COMPUTER-IMPLEMENTED GEOMETRY MONITORING METHOD

Description

Domaine technique

[0001] The invention relates to a computer-implemented geometry control method for verifying the geometry of a structure, in particular for verifying the geometry of a load-bearing structure for a liquefied gas storage facility. More specifically, the structure to be verified may have a generally vertical wall with a generally regular polygonal shape. Arrière-plan technologique

[0002] US patent 8,550,276 B2 describes a liquefied gas storage facility comprising a vertical wall and a bottom wall, where the bottom wall has a plurality of sectors that are rotationally mirror images of each other, and where the bottom wall is in the shape of a regular polygon, each side of which corresponds to one of these sectors. Such a structure is advantageous because it allows each sector to be constructed using identical elements, thus reducing the number of different elements required. In particular, a large portion of the bottom wall is constructed using rectangular elements of identical dimensions.

[0003] Another liquefied gas storage facility of this type is known from document WO 2011 / 048300 A1. In this document as well, the facility comprises a vertical wall and a bottom wall. The vertical wall has a plurality of vertical faces. The bottom wall includes a plurality of rectangular pieces arranged in sectors that are mirror images of each other by rotation, the edges of the rectangular pieces in one of these sectors being respectively parallel and perpendicular to one of the vertical faces. However, unlike document US 8,550,276 B2, the number of vertical faces is twice the number of sectors. The number of vertical faces is, for example, set at 56.As described in this document, planning for a high number of vertical panels, in particular twice the number of sectors, helps to limit the amount of material needed to create the load-bearing structure that will receive the vertical wall and the bottom wall, for the same storage volume.

[0004] In both of these documents, the load-bearing structure is, for example, made of concrete.

[0005] As for document KR20200039948A, it describes a computer-implemented geometry control method for measuring the distance between two facing walls of an orthogonally shaped tank, the tank being used to store liquefied natural gas by means of a laser distance sensor. Résumé

[0006] Certain aspects of the invention are based on the observation that the load-bearing structure, in practice, exhibits some dimensional deviations from the ideally intended regular polygon shape. Such dimensional deviations can make construction of the installation difficult. It is therefore necessary that the dimensional deviations not exceed pre-specified tolerances. However, if the load-bearing structure is large, manually verifying that it falls within the specified tolerances can be time-consuming, during which construction of the installation cannot proceed, since it is unknown whether the load-bearing structure conforms to the specifications.

[0007] An idea underlying the invention is to provide a method for controlling the geometry of a structure, which is implemented by computer from a plurality of measurements, each representing a three-dimensional measured position of a point located on the structure to be controlled.

[0008] The invention thus proposes a computer-implemented geometry control method, comprising: specify a structural model, for example representing a load-bearing structure for a liquefied gas storage facility, the structural model having geometric and dimensional attributes, the geometric attributes having a number N of sides of a polygon, N being an integer greater than or equal to 3, the dimensional attributes including an acceptable diameter range; provide a plurality of first measurements, each representing a measured position in a three-dimensional coordinate system of a point located on a structure to be controlled; determine a cylindrical coordinate system having a vertical axis parallel to an axis of said three-dimensional coordinate system, said vertical axis having a position that minimizes a dispersion of said first measurements along a radial coordinate in said cylindrical coordinate system;specify first disjoint sectors of space in the cylindrical coordinate system, said first sectors being defined by azimuthal coordinate increments and axial coordinate increments in said cylindrical coordinate system; calculate, from the first measurements, a plurality of first discretization points, each first discretization point representing an average position of the structure to be controlled in one of said first sectors; calculate, from the first discretization points, estimated positions of N vertical edges of the structure to be controlled; calculate, from the estimated positions of the N vertical edges, estimated positions of N medians, each median being a median of a vertical face of the structure to be controlled; and from the estimated positions of the N medians, compare distances between diametrically opposite medians with said acceptable diameter range.

[0009] With the geometry control process as defined above, the geometry of the load-bearing structure can be checked automatically from the initial measurements, meaning without user intervention except for specifying the load-bearing structure model. This makes geometry control easy to implement even when the load-bearing structure is very large.

[0010] The number N represents the number of sides of a polygon that will serve as the directrix for a vertical wall of the structure to be controlled. The vertical wall, if compliant, will closely resemble an ideal shape composed of N vertical faces separated by N vertical edges, forming a polygonal cylindrical surface with a regular N-sided polygon as its directrix. In other words, the acceptable diameter range can represent an acceptable distance between the respective medians of two diametrically opposed vertical faces, ensuring sufficient resemblance to the ideal shape.

[0011] According to embodiments, the geometry control process may include one or more of the following characteristics.

[0012] According to one embodiment, said axis of said three-dimensional frame is parallel to the direction of the Earth's gravitational field at the location of the structure to be controlled.

[0013] According to one embodiment, the dimensional attributes further include an acceptable radius range, and the geometry control method further includes comparing distances between said estimated positions of the median N and said vertical axis with said acceptable radius range.

[0014] In other words, the acceptable radius range can represent an acceptable distance between the medians and said vertical axis, which ensures a sufficient resemblance to the ideal shape.

[0015] According to one embodiment, the dimensional attributes further include an acceptable range of panel width, and the geometry control method further includes comparing distances between the estimated positions of two adjacent vertical edges with said acceptable range of panel width.

[0016] In other words, the acceptable range of pan width can represent an acceptable distance between two adjacent vertical edges, which ensures sufficient resemblance to the ideal shape.

[0017] According to one embodiment, the dimensional attributes further include an acceptable range of edge non-verticality, and the geometry control method further includes comparing distances, perpendicular to the vertical axis of said cylindrical coordinate system, between endpoints of each vertical edge with said acceptable range of edge non-verticality.

[0018] In other words, the acceptable range of edge non-verticality can represent an acceptable distance, perpendicular to a vertical axis of the structure to be controlled, between two endpoints of a given vertical edge, which ensures sufficient resemblance to the ideal shape.

[0019] According to one embodiment, the dimensional attributes further include an acceptable range of ovality, and the geometry control method further includes comparing a diameter difference between a larger inscribed circle of the vertical wall and a smaller circumscribed circle of the vertical wall of the structure to be controlled with said acceptable range of ovality.

[0020] In other words, the acceptable range of edge non-verticality can represent an acceptable diameter difference between a larger inscribed circle of the vertical wall and a smaller circumscribed circle of the vertical wall, which ensures sufficient resemblance to the ideal shape.

[0021] According to one embodiment, the dimensional attributes further include an acceptable range of local deformation of the vertical panels, and the geometry control method further includes comparing local deformations of the structure to be controlled with said acceptable range of local deformation of the vertical panels.

[0022] According to one embodiment, the dimensional attributes further include an acceptable range of non-verticality of the vertical panels, and the geometry control method further includes: select a subset of the first discretization points, in which subset the azimuthal coordinates of the first discretization points are equal to each other; determine radial coordinate deviations between the first discretization points of the subset; and compare the radial coordinate deviations with said acceptable range of non-verticality of the vertical slabs.

[0023] According to one embodiment, radial coordinate deviations are radial coordinate deviations between a first discretization point of the subset that has the lowest vertical coordinate and the other first discretization points of the subset.

[0024] According to one embodiment, the dimensional attributes further include an acceptable range of non-planarity of the vertical surfaces, and the geometry control method further comprises: from the estimated positions of the N vertical edges, select the first measurements corresponding to a vertical face of the structure to be checked; calculate a vertical face non-planarity parameter from the first selected measurements, and compare the vertical face non-planarity parameter with said acceptable range of vertical face non-planarity.

