent, are attained and can be understood in detail, a more particular description of the invention briefly summarized above may be had by reference to the exemplary preferred embodiments illustrated in the drawings, which form a part of this specification. However, the appended drawings illustrate only typical preferred embodiments of the invention and are not to be considered limiting of its scope as the invention may admit to other equally effective embodiments. In the drawings:

Figure 1 is side view of the preferred embodiment of the invention with a cutaway illustrating electronic components housed in the body thereof and also illustrating several of the quantities measured or calculated by the invention.

Figure 2 is a block diagram of the electronics and instrumentation subsystem of the embodiment in Figure 1.

Figure 3A illustrates an "ovality" condition and Figure 3B illustrates a "dent" or "mash" condition, both of which are more fully described below;

Figures 4-5 illustrate various pig geometries when the pig is in the bend of a pipe;

Figures 6A-C illustrate a piece of pipe measured by the invention and demonstrate graphical display of the data gathered therefrom;

Figures 7A-B are a flow charts illustrating the formulation of correction factor tables;

Figure 8 illustrates an alternative embodiment of the invention; and

Figure 9, like Figure 6B, demonstrates one method of displaying the data gathered or derived by the invention.

DESCRIPTION OF THE PREFERRED EMBODIMENT The invention comprises inspection pig 10 shown in Figure 1 and the analysis system discussed below. Inspection pig 10, as illustrated in Figure 1 in a preferred embodiment, comprises two orthogonal pairs of diametrically opposed measuring arms 14a-d circumferentially, pivotably mounted on central mandrel 16. Central mandrel 16 is supported and substantially centered in pipe section 12 of a pipline by support members 22a-b, which in the embodiment of Figure 1 are pipeline scraper cups made of an elastomeric material whose construction and use are well known to those in the art. However, support members 22a-b may alternatively be spring-loaded members terminating in wheels so that pig 10 may be pulled through pipe section 12 in some applications. Arms 14a-d are spring-loaded and are urged outwardly to mechanically engage inner wall 18 of pipe section 12 through contacts 20a-d although arm 14d is not shown for the sake of clarity. Contacts 20a-d may also be rolling contacts such as are known in the art or prolate spheroid, sliding wear-resistant surfaces such as those in Figure 8 discussed below. Arms 14a-d in the embodiment of Figure 1 are mounted on central mandrel 16 at pivots 38a-d so that both arms I4a-d and pivots 38a-d are intermediate support members 22a-b. However, arms 14a-d and pivots 38a-d nay be mounted at any point along the length of central πandrel 16 for detecting ovality or mash conditions. If bend radius measurements are desired, however, arms I4a-d must be mounted in the embodiment of Figure l to central mandrel 16 intermediate support members 22a-b but may be mounted anywhere intermediate support members 22a-b.

Odometer 24 includes spring-loaded arms 26a-b and odometer wheels 28a-b and spring-loaded arms 26a-b are pivotably attached at pivot 30 to the terminus of central mandrel 16 by a universal-joint. The construction and operation of odometer 24 are well known and understood by those in the art. To summarize, arms 26a-b are sprung outward to maintain the rolling contact between wheels 28a- b and inner wall 18 of pipe section 12. Pivot 30 allows measuring wheels 28a-b to track inner wall 18 without affecting the alignment of pig 10 within pipe section 12 relative to pipe centerline 32.

At least one magnet 34 is mounted in one or both of wheels 28a-b, whose circumferences are known. Magnet 34 rotates with wheels 28a-b and triggers a counter (not shown) as wheels 28a-b rotate. Since the circumference of wheels 28a-b is known, each count is proportional to the distance traveled by pig 10 between counts and the total count is proportional to the total distance pig 10 travels through pipe section 12.

Pig 10 carries two principal types of sensors aside from odometer 24. First, pig 10 carries radial sensors 36a-d (shown in Figure 2) . Radial sensors 36a-d in the preferred embodiment include arms 14a-d in Figure 1 and rotational sensors mounted at pivots 38a-d, respectively, where arms 14a-d, respectively, are mounted to central mandrel 16. The invention requires at least one set of two orthogonal pairs of diametrically opposed radial sensors but additional sets as desired may be added to obtain the desired circumferential coverage. Second, pig 10 carries orientational sensor 40, a pendulum sensor whose output signal is proportional to the circumferential rotation, or orientation, of pig 10 with respect to vertical during transit through pipe section 12.