[0025] The vertical panel non-planarity parameter can be calculated for one or some of the vertical panels or preferably for all the vertical panels of the structure to be controlled.

[0026] In one embodiment, the geometric attributes further include a flat bottom wall, the dimensional attributes further include an acceptable range of bottom wall non-planarity, and the geometry control method further includes: provide a plurality of second measurements, each representing a position measured in said three-dimensional coordinate system of a point located on a bottom wall of the structure to be controlled; associate to the second measurements a Cartesian coordinate system, said Cartesian coordinate system having a first axis, a second axis and a third axis perpendicular to each other, the third axis coinciding with said vertical axis of said cylindrical coordinate system; specify second disjoint sectors of space in said Cartesian coordinate system, said second sectors being defined by coordinate increments along the first and second axes of said Cartesian coordinate system;calculate, from the second measurements, a plurality of second discretization points, each second discretization point representing an average position of the bottom wall of the structure to be controlled in one of said second sectors; calculate a bottom wall non-planarity parameter from the second discretization points, and compare the bottom wall non-planarity parameter with said acceptable bottom wall non-planarity range.

[0027] According to one embodiment, the dimensional attributes further include an acceptable range of local deformation of the bottom wall, and the geometry control method further includes comparing local deformations of the bottom wall of the structure to be controlled with said acceptable range of local deformation of the bottom wall.

[0028] According to one embodiment, N is an even number, more particularly N is an even number between 8 and 56, more particularly N = 8 or N = 56.

[0029] According to one embodiment, the geometry control process further includes a generation step consisting of generating a visual representation of a result of at least one said comparison, or of all said comparisons or of some of said comparisons.

[0030] Such a visual representation can be displayed to a user or saved for later display, for example. This allows the user to have usable information to decide on corrective actions to be taken on the vertical and / or back wall of the structure being inspected.

[0031] According to one embodiment, the geometry control process further includes a display step consisting of displaying said visual representation on a display device for a user.

[0032] According to one embodiment, the geometry control process further includes the step of assigning a conformity qualification to the structure to be controlled, the conformity qualification being chosen from conforming and non-conforming.

[0033] According to one embodiment, the plurality of first measurements comes from an acquisition of measurements by means of a measuring instrument in the structure to be controlled.

[0034] According to one embodiment, the measuring instrument comprises a laser remote sensing device disposed in an internal space of the structure to be controlled.

[0035] According to one embodiment, the invention further provides a computer program comprising instructions which, when the program is executed by a computer, cause the computer to implement the geometry control method according to any one of the embodiments described above.

[0036] According to one embodiment, the invention further provides a computer-readable data carrier on which said computer program is recorded.

[0037] According to one embodiment, the invention further provides a geometry control device comprising at least one processor and at least one memory containing a computer program, wherein said at least one memory and the computer program are configured, with said at least one processor, to cause the geometry control method to be executed according to any one of the embodiments described above by the geometry control device.

[0038] The measurements can be provided to the computer program and the geometry control device in various ways. In one embodiment, the geometry control device includes the aforementioned measuring instrument, the measuring instrument being configured to record the first plurality of measurements and, if applicable, the second plurality of measurements in the memory of the geometry control device. In another embodiment, the geometry control device is configured to receive the first plurality of measurements and, if applicable, the second plurality of measurements on a data carrier or via a network interface.

[0039] According to one embodiment, the invention further provides a conformity control method for verifying the geometric conformity of a structure to be inspected, the conformity control method comprising: an acquisition step consisting of acquiring a plurality of measurements each representing a three-dimensional position measured from a point located on the structure to be controlled; and the implementation of the geometry control process according to any one of the embodiments described above.

[0040] According to one embodiment, the acquisition step is carried out in several sub-steps, the laser remote sensing device being placed at each sub-step in a different location within the internal space of the structure to be controlled.

[0041] According to one embodiment, during the acquisition step, the measurements are associated with a measurement of the direction of the Earth's gravity field at the location of the structure to be controlled.

[0042] According to one embodiment, the structure to be controlled is a load-bearing structure for a liquefied gas storage facility.

[0043] According to one embodiment, the structure to be controlled is made of concrete.

[0044] In one embodiment, the liquefied gas storage facility is intended to be installed on land. In this case, the structure to be controlled can be made of concrete.

[0045] In another embodiment, the liquefied gas storage installation is intended to be installed on board a floating structure, such as a ship. In this case, the structure to be controlled may be a portion of a double hull present on the floating structure.

[0046] In one embodiment, the liquefied gas is LNG, namely a mixture with a high methane content stored at a temperature of approximately -162°C at atmospheric pressure. Other liquefied gases can also be considered, including ethane, propane, butane, or ethylene. Liquefied gases can also be stored under pressure, for example, at a relative pressure between 2 and 20 bar, and in particular at a relative pressure close to 2 bar. The tank can be constructed using various techniques, including as an integrated membrane tank or a self-supporting tank.In particular, the vertical wall of the tank can be made by juxtaposing vertical rows of flat insulating wall modules and vertical rows of corner insulating wall modules as described for example in International Application No. PCT / EP2022 / 057845, published under No. WO 2022 / 200536 A1, or in International Application No. PCT / EP2022 / 057848, published under No. WO 2022 / 200539 A1. Brève description des figures

[0047] The invention will be better understood, and other objects, details, features and advantages thereof will become more apparent from the following description of several particular embodiments of the invention, given solely by way of illustration and not limitation, with reference to the accompanying drawings. [ Fig.1 ] There [ Fig.1 ] represents, in cross-section perpendicular to its vertical axis, the shape of a polygonal load-bearing structure for a liquefied gas storage facility. Fig.2 ] There [ Fig.2 ] is a principled illustration of the non-planarity of a vertical load-bearing panel of the load-bearing structure of the [ Fig.1 ]. Fig.3 ] There [ Fig.3 ] is a block diagram representing the steps in a conformity control process for the load-bearing structure of the [ Fig.1 ]. Fig.4 ] There [ Fig.4 ] is a partial, perspective view from within the load-bearing structure, illustrating a measurement acquisition step carried out within the structure. Fig.5 ] There [ Fig.5 ] is a diagram illustrating, for explanatory purposes, a Cartesian coordinate system in which the measurements obtained during the acquisition stage are expressed. Fig.6 ] There [ Fig.6 ] is a scheme analogous to the [ Fig.1 ], illustrating various dimensional attributes of the load-bearing structure. Fig.7 ] There [ Fig.7 ] is a scheme analogous to the [ Fig.1 ], illustrating, by way of explanation, a parameter of ovalization of the load-bearing structure. Fig.8 ] There [ Fig.8 [ ] is a diagram illustrating, for the purpose of explanation, a parameter of edge non-verticality of the load-bearing structure. ] Fig.9A ] There [ Fig.9A ] is a diagram illustrating, for the purpose of explanation, together with the [ Fig.9B ], the principle of a calculation step used to express the coordinates of the measurements obtained during the acquisition step in the cylindrical coordinate system of the [ Fig.10 ]. Fig.9B ] There [ Fig.9B ] is a diagram illustrating, for the purpose of explanation, together with the [ Fig.9A ], the principle of a calculation step used to express the coordinates of the measurements obtained during the acquisition step in the cylindrical coordinate system of the [ Fig.10 ]. Fig.10 ] There [ Fig.10 [ ] is a diagram illustrating, for explanatory purposes, a cylindrical coordinate system used to express the coordinates of measurements obtained during the acquisition stage. Fig.11A ] There [ Fig.11A ] is a diagram illustrating, for the purpose of explanation, together with the [ Fig.11B ] and the [ Fig.11C ], the principle of discretization performed on measurements of the vertical load-bearing wall. [ Fig.11B ] There [ Fig.11B ] is a diagram illustrating, for the purpose of explanation, together with the [ Fig.11A ] and the [ Fig.11C ], the principle of discretization performed on measurements of the vertical load-bearing wall. [ Fig.11C ] There [ Fig.11C ] is a diagram illustrating, for the purpose of explanation, together with the [ Fig.11A ] and the [ Fig.11B ], the principle of discretization performed on measurements of the vertical load-bearing wall. [ Fig.12 ] There [ Fig.12 [ ] is a diagram illustrating, for the purpose of explanation, the principle of calculating the estimated positions of vertical edges and medians of the vertical faces of the vertical load-bearing wall. Fig.13A ] There [ Fig.13A ] is a scheme analogous to the [ Fig.1 ] and to the [ Fig.6 ], illustrating, by way of explanation, the principle of controlling various parameters of the vertical load-bearing wall through calculation. Fig.13B ] There [ Fig.13B [ ] is a diagram illustrating, by way of explanation, the principle of controlling local deformations of the vertical load-bearing wall by calculation. Fig.13C ] There [ Fig.13C [ ] is a diagram illustrating, by way of explanation, the principle of checking the verticality of the vertical load-bearing wall by calculation. Fig.14 ] There [ Fig.14 ] is a diagram illustrating, for the purpose of explanation, the principle of a discretization performed on the measurements of the load-bearing wall of the base visible on the [ Fig.4 ]. Fig.15A ] There [ Fig.15A [ ] is a graph illustrating, as an explanation, a parameter of non-planarity of the load-bearing bottom wall. Fig.15B ] There [ Fig.15B [ ] is a graph illustrating, by way of explanation, another parameter of non-planarity of the load-bearing bottom wall. ] Fig.16 ] There [ Fig.16 [ ] is a diagram illustrating, by way of explanation, the principle of controlling local deformations of the load-bearing base wall by calculation. Fig.17 ] There [ Fig.17 ] is a functional block diagram of a geometry control device to implement the conformity control process of the [ Fig.3 ]. Fig.18 ] There [ Fig.18 [ ] is a diagram illustrating, by way of explanation, the principle of controlling a non-planarity parameter of the vertical load-bearing wall by calculation. Fig.19 ] There [ Fig.19 [ ] is a diagram illustrating, for explanatory purposes, reference lines that can be used to position insulating wall modules on the vertical load-bearing wall. Fig.20 ] There [ Fig.20 ] is a diagram illustrating, for the purpose of explanation, reference lines that can be used to position insulating blocks on the load-bearing back wall. Description des modes de réalisation

[0048] As mentioned above, the invention relates to the construction of a liquefied gas storage facility, which is referred to as 1 in the description that follows.

[0049] According to one variant, installation 1 is suitable for storing a liquefied gas, in particular liquefied natural gas (LNG) at a temperature of about -162°C and at atmospheric pressure or other liquefied gases.

[0050] The supporting structure 10 is described first. The supporting structure 10 comprises at least one load-bearing wall that defines a cavity for receiving the sealed tank 20. In one embodiment, a main load-bearing wall 12 has an approximately cylindrical geometry surrounding the cavity. Such a main load-bearing wall 12 may also be closed by another load-bearing wall at at least one end in the direction shown. In one embodiment, such a main load-bearing wall 12 may extend between a bottom load-bearing wall and a lid load-bearing wall.

[0051] Installation 1 can be designed to be located on land. The main load-bearing wall 12 is then typically vertical, that is, located in a plane parallel to the direction of gravitational acceleration, within dimensional tolerances. The load-bearing structure 10 is, for example, made of concrete. Not shown in the drawings, the bottom load-bearing wall may be located at ground level or possibly below ground level. Also not shown in the drawings, at the end of the main load-bearing wall 12 opposite the bottom load-bearing wall, the load-bearing structure 10 includes a lid-bearing wall closing the internal space 11 delimited by the bottom load-bearing wall and the vertical load-bearing wall 12. This lid-bearing wall can support various equipment that can be used to convey the liquid product to or from this internal space 11. The bottom load-bearing wall and / or the lid-bearing wall may, for example, be flat.However, other shapes are possible for the bottom load-bearing wall and the lid load-bearing wall, including spherical dome shapes.

[0052] Alternatively, the installation 1 can be designed to be installed on board a floating structure, such as a ship. In this case, the load-bearing structure 10 is a portion of a double hull formed by the floating structure. The main load-bearing wall 12 may optionally be non-vertical, and may even have a direction perpendicular to the direction of gravitational acceleration when the floating structure is at rest.

[0053] In what follows, we will consider more specifically the case of an installation 1 located on land where the main load-bearing wall 12 is vertical. We will therefore refer to it as a vertical load-bearing wall 12. It should be noted, however, that the following description applies to any orientation of the main load-bearing wall 12 with respect to the direction of gravitational acceleration.

[0054] There [ Fig.1 ] is a schematic cross-sectional view of the load-bearing structure 10, taken perpendicular to a vertical axis 9 of the vertical load-bearing wall 12. The vertical load-bearing wall 12 is shown in solid line on the [ Fig.1 The vertical load-bearing wall 12 forms a polygonal cylindrical surface and was typically constructed using civil engineering techniques. Thus, the vertical load-bearing wall 12 has vertical faces 14 separated from each other by edges 13.

[0055] The watertight tank 20 (shown in dotted lines on the [ Fig.1 ]) is intended to be installed in the internal space 11 of the supporting structure 10. The tank 20 has a vertical peripheral wall 22 intended to be opposite the vertical supporting wall 12. Not shown in the drawings, the tank 20 also has a bottom wall opposite the bottom supporting wall and a lid wall opposite the lid supporting wall.

[0056] The vertical peripheral wall 22 can be formed of vertical rows of planar insulating wall modules and vertical rows of corner insulating wall modules as described, for example, in international application no. PCT / EP2022 / 057845, published under no. WO 2022 / 200536 A1, or in international application no. PCT / EP2022 / 057848, published under no. WO 2022 / 200539 A1. The bottom wall can comprise a plurality of angular sectors that are rotationally mirror images of each other as described in WO 2022 / 200536 A1 or WO 2022 / 200539 A1.

[0057] The position and orientation of the vertical panels 14 may exhibit dimensional deviations from an intended regular polygon shape. Furthermore, as schematically represented on the [ Fig.2 Each vertical panel 14 may exhibit dimensional deviations from an ideal planar shape 14P. These dimensional deviations may, for example, be due to dimensional tolerances in a concrete structure. Such dimensional deviations can make the construction of the installation 1 difficult. It is therefore important that the dimensional deviations do not exceed tolerances specified in advance. However, if the load-bearing structure 10 is large, manually verifying that the load-bearing structure 10 falls within the specified tolerances can be time-consuming, during which construction of the installation 1 cannot continue since it is unknown whether the load-bearing structure 10 conforms to the specifications.

[0058] The following is described, with reference to figures 3 à 16 , a conformity control method 1000 (hereinafter referred to as "the 1000 method" for convenience) to check the geometric conformity of the load-bearing structure 1 with respect to a model.

[0059] In a first step 1001 (cf. [ Fig.3 ]) of process 1000, the load-bearing structure 1 having been constructed, a plurality of measurements 50 are acquired (cf. [ Fig.4 ] And [ Fig.5 Each measurement 50 represents a three-dimensional position measured from a point located on the vertical load-bearing wall 12. In principle, each measurement 50 can be acquired by any suitable remote sensing technique. However, it is particularly advantageous to acquire the measurements 50 using a laser remote sensing (lidar) device 91 located in the internal space 11. The operating principle of such a device 91 is known as such and is therefore not described in detail here.