Each of the rotational sensors in the preferred embodiment of radial sensors 36a-d is a 7/8" diameter, single turn potentiometer having a conductive plastic element and a bushing mount manufactured by Bourns, Inc. and sold under the Model No. 6638. This potentiometer is commercially and readily available to those in the art as are the operational specifications and their use is well understood by those in the art. This same potentiometer is used to implement orientational sensor 40 as a pendulum sensor by hanging a concentric weight from the shaft, preferably with some damping mechanism. These rotational sensors must insulated from the fluid in pipe section 12 but may be exposed to the pipeline pressure provided they are pressure compensated in a manner well known to the art. However, there may be other, equally suitable embodiments for radial sensors 36a-d and orientational sensor 40. Radial sensors 36a-d may also be linear sensors, optical sensors, magnetic sensors or any sensor capable of measuring the distance to inner wall 18 of pipe 12 and generating an output proportional thereto. Radial sensors 36a-d therefore need not be limited to the measuring arm/rotational sensor combination of the preferred embodiment disclosed herein. Orientational sensor 40, may alternatively be a two-axis accelerometer or a plurality of equi-spaced mercury switches. These and other alternative embodiments will be apparent to those in the art having the benefits of the teachings herein.

Contact surfaces 20a-d at the distal terminus of arms 14a-d trace the contour of inner wall 18 while pig 10 is in transit during a pass as described below by virtue of being spring loaded and pivotably mounted to central mandrel 16. This is true regardless of whether the deflection is outward, as when ovality (shown in Figure 3A) is encountered, or inward, as when a dent or mash (show in Figure 3B) is encountered, relative to central mandrel 16. The rotational sensors comprising part of radial sensors 36a-d measure angular deflections ©_a-θ_d of arms I4a-d. Since the geometry of pig 10 is known, the distances r_a-r_d between pig centerline 32 and inner wall 18 corresponding to angular deflections e_a-© can be calculated and the rotational sensors are calibrated to output signals proportional to r_a-r_d, respectively, rather than θ_a-θ_d. Thus, in response to the angular deflections, each of radial sensors 36a-d output an electrical signal proportional to the distance between inner wall 18 of pipe 12 and the centerline 32 of pig 10.

Central mandrel 16 includes fluid-tight chamber 42 as shown in Figure 1 such as is well known in the art. Orientational sensor 40 as well as electronics package 44 shown in Figure 2 are housed within chamber 42. Referring now to Figure 2, the preferred embodiment of electronics package 44 includes multiple, high power NiCad cell batteries 48 to power the electronics including processor 50, either a Hitachi 64180 or Zilog Z8018 microprocessor, and memory 52 including both RAM and ROM for processing and storing the data from radial sensors 36a-d, odometer 24, and orientational sensor 40. Electronics package 44 receives the data from radial sensors 36a-d, odometer 24, and orientational sensor 40 via electrical leads 46a-f in a manner apparent to those in the art having the benefits of the teachings herein. Thus, processor 50 and memory 52 comprise a means for receiving and storing data from the various sensors, and in alternative embodiments may be a magnetic tape or some other electro-mechanical recorder. As those of ordinary skill in the art having the benefit of the teachings herein will appreciate, the particular structure of radial sensors 36a-d, orientational sensor 40, and odometer 24 in the implementation of the method of the invention is subsidiary to the sensed attributes. Radial sensors 36a-d, regardless of whether they are comprised of spring-loaded arms and rotational sensors, must sense the distance between inner wall 18 and centerline 32 of pig 10 and output an electrical signal proportional thereto. Orientational sensor 40 likewise may be the pendulum sensor of the preferred embodiment but need not necessarily be so provided that it outputs a signal proportional to the orientation of pig 10 relative to the vertical or gravity. Odometer 24 likewise may be any suitable structure measuring and outputting a signal proportional to the distance traveled by pig 10.

Thus, each of radial sensors 36a-d, orientational sensor 40, and odometer 24 generates an electrical signal containing data proportional to some particular, measured quantity as discussed above. These electrical signals are transmitted to electronics package 44 via leads 46a-f where they are sequentially sampled by processor 52 at a predetermined sample rate. The sampled data is monitored by processor 52 to detect any change greater than a predetermined amount in the output of radial sensors 36a-d. In the event there is such a change, a datapoint comprising data from the respective radial sensor, orientational sensor 40, and odometer 24 is stored to memory 52.