[0060] In a very simple execution variant, the acquisition step 1001 is carried out in one go, the device 91 having been placed at a single point in the internal space 11.

[0061] However, it is preferable for acquisition step 1001 to be carried out in several sub-steps, with the device 91 positioned in a different location within the internal space 11 at each sub-step. This can allow for acquisition with satisfactory resolution over the entire vertical load-bearing wall 12. Of course, in this case, the measurements 50 obtained at each sub-step are post-processed so that they are all expressed in the same spatial coordinate system.

[0062] In any case, at the end of acquisition step 1001, a large number of measurements 50 are available, expressed in the same spatial frame of reference. According to one embodiment, as schematically represented on the [ Fig.5 ], the measurements 50 are expressed in the same Cartesian orthogonal coordinate system (x1, y1, z1), whose origin A1 is arbitrarily fixed. The third coordinate z1 is preferably expressed along an axis corresponding precisely to the direction G (cf. [ Fig.4 ] And [ Fig.5 ]) of the local Earth gravity field, i.e., the Earth gravity field at the location of the supporting structure 10. The direction G of the local Earth gravity field may have been acquired during step 1001 by means of a suitable sensor 92, for example integrated into the device 91 (cf. [ Fig.4 ]). It should be noted that although the figures 4 And 5To illustrate, some 50 measurements are shown. The result of acquisition step 1001 can include a very large number of measurements 50, on the order of 10⁷ to 10⁹ measurements 50 in the case of a load-bearing structure 10 with internal dimensions on the order of 10¹ meters. The number of measurements 50 depends, as is known, on the scanning parameters of device 91 and the number of sub-steps in acquisition step 1001.

[0063] After acquisition step 1001, a geometry control method 100 (hereinafter referred to as "method 100") is implemented, the steps of which are described below. In one embodiment, the geometry control method 100 is implemented by a computer running a suitable computer program. However, method 100 can be implemented by any suitable combination of hardware and software, including in a distributed computing environment.

[0064] The process 100 includes a first step 101 consisting of specifying a model to which the load-bearing structure 10 will be compared.

[0065] The model specifies the number N of sides of the polygon that will serve as the directrix for the vertical load-bearing wall 12. N is an integer greater than or equal to 3. For example, N is an even number. More specifically, N is, for example, an even number between 8 and 56. In one particular embodiment illustrated in the drawings, N is equal to 8. In another particular embodiment, N is equal to 56.

[0066] In what follows, radius Q, diameter D, pan width W, ovality OV, and edge non-verticality NV denote physical quantities of the load-bearing structure 10. The model includes dimensional attributes, including ranges of acceptable values ​​for each of these quantities. figures 6 à 8 illustrate these quantities in the load-bearing structure 10.

[0067] There [ Fig.6 ] illustrates, by way of explanation, the radius Q and the diameter D. As the [ Fig.1 ], there [ Fig.6 ] is a cross-sectional view of the load-bearing structure 10, taken perpendicular to a vertical axis of the vertical load-bearing wall 12. The vertical axis bears the reference 9 on the [ Fig.6 ]. To each vertical panel 14, a midpoint 14A can be associated, located in the cutting plane of the [ Fig.6 The radius Q is the distance between the midpoint 14A and the vertical axis 9 in this cutting plane. The diameter D is the distance between the midpoints 14A of two diametrically opposed vertical faces 14.

[0068] There [ Fig.6 ] also illustrates, by way of explanation, the width of panel W. Considering a vertical panel 14, the width of panel W associated with this vertical panel is the distance, in the cutting plane of the [ Fig.6 ], between the two vertical edges 13 delimiting the panel 14.

[0069] There [ Fig.7 ] illustrates, by way of explanation, the ovalization OV. As the [ Fig.1 ] and the [ Fig.6 ], there [ Fig.7 ] is a cross-sectional view of the load-bearing structure 10, taken perpendicular to the vertical axis 9. Given an actual contour 12R of the vertical load-bearing wall 12 in the cutting plane of the [ Fig.7 We can define a greatest inscribed circle IC in this real contour 12R and a smallest circumscribed circle CC of the real contour 12R. The greatest inscribed circle IC has a diameter denoted D_inner and the smallest circumscribed circle CC has a diameter D_outer. The ovality OV is the difference between these two diameters, that is, OV = D_outer - D_inner.

[0070] There [ Fig.8 ] illustrates, by way of explanation, the non-verticality of edge NV. The [ Fig.8 ] is a schematic front view of an edge 13 separating two vertical faces 14 (not shown on the [ Fig.8 Given two endpoints 13-1 and 13-2 of edge 13, the edge non-verticality NV is the absolute value of a distance, perpendicular to the vertical axis 9, between the endpoints 13-1 and 13-2. It is specified that the endpoints 13-1 and 13-2 can be points of edge 13 in two reference planes, spaced parallel to the vertical axis 9, in the vertical load-bearing wall 12.

[0071] The acceptable range for each of the quantities mentioned above can be specified as a range of numerical values. Alternatively, an acceptable range can be specified as a reference value V and a tolerance T on this reference value V; the acceptable range is then the interval between VT and V+T, this interval being open, closed, or semi-closed, depending on the implementation requirements. When the reference value V is a non-zero value, the tolerance T can, in particular, be specified as a percentage of V. With regard to the diameter D, the radius Q, and the pan width W, the reference value is a non-zero value specified in advance, taking into account the desired dimensions and proportions of the supporting structure 10, and the tolerances are non-zero values ​​specified in advance, taking into account the acceptable tolerances on these desired dimensions and proportions of the supporting structure 10.With regard to ovality OV and edge non-verticality NV, the reference value is zero, and the tolerance is a non-zero value specified in advance.

[0072] The geometric and / or dimensional attributes of the model can be specified by a user, for example, through a computer-implemented user interface. Alternatively, the model can also be specified in advance as a parameter file, which may be editable by the user. Advantageously, the acceptable ranges are editable by the user. In particular, when tolerances are specified as a percentage, as mentioned above, the percentage is advantageously editable by the user. This allows the user to exercise more or less stringent control over the conformity of the load-bearing structure being inspected by changing the percentage. More stringent control would use a lower percentage, for example, advantageously 10%, 5%, or 3%, which amounts to accepting a deviation from the reference value of 10% or less, 5% or less, or 3% or less.A more lenient control will use a higher percentage, for example, advantageously 15%, 20%, or 25%, which amounts to accepting a deviation from the reference value of 15% or less, 20% or less, or 25% or less. In any case, for each of the quantities mentioned above, the acceptable range specifies the deviation that will be accepted between the model and the load-bearing structure to be controlled.

[0073] The process 100 further includes a step 102 in which the measurements 50 described above, taken on the load-bearing structure 10 to be inspected, are provided. Specifically, the measurements 50 are provided as a computer file, on a data storage medium, or via a network interface. For example, the computer file is provided in the .pts file format, which is well-known in the field of laser remote sensing devices.

[0074] The process 100 further includes a step 103 in which a cylindrical coordinate system (r, θ, z) is determined from the measurements 50.

[0075] As is well known, defining a cylindrical coordinate system involves defining a vertical axis along which a z-coordinate is measured, and polar coordinates r and θ perpendicular to this vertical axis. Therefore, in implementing step 103, it is necessary to first define a vertical axis along which the z-coordinate is measured.