The predetermined sampling rate and change amount will vary from application to application in accord with constraints well known to those in the art familiar with other data acquisition pigs. For instance, sampling rate will be inherently limited by microprocessor speed if a microprocessor is used. Resolution will also be limited by weight, volume, and power constraints on electronics package 44 and the condition to be detected. As an example, for a pipe known to have accretions or deposits on the inner wall, a user searching for mashes will probably want resolution to be sufficiently low to avoid recording incidences of accretions and deposits. Consequently, the value of these parameters will depend on the particular implementation of pig 10 and its application.

Pig 10 as used in the preferred embodiment is inserted into pipe section 12 and is then propelled through pipe section 12 by the fluid, whether liquid or gas, moving within pipe section 12 operating against support members 22a-b. At the end of the pass, pig 10 is removed from pipe section 12. The mechanics and means of insertion into, propulsion through, and removal from pipe section 12 are well known to those in the art and any of various techniques may be employed. The instrumentation aboard pig 10 collects and stores data during the pass in the manner described above and the data generated by radial sensors 36a-d is then processed with the data generated by orientational sensor 40 and odometer 24 in a previously unknown manner to obtain results superior to those currently found in the art. Once the pass is completed, the stored data is retrieved from electronics package 44. In the preferred embodiment, this is performed by interfacing pig 10 with a computer and uploading the data stored in memory 52 in electronics package 44 to the computer's memory. The data set is a plurality of data points, each data point including a sampled output of each of radial sensors 36a-d, orientational sensor 40, and odometer 24.

At each stored data point each of the radii from the pipe centerline to inner wall 18, the pig orientation in the pipe 12, and the distance from the start of pipe 12 is calculated. The distance traveled by pig 10 through pipe section 12 is calculated for each datapoint in a manner known to the art from the data generated by odometer 24. So, too, the orientation of pig 10 within pipe 12 relative to the vertical, or gravity, is calculated as known to those in the art from data generated by orientational sensor 40. From the distance traveled by pig 10 and the orientation of pig 10 within pipe 12 relative to the vertical, the location and orientation of the condition represented by the datapoint can be determined.

The nature of the condition represented by the datapoint is identified from the data generated by radial sensors 36a-d. Referring now to both Figure 1, the geometry of pig 10 is known and, once calibrated, the outputs of radial sensors 36a and 36c are proportional to the change in the radii r_a and r_c between inner wall 18 of pipe 12 and the centerline of pig 10, radii r_a and r_c further comprising d_ac. Radii r_a and r_c are not "radii" in the technical sense of the word but shall be so called herein as a convenient label for the radial distance between the centerline of pig 10 and inner wall 18 of pipe 12. Since support members 22a-b center central mandrel 16 in pipe 12, the centerline of pig 10 is co-linear with centerline 32 of pipe 12 and the two shall be treated herein synonymously unless otherwise noted. One such exception is in the presence of a bend, as is discussed below.

The measured radii r_a and r_c are derived from the data generated by radial sensors 36a-d and the known geometry of pig 10. Furthermore, the process described immediately above is readily extrapolated to radial sensors 36b and 36d and still other radial sensors that might be added in alternative embodiments. Still further, when the data from one radial sensor is coupled with the data from a diametrically opposed counterpart, the raw diameters are calculable as in the manner of d_ac. The constituent radii can also be used to calculate correction factors for obtaining corrected diameter measurements as explained in detail below.

Figures 3A-3B illustrate a "dent" or "mash" condition and an "ovality" condition, respectively. As illustrated, these conditions are readily detectable once diameters D_ac and D_bd are calculated, a condition existing when either D_ac>D_bd or when D_bd>D_ac. When pig 10 is in a bend of pipe 12 as illustrated in Figures 4-5, the contact surfaces 20a and 20c of measuring arms 14a and 14c will not be positioned on a true pipe diameter (D_TRϋE) , but rather the diameter (d^^) of a non-circular conic section of the pipe 12 skewed relative to true pipe diameter D_TRUE. Furthermore, as shown in Figure 5, the pair of arms perpendicular to r_a and r_c (i.e.. those measuring r_b and r_d) , which are not illustrated in Figure 4 for the sake of clarity, will not measure a true pipe diameter, but a chord (C) less than the true diameter D_TRUE.

Referring still to Figure 5, the relationship between a chord (r_b+r_d) measured by a pair of diametrically opposed arms displaced orthogonally from centerline 32 of pipe 12 and the true diameter of pipe 12 is :

^DTR_UE ⁼ K x d_RAW

where D_TRUE is the true diameter of pipe 12, K is a correction factor, and d^^ is the chord length C. This relationship can be extrapolated to derive the value of ^DTRUE ^{from d} _RAW ^usiⁿ9 ^tne constituent radii of d_RAW as discussed below.

Utilizing the known geometry of pig 10, knowing how pig 10 will position itself within pipe 12, and knowing how the range of pipe internal diameters that pig 10 will measure, diameter correction factors can be calculated as functions of constituent radii. Also derivable from the set of measured radii is the bend radius of pipe 12 since the amount of offset of the centerline of the pig 10 from centerline 32 of pipe 12 at the center of central mandrel 16 in a bend is a function of r_a/r_c, the pipe bend radius is therefore a function of the pig 10 centerline offset.

Each calculated diameter is therefore corrected for "skew", i.e.. error in the direction co-linear with the calculated diameter, and "orthogonality", i.e.. error in direction orthogonal to the calculated diameter. The skew correction factor (K_s) of any pair of arms is a function of the ratio of that pair of arms. The bend ratio may also be calculated for each ratio of constituent radii. The orthogonal correction factor (K₀) relates the ratio of any pair of opposing arms to the correction factor for the orthogonal pair of arms for correcting the measured chord of a pipe diameter to closely approximate a true diameter

^DTRϋE* The set of measured radii r_a-r_d can therefore be used to detect deformations in pipe 12 such as ovality and dents and bends. This information, coupled with the data generated by odometer 24 and pendulum sensor 40 allows not only detection and calculation of these parameters, but also enables mapping of interesting pipe conditions.

The skew and orthogonal correction factors are functions of the measured radii r_a-r_d. A corrected diameter nearly approximating the true diameter is a product of a raw diameter and its respective skew and orthogonal correction factors. Stated mathematically in terms illustrated in Figures 4-5, the corrections for raw diameters d_ac and d_bd (i.e. , r_a+r_c and r_b+r_d, respectively) yielding corrected diameters D_ac and D_bd are:

D_ac = [r_a+r_c] x K_a__ac x K₀__bd

^Dbd = C^rb^+rd^{] x κ}s-bd * ^κo-ac

wherein the correction factor functions are obtained from tables constructed and indexed as described below.

In the preferred embodiment, processing is initiated by constructing a skew correction factor table and an orthogonal correction factor table for prospective radii ratios. The tables may be constructed either before or after pig 10 is passed through pipe 12 as the calculations are independent of actual measurements taken by pig 10. The nominal outer diameter of pipe 12 is known and a nominal wall thickness is assumed, in the preferred embodiment a wall thickness of 3/8" is assumed, to calculate a nominal inner diameter D_TRUE for pipe 12.

Referring to Figure 7A and employing the nomenclature of Figure 5, a skew correction factor table containing r_a/r_c and K_s__ac is calculated over a range of bend radii (R_i) determined by a resolution index i that increases in predetermined increments. Upon completion of the calculation in Figure 7A, a table comprising the prospective radii ratios, their respective skew correction factors (K_s__ac) , and the bend radii (R_L) is indexable by the ratio and is usable as a look-up table. Referring now to Figure 7B, an orthogonal correction factor table for a range of prospective radii ratios is calculated by a resolution index i that increases in predetermined increments. A table comprising the prospective radii ratios and their respective orthogonal correction factors ^κ _o-_bd ^are indexable by the ratio and are also usable as a look-up table.

The calculations presented in Figures 7A-B need only be performed once since the calculations are dependent on prospective radii ratios rather than measured ratios. For instance, the skew correction factor table is equally applicable to corrections of raw diameters d_ac and d_bd regardless of whether prospective radii r_a and r_c are used or whether prospective radii r_b and r_d are used. Similarly, the orthogonal correction factor table is equally applicable to corrections of raw diameters d_ac and d_bd regardless of whether prospective radii r_a and r_c are used or whether prospective radii r_b and r_d are used.