[0076] To this end, the Cartesian orthogonal coordinate system (x1, y1, z1) mentioned above, in which the measurements 50 are expressed, is defined such that its vertical axis Z1 is parallel to the direction G of the local Earth's gravitational field. The vertical axis 19 of the cylindrical coordinate system is chosen to be parallel to the vertical axis Z1. It is clear that the vertical axis 19 is thus parallel or nearly parallel to the vertical axis 9 of the supporting structure 10 to be controlled.

[0077] Furthermore, the position of the vertical axis 19 minimizes the dispersion of measurements 50 along the radial coordinate r. It is clear that the vertical axis 19 is thus very close to the vertical axis 9 of the supporting structure 10 to be controlled. Moreover, as will be detailed below, the edges 13 are precisely the locations where local maxima of the radial coordinate r exist.

[0078] Various mathematical methods are appropriate for finding a position of the vertical axis 19 that minimizes a dispersion along the radial coordinate r of the measurements 50. In a simple example of execution, as schematically represented on the [ Fig.9A ], we select several (here, four) subsets T1, T2, T3, T4 of measure 50, the subsets T1, T2, T3, T4 being defined by increments Δz1 of vertical coordinates z1; and we calculate for each subset T1, T2, T3, T4 an interpolation circle C1, C2, C3, C4 (cf. [ Fig.9B ]) associated and perpendicular to the vertical axis Z 1 , then we calculate a barycenter C of the centers of the circles C 1 , C 2 , C 3 , C 4 . The vertical axis 19 is defined as passing through this barycenter C and being parallel to the vertical axis Z 1 .

[0079] Next, the Cartesian coordinates of the measurements 50 in the Cartesian coordinate system (x1, y1, z1) (cf. [ Fig.5 ]) are converted by calculation into cylindrical coordinates (r, θ, z) (cf. [ Fig.10 ]) according to known mathematical relationships. The choice of the direction in which θ = 0° can be made either arbitrarily, or based on the position of a reference point drawn on one of the sides 14 or of a structural element previously chosen on one of the sides 14. The positive direction of the vertical axis 19 is preferably chosen to be opposite to the direction of the acceleration of Earth's gravity.

[0080] The process 100 further includes a step 104 in which disjoint sectors 200 are specified (cf. [ Fig.11C ]) of space in the cylindrical coordinate system (r, θ, z). Concretely, these sectors 200 are defined by an increment δθ of coordinates θ (cf. [ Fig.11A ] And [ Fig.11C ]) and by an increment δz of coordinates z (cf. [ Fig.11B] et [Fig.11C ]).

[0081] Like the numerical quantities of the model described above in relation to step 101, the increments δθ and δz can be specified by a user, for example, using a computer-implemented user interface. Alternatively, the model may also have been specified in advance as a parameter file, possibly editable by the user. The increments δθ and δz can be specified a priori based on knowledge of the reference value Q0 of the radius Q described above. For example, the increments δθ and δz can be specified such that an area of ​​the vertical load-bearing wall 12 included in a given sector 200 is between 1 cm² and 10 cm².

[0082] The process 100 further includes a step 105 in which, from the measurements 50, discretization points 250 are calculated, one of which is shown as an illustration on the [ Fig.11C More specifically, a discretization point 250 is calculated for each sector 200. Each sector 200 is associated with a range of cylindrical coordinates {Θ ± δθ / 2, Z ± δz / 2}. The discretization point 250 is the center of the sector 200. Its cylindrical coordinates are {R, Θ, Z}, where R is the average of the coordinates r of all the measurements 50 located within the sector 200.

[0083] The process 100 further includes a step 106 in which, from the discretization points 250 calculated in step 105, estimated positions 300 of the N vertical edges 13 are calculated.

[0084] This calculation can be carried out in a large number of ways. Advantageously, this calculation consists of searching, using the knowledge of the reference value W 0 of the width of pan W described above, for a set of N vertical lines 13 approximately spaced by W 0 and whose positions correspond to N local maxima of r.

[0085] The principle of such a calculation will be better understood by referring to the [ Fig.12 ]. In this figure, some of the discretization points 250 calculated in step 105 are shown very schematically in the cylindrical coordinate system (r, θ, z), together with estimated positions 300 of the N vertical edges 13. The calculation consists of finding a position of a set of N vertical lines 300 subject to the following two criteria: (i) each of the N 300 lines is sufficiently close to local maxima of r as a function of θ; (ii) each 300 line is separated from a neighboring 300 line by a distance (perpendicular to the vertical axis 19 of the cylindrical coordinate system (r, θ, z)) which is included in the interval [W 0 - W 2 ; W 0 + W 2 ], where W 2 is a non-zero value.

[0086] Note that W2 is not necessarily equal to the tolerance value on W0; on the contrary, W2 can, for example, be chosen to be 2.5 times or 3.0 times this tolerance. Criterion (ii) can then prevent a local depression on one of the vertical faces 14 from being incorrectly considered a vertical edge 13. Note also that the lines 300 do not necessarily pass through discretization points 250, and may even pass through no discretization points 250 at all.

[0087] In any event, at the end of step 106, the estimated positions 300 of the vertical edges 13 are available. The process 100 further includes a step 107 in which, from the estimated positions 300 of the vertical edges 13, estimated positions of the medians 500 of the vertical faces 14 of the load-bearing structure 10 to be checked are calculated. In an example of implementation, still with reference to the [ Fig.12 This step 107 involves calculating the positions of points 400 that are equidistant, perpendicular to the vertical axis 19, from the lines 300. Preferably, to simplify the calculation, the positions of the points 400 are calculated at the same z-coordinate as the discretization points 250. If it turns out that at this z-coordinate, a point 400 does not coincide with a discretization point 250, the r-coordinate of this point 400 is calculated by interpolation (for example, by linear interpolation) from the coordinates of neighboring discretization points 250. The positions of the medians 500 are then calculated by interpolation from the positions of the points 400.

[0088] The process 100 further includes a step 108 in which, from the estimated positions of the medians calculated in step 107, distances between diametrically opposite medians are compared with the acceptable range of diameter D. With reference to the [ Fig.13A ], this calculation can consist of calculating a distance D c between two diametrically opposed points 400 located in the same vertical plane (in other words located at the same z coordinate), and checking that this distance D c falls within the acceptable range of diameter D.

[0089] Some or all of the comparisons listed below can also be made during step 108.

[0090] In particular, with reference to the [ Fig.13A ], from the estimated positions of the medians 500 calculated in step 107, and more specifically the positions of the points 400, we compare the distances Q c between the vertical axis 19 and each of the N points 400 with the acceptable range of radius Q.

[0091] Furthermore, still with reference to the [ Fig.13A ], we calculate the distances between the estimated positions 300 of the vertical edges 13 calculated in step 106, and we compare these distances with the acceptable range of pan width W.

[0092] Furthermore, returning to the [ Fig.7 ], we calculate the positions and diameters D_inner and D_outer of the largest inscribed circle IC and the smallest circumscribed circle CC, we calculate the value of the ovalization OV = D_outer - D_inner, and we compare this calculated value with the acceptable range of ovalization OV.

[0093] In addition, from the lines 300 calculated in step 106, we calculate the positions of two endpoints located on the lines 300, we calculate the absolute value of the distance NV, perpendicular to the vertical axis 19, between these two endpoints, and we compare NV with the acceptable range of edge non-verticality.

[0094] Furthermore, local deformations of the vertical load-bearing panels 14 are compared with an acceptable range of local deformation of the vertical load-bearing panels 14, which was previously specified in step 101. This comparison will be explained in more detail with reference to the [ Fig.13B For a given sector 200, a local interpolation plane 600 is calculated from all the measurements 50 included in this sector 200. A local Cartesian coordinate system (a, b, c) is then associated with this plane 600. This system is orthogonal and defined as follows: axis (c) is perpendicular to the plane 600; axis (b) is parallel to the vertical axis 19 of the cylindrical coordinate system (r, θ, z); and axis (a) is orthogonal to axes (b) and (c). The coordinates of the measurements 50 are then expressed in this local Cartesian coordinate system (a, b, c) according to known mathematical relationships.

[0095] Next, having specified a value σa, we search in sector 200 for measurements 50 whose coordinates (a) are included in an interval of width σa (cf. [ Fig.13B Among these 50 measurements, we find the two measurements 50 furthest from the 600 plane, perpendicular to the axis (c), one in the positive direction of the axis (c) and the other in the negative direction of the axis (c); we calculate the difference σc between the coordinates along the axis (c) of these two measurements 50; and we compare this difference σc with the acceptable range of local deformation. This process is repeated until the sector 200 is exhausted, each time shifting the interval of width σa by a fixed increment; in other words, the interval of width σa is used as a sliding window, which can be of any width along the axis (b) of the local Cartesian coordinate system (a, b, c). The preceding operations are repeated for each of the 200 sectors.

[0096] In addition, a radial coordinate deviation is calculated between all discretization points 250 having the same azimuthal coordinate Θ and the lowest of them, and this deviation is compared with an acceptable range of non-verticality of the vertical load-bearing panels 14, which was previously specified in step 101. This comparison will be explained in more detail with reference to the [ Fig.13C Among all the discretization points 250 having a given coordinate Θ, we obtain the radial coordinate Rb of the discretization point 250b which has the smallest coordinate Zb along the vertical axis 19. Then, for each discretization point 250 having the same coordinate Θ, we calculate the radial coordinate deviation ΔR = R - Rb, and we compare ΔR with the acceptable range of non-verticality of the vertical load-bearing faces 14. The preceding operations are repeated for each coordinate Θ between 0 and 360°.

[0097] Alternatively, the deviation ΔR can be calculated only for certain discretization points 250 having a given coordinate Θ. Alternatively, the deviation ΔR can be calculated only for certain discretization points 250 having a Z coordinate included in one or more given ranges. This can allow for the evaluation of the non-verticality of the vertical load-bearing panels 14 in specific areas thereof, and / or prevent the evaluation of the non-verticality of the vertical load-bearing panels 14 from being distorted by the connection between the bottom load-bearing wall 15 and the vertical load-bearing panels 14, particularly when this connection has the form of a chamfer. Alternatively, the radial coordinate deviation ΔR can be calculated using the radial coordinate of a discretization point 250 other than discretization point 250b.

[0098] The process 100 may optionally include a step 109 of generating at least one visual representation of the conformity of the load-bearing structure with the results of the checks carried out in step 108. For example, this visual representation may include or consist of graphs indicating by colors the regions of the vertical load-bearing wall 12 where the various parameters calculated in step 108 do not fall within the acceptable ranges specified in step 101. To this end, more specifically, for each sector 200 of the vertical load-bearing wall 12 and for each parameter calculated in step 108, a boolean variable is calculated, and takes the value NO (for non-compliant) if the parameter does not fall within the associated acceptable range specified in step 101, and YES (for compliant) otherwise.In addition or as an alternative, the visual representation may include or consist of graphs indicating by colors the various parameters calculated in step 108 on a graphic representation of the vertical load-bearing wall 12.

[0099] In step 1010 of process 1000, the visual representation(s) generated in step 109 can be displayed to a user, providing them with usable information to decide on corrective actions to be taken on the vertical load-bearing wall 12. The process can also result in generating a conformity rating chosen from conforming and non-conforming, for example, a Boolean variable. In one example, the Boolean variable could take the value YES (for conforming) if a percentage of non-conforming sectors of the vertical load-bearing wall 12 is less than a predetermined percentage, and take the value NO (for non-conforming) otherwise.

[0100] So far, only a geometry check of the vertical load-bearing wall 12 has been described. However, as mentioned previously, the load-bearing structure 10 may also include a bottom load-bearing wall. A variant of the method 100 is described below which is applicable in the case where this bottom load-bearing wall 15 (cf. [ Fig.4 ]) is flat.

[0101] In this variant, at acquisition step 1001, the apparatus 91 also acquires a plurality of measurements 51 (cf. [ Fig.4 ]), where each measurement 51 represents a three-dimensional position measured from a point located on the bottom load-bearing wall 15 of the load-bearing structure 10 to be controlled. In step 102, these measurements 51 are provided at the same time as the measurements 50.

[0102] The measurements 51 are expressed in the same Cartesian orthogonal coordinate system (x 1 , y 1 , z 1 ) as the measurements 50.

[0103] In step 103, the Cartesian coordinates of the measurements 51 in the Cartesian coordinate system (x1, y1, z1) (cf. [ Fig.5 ]) are converted by calculation, according to known mathematical relations, into Cartesian coordinates in a Cartesian coordinate system (x, y, z) whose vertical axis z coincides with the vertical axis 19 of the cylindrical coordinate system for measurements 50.

[0104] In step 104, disjoint sectors 201 of space are further specified in the Cartesian coordinate system (x, y, z). Specifically, these sectors 201 are defined by an increment δx of x coordinates and by an increment δy of y coordinates (cf. [ Fig.14 ]). Like the numerical quantities of the model described above in relation to step 101, the increments δx and δy can be specified by a user, for example, by means of a computer-implemented user interface. Alternatively, the model may also have been specified in advance as a parameter file, possibly editable by the user. The increments δx and δy can be specified a priori based on knowledge of the reference value Q0 of the radius Q described above. For example, the increments δx and δy can be specified such that an area of ​​the load-bearing bottom wall 15 included in a given sector 201 is between 5 cm² and 50 cm². In step 105, again with reference to the [ Fig.14 A discretization point 251 is calculated for each sector 201. Each sector 201 is associated with a range of Cartesian coordinates {X ± δx / 2, Y ± δy / 2}. The discretization point 251 is the center of sector 201. Its Cartesian coordinates are {X, Y, Z}, where Z is the average of the z-coordinates of all the measurements 51 located within sector 201.

[0105] In step 108, a dispersion Δz is calculated along the vertical z-axis of the measurements 51; in other words, the difference between the highest and lowest z-coordinates of the measurements 51 is calculated. With reference to the [ Fig.15A [ ], which is a graph representing a statistical distribution of the z coordinates of the measurements 51 in an example, we understand that the higher Δz is, the higher the highest asperity of the bottom load-bearing wall 15. Δz is therefore a parameter that quantifies the non-planarity of the bottom load-bearing wall 15. Still in step 108, we compare Δz with an acceptable range of non-planarity of the bottom load-bearing wall 15, which was previously specified in step 101.

[0106] In step 108, a global interpolation plan 601 is also calculated from all the measurements 51 (cf. [ Fig.15B ]). Next, we find the two measures 51 furthest from the plane 601, parallel to the vertical axis z, one in the positive direction of the vertical axis z and the other in the negative direction of the axis z; and we calculate the difference Δd between the z coordinates of these two measures 51. With reference to the [ Fig.15B [ ], which is a graph representing a statistical distribution of the z-coordinates of the measurements 51 in another example, we understand that in this way, Δd quantifies the non-planarity of the load-bearing bottom wall 15 differently from Δz. Still in step 108, we compare Δd with an acceptable range of non-planarity of the load-bearing bottom wall 15, which may or may not be identical to that to which we compare Δz.

[0107] Alternatively, only one of the dispersions Δz or Δd can be calculated.