Once the skew correction factor table and the orthogonal correction factor table are complete, the analysis for each datapoint then proceeds by calculating the raw diameters d_ac and d_bd as the sum r_a+r_c and r_b+r_d, respectively, for each respective pair of constituent radii. The skew correction factor and the orthogonal correction factor for each raw diameter are then retrieved from the correction tables. For instance, the skew correction factor K₈__ac is obtained by retrieving the correction factor K_s in the skew correction factor table corresponding to the ratio r_a/r_c and the orthogonal correction factor K₀__ac by retrieving the correction factor K₀ in the orthogonal correction factor table corresponding to the ratio r_b/r_d. Similarly, the skew correction factor ^κ _s-_bd *^s obtained by retrieving the correction factor K_s in the skew correction factor table corresponding to the ratio r_b/r_d and the orthogonal correction factor K₀__bd by retrieving the correction factor K₀ in the orthogonal correction factor table corresponding to the ratio r_a/r_c. D_ac and D_bd are then calculated by multiplying d_ac by K_s__ac and K_Q__bd and by multiplying d_bd by K_B__bd and K₀._ac, respectively.

An apparent bend radius is obtained for each pair of arms by extracting the bend radius for each arm ratio from the previously described skew correction factor table. The actual bend radius is then approximated by the arm pair indicating the smallest bend radius. Furthermore, since the orientation of pig 10 to orientation sensor 40 is known and orientational sensor 40 provides an output proportional to the gravity vector, the orientation of any pair of arms with respect to the gravity vector may be readily calculated. Knowing the orientation of the pair of arms indicating the bend radius, allows determination of the direction of the bend, i.e., over-bend, under-bend, left- bend, right-bend, etc. The directions of each corrected diameter may also be used to deduce the orientation of an ovality or a mash.

One effective way of presenting and analyzing the information resulting from the previously described techniques is to plot it on a graph, with the diameters and bend radius information plotted vertically verses the distance along the pipeline plotted horizontally. Examples may be found in Figures 6B, 6C, and 9. If both the minimum and maximum diameters are plotted at any one distance as shown in Figures 6C and 9, the ovality or mash conditions may readily be seen by the vertical distance between the minimum and maximum diameters on the graph.

A 12" pipe test section, with the geometry shown in Figure 6A, consists of a 90° 1.5R bend at the beginning and a mash near the end with a minimum inside dimension of 10.24 inches. Figure 6B shows the plotted diameter information, and Figure 6C shows the plotted bend radius information. These graphs show that while pig 10 is measuring the information in the bend, as in data set 58, due to the unique correction factors for skew and orthogonality, the calculated diameters in the bend reflect the actual pipe profile. Data set 54 in Figure 6B reflects the mash near the end of the pipe.

Looking now at Figure 9, a pipe ovality condition is portrayed in data set 55, wherein the pipe is deformed into an oval shape with the resultant data showing an increase in maximum diameter and a decrease in minimum diameter. In contrast, a mash condition as in data set 57 and data set 54 of Figure 6B shows little change in pipe maximum diameter with decrease in pipe minimum diameter. A display of the bend radius information for the pipe geometry of Figure 6A is shown in Figure 6C and is correlatable to the diameter information as both are plotted versus distance travelled. The absence of extraneous diameter excursions from the nominal diameter in data set 56 in Figure 6B corresponding to bend data set 58 in Figure 6C indicates a true bend, since the diameters remain relatively constant, is a bend condition. Note that the bend data indicated by data set 60 is erratic, and also corresponds with the mash condition of data set 54, indicating there is no bend at that location.

Figure 8 illustrates an alternative embodiment for pig 10 with like parts having like numbers. Central mandrel 16 is segmented and comprises forward segment 70 connected through universal joint 72 with rear segment 74. Odometer 24 is mounted to the terminus of rear segment 74 and both electronics package 44 and orientational sensor 40 are housed in forward segment 70. However, electronics package 44 and orientational sensor 40 may be housed in rear segment 74 in other embodiments. Perhaps the greatest difference in the embodiment of Figure 8 is the implementation of the radial sensors. Contacts 20a-d shown in Figure 1 are replaced by sliding wear surfaces 76a-d in Figure 8. Surfaces 76a-d may be constructed of any suitable wear-resistant material as is known in the art. Surfaces 76a-d are prolate spheroids having a major radius less than 1/2 the internal diameter of pipe 12 and an arc width yielding full circumferential calipering of wall 18 with multiple sets of arms. The signals output by radial sensors 36a-d (shown in Figure 2) mounted on rear segment 74 are transmitted to electronics package 44 housed in forward segment 70 through an umbilical cable (not pictured) .