[0108] Advantageously, the dispersion Δz can be calculated only on the measurements 51 that are located sufficiently far from the vertical load-bearing panels 14. More specifically, thanks to prior knowledge of the reference value Q0 of the radius Q, measurements 51 whose cylindrical coordinate r satisfies r > R0 - R2, where R2 is a predetermined threshold, can be excluded from the dispersion Δz calculation. The same is advantageously true for the dispersion Δd. This can prevent the evaluation of the non-planarity of the bottom load-bearing wall 15 from taking into account the connection between the bottom load-bearing wall 15 and the vertical load-bearing panels 14, particularly when this connection has the form of a chamfer.

[0109] A similar comparison with an acceptable range of non-planarity of the vertical load-bearing panels 14 can also be carried out in step 108. From the estimated positions 300 of the vertical edges 13 calculated in step 106, all the measurements 50 corresponding to a given vertical load-bearing panel 14 are selected, and a global interpolation plane 701 is calculated (cf. [ Fig.18 ]) from the 50 measurements thus selected. It should be noted that the global interpolation plane 701 is not necessarily parallel to the local interpolation plane 600 represented on the [ Fig.13B ]. Next, a Cartesian coordinate system (i, j, k) is associated with the global interpolation plane 701. This system is orthogonal and defined as follows: the (k) axis is orthogonal to the 701 plane; and the (i) and (j) axes are included in the 701 plane. Still with reference to the [ Fig.18 We search among the measurements 50 used to calculate plane 701 for the two measurements 50 furthest from plane 701, parallel to axis (k), one in the positive direction of axis (k) and the other in the negative direction of axis (k); and we calculate the difference Δk between the k coordinates of these two measurements 50. Still in step 108, we compare Δk with an acceptable range of non-planarity of the vertical load-bearing panels 14, which was previously specified in step 101. It will be understood that, analogously to Δd, Δk quantifies the non-planarity of the vertical load-bearing panel 14. Furthermore, it is understood that Δk can be calculated and compared with the acceptable range of non-planarity of the vertical load-bearing panels 14 for one, some, or all of the vertical load-bearing panels 14.

[0110] The global interpolation plane 601 can also be used to control local deformations of the bottom load-bearing wall 15, as will be described with reference to the [ Fig.16 ]. For a given sector 201, a value σx having been specified, we search for the measurements 51 whose x coordinates are included in an interval of width σx (cf. [ Fig.16 Among these measurements 51, the two measurements 51 furthest from plane 601, parallel to the vertical z-axis, are sought in sector 201, one in the positive direction of the vertical z-axis and the other in the negative direction of the z-axis. A difference σz is calculated between the coordinates along the vertical z-axis of these two measurements 51, and this difference σz is compared with an acceptable range of local deformation of the bottom wall, which was previously specified in step 101. This process is repeated until sector 201 is exhausted, each time shifting the width interval σx by a fixed increment. In other words, the width interval σx is used as a sliding window, which can be of any width along the y-axis of the Cartesian coordinate system (x, y, z). The preceding operations are repeated for each of the sectors 201.Alternatively, the two measurements 51 used to calculate σz can be the two most distant measurements of a local interpolation plane which is calculated from all the measurements 51 included in sector 201. It should be noted that this local interpolation plane is not necessarily parallel to the global interpolation plane 601.

[0111] The principles described above for a flat bottom load-bearing wall 15 are also applicable to a flat lid wall of the load-bearing structure 10.

[0112] The geometry control procedures described above can be applied to any civil engineering structure or any welded structure to be built in accordance with the specified model.

[0113] As mentioned above, the vertical perimeter wall 22 can be formed of vertical rows of flat insulating wall modules and vertical rows of corner insulating wall modules as described in WO 2022 / 200536 A1 or in WO 2022 / 200539 A1. As described in these documents, knowledge of the three-dimensional positions of the N vertical edges 13 makes it possible to draw, on the vertical load-bearing panels 14, vertical reference lines that can be used to position the vertical rows of flat insulating wall modules and the vertical rows of corner insulating wall modules on the vertical load-bearing panels 14. On the [ Fig.19 ], we have schematically represented, for the purpose of explanation, such vertical reference lines 800, together with orthoradial reference lines 825, for a vertical load-bearing panel 14. With reference to the [ Fig.19 ], the vertical reference lines 800 and the orthoradial reference lines 825 together delimit disjoint sectors 830, each sector 830 corresponding to the intended location of a planar insulating wall module or a corner insulating wall module on the vertical load-bearing panel 14.

[0114] According to a variant of process 100, from the estimated positions 300 of the N vertical edges 13 calculated in step 106 and the measurements 50, the positions of the vertical reference lines 800 are calculated as described in WO 2022 / 200536 A1 or in WO 2022 / 200539 A1; from the calculated positions of the vertical reference lines 800 and the planned dimensions of the insulating modules, the positions of the orthoradial reference lines 825 are further calculated; and in step 108, local deformations of the vertical load-bearing panels 14 are compared with the acceptable range of local deformation of the vertical load-bearing panels 14 as described above, except that, instead of sectors 200, sectors 830 are used or sectors of a predetermined radius around the geometric center of each sector 830.

[0115] In this way, the comparison of the local deformations of the vertical load-bearing panels 14 with the acceptable range of local deformation of the vertical load-bearing panels 14 is better indicative of corrective actions to be taken on the vertical load-bearing wall 12 so that the flat insulating wall modules and the corner insulating wall modules can be positioned as desired on the vertical load-bearing panels 14.

[0116] As described in WO 2022 / 200536 A1 or in WO 2022 / 200539 A1, knowledge of the three-dimensional positions of the N vertical edges 13 also makes it possible to draw, on the flat load-bearing base wall 15, horizontal reference lines that can be used to position the angular sectors that are images of each other by rotation and the insulating blocks forming these angular sectors. On the [ Fig.20 ], we have schematically represented, for the purpose of explanation, such horizontal reference lines 900, together with orthoradial reference lines 925. With reference to the [ Fig.20 ], the horizontal and orthoradial reference lines 925 together delimit disjoint sectors 930, each sector 930 corresponding to the intended location of an insulating block on the load-bearing bottom wall 15. It should be noted that the sectors 930 can have various shapes, including rectangles, trapezoids, etc. The [ Fig.20 ] is therefore in no way limiting in this respect.

[0117] According to a variant of process 100, from the calculated positions of the vertical reference lines 800, the positions of the horizontal reference lines 900 are calculated as described in WO 2022 / 200536 A1 or in WO 2022 / 200539 A1; from the calculated positions of the horizontal reference lines 900 and the planned dimensions of the insulating blocks, the positions of the orthoradial reference lines 925 are further calculated; and in step 108, local deformations of the flat bottom load-bearing wall 15 are compared with the acceptable range of local deformation of the bottom load-bearing wall as described above, except that, instead of sectors 201, sectors 930 are used or sectors of a predetermined radius around the geometric center of each sector 930.

[0118] In this way, comparing the local deformations of the flat bottom load-bearing wall 15 with the acceptable range of local deformation of the bottom load-bearing wall is better indicative of corrective actions to be taken on the vertical load-bearing wall 12 so that the insulating blocks can be positioned as desired on the bottom load-bearing wall 15.

[0119] Since sectors 830 and 930 correspond to the intended location of insulating wall modules or insulating blocks, measurements 50 and 51 included in sectors 830 and 930 can also be used to calculate the dimensions of spacers. For example, spacers include shims and / or sealant beads. These spacers are intended to compensate for any unevenness in the vertical load-bearing panels 14 and the bottom load-bearing wall 15. For this purpose, the spacers are positioned between an insulating wall module and a vertical load-bearing panel 14, or between an insulating block and the bottom load-bearing wall 15. The dimensions of the spacers can be calculated as described, for example, in WO 2023 / 073201 A1.