It will furthermore be evident to those of ordinary skill in the art having the benefits of the teachings above that the invention claimed below may be embodied in alternative and equally satisfactory embodiments without departing from the spirit or essential characteristics thereof. By way of illustration, electronics package 44 can include a clock synchronized to an external clock with which the passage of pig 10 past preselected points in the pipeline can be manually observed and correlated to recorded datapoints. Alternatively, external markers may be used to correlate pig 10's position in pipe section 12 with a clock in manner those of ordinary skill in the art having the benefits of the teachings herein will readily recognize. Thus, the preferred embodiments set forth above must consequently be considered illustrative and not limiting of the scope of the invention claimed below.

Claims

WHAT IS CLAIMED IB:

1. An instrumented inspection pig for inspecting pipe in a pipeline, comprising: a central mandrel; two supporting members affixed to the central mandrel; two pairs of diametrically opposed radial sensors mounted to the central mandrel, each radial sensor measuring a quantity proportional to the respective distance between the pipe wall and the axial centerline of the pig at the location of the radial sensor and outputting an electrical signal proportional to the respective distance; an orientational sensor measuring the orientation of the pig relative to the gravitational vector within the pipe and outputting an electrical signal proportional to the same; means for receiving the electrical signals from the radial sensors and the orientational sensor and for storing the data therein; and a power supply providing power to the radial sensors, the orientational sensor, and the receiving and storing means.

2. The pig of claim 1, wherein at least one of the radial sensors comprises: a spring-loaded measuring arm pivotably mounted to the central mandrel to mechanically engage the wall of the pipe and trace the contour of the wall; a rotational sensor measuring the angular deflection of the measuring arm relative to the central mandrel and outputting an electrical signal proportional to the distance between the wall and the pig's axial centerline as the measuring arm traces the contour of the wall.

3. The inspection pig of claim 2, wherein the measuring arms mechanically engage the pipe through prolate spheroids.

4. The inspection pig of claim 3, wherein the pairs of measuring arms are orthogonal.

5. The pig of claim 4, wherein the radial sensors are mounted intermediate the support members.

6. The pig of claim 5, including an odometer measuring the distance traveled from the start of the pipe and generating an electrical signal proportional thereto, which signal is received by the receiving means, the data in which signal being stored by the receiving means.

7. The inspection pig of claim 6, wherein the odometer further comprises: two arms pivotably mounted to a terminus of the central mandrel at a point substantially collinear to the central, longitudinal axis of the central mandrel; two wheels, each wheel rotatably mounted to one of the arms at the end most distal to the central mandrel; and a magnet mounted to the circumference of at least one of the wheels at evenly space intervals.

8. The inspection pig of claim 7, wherein the receiving and storing means comprises a processor and a memory.

9. The inspection pig of claim 8, wherein the central mandrel includes a fluid-tight compartment in which at least one of the power supply, the receiving means, and the orientational sensor are located.

10. The inspection pig of claim 9, wherein the supporting members are mounted at points spaced apart on the central mandrel.

11. The inspection pig of claim 10, wherein the storing and receiving means samples data at a predetermined rate.

12. The inspection pig of claim 11, wherein the orientational sensor is a pendulum sensor.

13. The pig of claim l, including an odometer measuring the distance traveled from the start of the pipe and generating an electrical signal proportional thereto, which signal is received by the receiving means, the data in which signal being stored by the receiving means.

14. The inspection pig of claim 13, wherein the odometer further comprises: two arms pivotably mounted to a terminus of the central mandrel at a point substantially collinear to the central, longitudinal axis of the central mandrel; two wheels, each wheel rotatably mounted to one of the arms at the end most distal to the central mandrel; and a magnet mounted to the circumference of at least one of the wheels at evenly space intervals.

15. The inspection pig of claim 1, wherein the receiving and storing means comprises a processor and a memory.

16. The inspection pig of claim 1, wherein the central mandrel is segmented.

17. The inspection pig of claim l, wherein the supporting members are mounted at points spaced apart on the central mandrel.

18. The inspection pig of claim 1, wherein the pairs of radial sensors are orthogonal.

19. The inspection pig of claim 1, wherein the storing and receiving means samples data at a predetermined rate.

20. The inspection pig of claim 1, wherein the orientational sensor is a pendulum sensor.

21. A method for detecting interesting conditions in the pipe of a pipeline comprising the steps of: passing a pig through the pipe; measuring the radial distance from the wall of the pipe to the axial centerline of the pig at two orthogonal pairs of diametrically opposed locations about the circumference of the pipe at a specific point in the pipeline, each distance being measured independently of the others, as the pig passes through the pipeline; and analyzing the four measured distances to detect whether there is an interesting condition at the specific point in the pipeline.

22. The method of claim 22, wherein the step of analyzing the four measured distances includes: deriving therefrom two raw diameters for the pipe at the specific point; correcting each of the two raw diameters for skew and orthogonality; and comparing the two diameters to detect whether there is at least one of a dent and an ovality condition in the pipe at the specific point.

23. The method of claim 23, wherein the step of correcting the raw diameters for skew and orthogonal error includes: determining skew and orthogonal correction factors from the measured radial distances; and applying the skew and orthogonal correction factors to the raw diameters.

24. The method of claim 24, wherein the skew correction factor is determined by: constructing a skew correction factor table from prospective radii ratios, an assumed nominal inner diameter, and a range of assumed bend radii, the table containing each prospective radii ratio, its respective bend radius, and its respective skew correction factor entry; taking the actual ratio of the measured distances co- linear with the respective raw diameter being corrected; comparing the actual ratio of the measured distances with the prospective radii ratios to determine which of the prospective radii ratios most nearly matches the actual ratio of the measured distances; retrieving the skew correction factor of the most nearly matching prospective radii ratio.

25. The method of claim 25, wherein the skew correction factor table is constructed by: determining the pig position in a theoretical pipeline having the nominal inner diameter over the range of bend radii; calculating the diametrically opposite radial distances from the pig centerline to the inner wall of the theoretical pipeline for each bend radius in the range of bend radii; calculating the skew correction factor for each pair of calculated radial distances by dividing the assumed nominal inner diameter by the sum of the calculated radial distances; calculating the prospective radii ratio from the calculated radial distances for each pair of radial distances; and tabulating each bend radius and its respective skew correction factor and prospective radii ratio.

26. The method of claim 26, wherein the approximate bend radius at the specific point of the pipeline is determined by: taking the actual ratio of each pair of the measured distances; comparing the actual ratios with the prospective radii ratios to determine which of the prospective radii ratios most nearly matches the actual ratios of the measured distances; retrieving the respective bend radii of the most nearly matching prospective radii ratios; comparing the respective bend radii; and designating the lesser of the respective bend radii as the approximate bend radius for the pipeline at the preselected point.

27. The method of claim 24, wherein the orthogonal correction factor is determined by: constructing an orthogonal correction factor table from prospective radii ratios and an assumed nominal inner diameter, the table containing each prospective radii ratio and its respective orthogonal correction factor entry; taking the actual ratio of the measured distances orthogonal to the respective raw diameter being corrected; comparing the actual ratio of the measured distances with the prospective radii ratios to determine which of the prospective radii ratios most nearly matches the ratio of the measured distances; retrieving the orthogonal correction factor of the most nearly matching prospective radii ratio.

28. The method of claim 28, wherein the orthogonal correction factor table is constructed by: calculating a plurality of prospective radii ratios and their respective orthogonal correction factors over a range of index values incremented in predetermined increments, each set of prospective radii ratios and orthogonal correction factors being calculated by: setting a first prospective radius to the value one-half the nominal inner diameter divided by the respective index value; setting a second prospective radius to the difference between the nominal inner diameter and the first prospective radius; calculating the prospective radii ratio by dividing the second prospective radius by the first prospective radius; calculating the centerline offset by halving the difference of the second prospective radius and the first prospective radius; calculating the orthogonal correction factor by dividing the nominal inner diameter by twice the square root of the difference between the square of the nominal inner diameter and the square of the centerline offset; and tabulating each prospective radii ratio and its respective orthogonal correction factor.

29. The method of claim 23, wherein the bend radius of the pipe is derivable from the measured radial distances.

30. The method of claim 23, wherein distances are measured and analyzed at a plurality of specific points.

31. The method of claim 23, wherein the pig comprises: a central mandrel; two supporting members affixed to the central mandrel; two pairs of diametrically opposed radial sensors mounted to the central mandrel and measuring a quantity proportional to the respective distance between the pipe wall and the axial centerline of the pig at the location of the radial sensor and outputting an electrical signal proportional to the respective distance; an orientational sensor measuring the orientation of the pig relative to the gravitational vector within the pipe and outputting an electrical signal proportional to the same; means for receiving the electrical signals from the radial sensors and the orientational sensor and for storing the data therein; and a power supply providing power to the radial sensors, the orientational sensor, and the receiving and storing means.