[0120] The geometry control procedures described above can be implemented using a 3000 geometry control device (see [ Fig.17The geometry control device 3000 comprises at least one processor 3001 and at least one memory 3002 containing a computer program; the at least one memory 3002 and the computer program are configured, along with said at least one processor 3001, to cause the execution of the geometry control method 100 described previously. The geometry control device 3000 can be implemented in various forms, either unitary or distributed, using hardware and / or software components. Usable hardware components include ASICs, FPGAs, or microprocessors. Software components can be written in various programming languages, for example, C, C++, Java (registered trademark), or VHDL. This list is not exhaustive.

[0121] Although the invention has been described in connection with several particular embodiments, it is clearly evident that it is by no means limited to them and that it includes all technical equivalents of the means described as well as their combinations if these fall within the scope of the invention.

[0122] The use of the verb "comporter", "comprendre" or "include" and its conjugated forms does not exclude the presence of other elements or steps than those stated in a claim.

[0123] In claims, any reference sign in parentheses shall not be interpreted as a limitation of the claim.

Claims

1. A computer-implemented geometry monitoring method (100) comprising: - specifying (101) a structure design (10), the structure design (10) having geometric attributes and dimensional attributes, the geometric attributes including a number N of sides of a polygon, N being an integer greater than or equal to 3, the dimensional attributes including an acceptable range of diameters (D); - obtaining (102) a plurality of first measurements (50) each representing a measured position in a three-dimensional frame of reference (x1, y1, z1) of a point situated on a structure to be monitored; - determining (103) a system of cylindrical coordinates (r, θ, z) having a vertical axis (19) parallel to an axis of said three-dimensional frame of reference, said vertical axis (19) having a position that minimizes a dispersion of said first measurements (50) with respect to a radial coordinate (r) of said system of cylindrical coordinates; - specifying (104) first disjoint sectors (200) of the space in the system of cylindrical coordinates (r, θ, z), said first sectors (200) being defined by azimuth coordinate increments (δθ) and axial coordinate increments (δz) in said system of cylindrical coordinates (r, θ, z); - calculating (105) on the basis of the first measurements (50) a plurality of first discretization points (250), each first discretization point (250) representing a mean position of the structure to be monitored in one of said first sectors (200); - calculating (106) on the basis of the first discretization points (250) estimated positions (300) of N vertical edges (13) of the structure to be monitored; - calculating (107) on the basis of the estimated positions (300) of the N vertical edges (13) estimated positions (400) of N medians, each median being a median of the vertical panel (14) of the structure to be monitored; and - comparing (108) on the basis of the estimated positions of the N medians the distances between diametrically opposite medians and an acceptable range of diameters (D).

2. The geometry monitoring method (100) as claimed in claim 1, in which the dimensional attributes further include an acceptable range of radii (Q) and in which the geometry monitoring method (100) further comprises comparing (108) the distances between said estimated positions (400) of the N medians and said vertical axis (19) with said acceptable range of radii.

3. The geometry monitoring method (100) as claimed in either one of claims 1 or 2, in which the dimensional attributes further include an acceptable range of panel widths (W) and in which the geometry monitoring method (100) further comprises comparing (108) the distances between the estimated positions (300) of two adjacent vertical edges (13) with said acceptable range of panel widths.

4. The geometry monitoring method (100) as claimed in any one of claims 1 to 3, in which the dimensional attributes further include an acceptable range of edge non-verticality (NV) and in which the geometry monitoring method (100) further comprises comparing (108) distances perpendicular to the vertical axis (19) of said system of cylindrical coordinates between end points of each vertical edge (13) with said acceptable range of edge non-verticality (NV).

5. The geometry monitoring method (100) as claimed in any one of claims 1 to 4, in which the dimensional attributes further include an acceptable range of ovalization (OV) and in which the geometry monitoring method (100) further comprises comparing (108) a diameter difference between an inscribed larger circle (CI) of the vertical wall (12) and a circumscribed smaller circle (CC) of the vertical wall (12) of the structure to be monitored with said acceptable range of ovalization.

6. The geometry monitoring method (100) as claimed in any one of claims 1 to 5, in which the dimensional attributes further include an acceptable range of local deformations of the vertical panels (14) and in which the geometry monitoring method (100) further comprises comparing local deformations of the structure to be monitored with said acceptable range of local deformations of the vertical panels (14).

7. The geometry monitoring method (100) as claimed in any one of claims 1 to 6, in which the dimensional attributes further include an acceptable range of non-verticality of the vertical panels (14) and in which the geometry monitoring method (100) further comprises: - selecting a subset of the first discretization points (250), in which subset the azimuth coordinates (Θ) of the first discretization points (250) are equal to one another; - determining the differences (ΔR) of radial coordinates (R) between the first discretization points (250) of the subsets; and - comparing the differences (ΔR) of radial coordinates (R) with said acceptable range of non-verticality of the vertical panels (14).

8. The geometry monitoring method (100) as claimed in any one of claims 1 to 7, in which the dimensional attributes further include an acceptable range of non-flatness of the vertical panels (14) and the geometry monitoring method (100) further includes: - on the basis of the estimated positions (300) of the N vertical edges (13), selecting the first measurements (50) corresponding to a vertical panel (14) of the structure to be monitored; - calculating a vertical panel non-flatness parameter (Δk) on the basis of the selected first measurements (50) and comparing the vertical panel non-flatness parameter (Δk) with said acceptable range of non-flatness of the vertical panels (14).

9. The geometry monitoring method (100) as claimed in any one of claims 1 to 8, in which the geometric attributes further include a plane bottom wall and the dimensional attributes further include an acceptable range of non-flatness of the bottom wall, and the geometry monitoring method (100) further includes: - obtaining (102) a plurality of second measurements (51) each representing a position measured in said three-dimensional frame of reference (x1, y1, z1) of a point situated on a bottom wall (15) of the structure to be monitored; - associating (103) with the second measurements (51) a system of cartesian coordinates (x, y, z), said system of cartesian coordinates having a first axis, a second axis and a third axis that are mutually perpendicular, the third axis coinciding with said vertical axis (19) of said system of cylindrical coordinates; - specifying (104) disjoint second sectors (201) of the space in said system of cartesian coordinates, said second sectors (201) being defined by increments (δx, δy) of coordinates along the first axis and the second axis of said system of cartesian coordinates (x, y, z); - calculating (105) on the basis of the second measurements (51) a plurality of second discretization points (251), each second discretization point (251) representing a mean position of the bottom wall (15) of the structure to be monitored in one of said second sectors (201); - calculating (108) a parameter of non-flatness of the bottom wall (Δz; Δd) on the basis of the second discretization points (251) and comparing (108) the parameter of non-flatness of the bottom wall (Δz; Δd) with said acceptable range of non-flatness of the bottom wall, and optionally in which the dimensional attributes further include an acceptable range of local deformations of the bottom wall and in which the geometry monitoring method (100) further comprises comparing local deformations of the bottom wall of the structure to be monitored with said acceptable range of local deformations of the bottom wall.

10. The geometry monitoring method (100) as claimed in any one of claims 1 to 9, in which N is an even number and N is more particularly an even number between 8 and 56, even more particularly N = 8 or N = 56.

11. The geometry monitoring method (100) as claimed in any one of claims 1 to 10, further comprising a generation step consisting in generating (109) a visual representation of a result of at least one of said comparisons (108) and / or further including the step of assigning a qualification of conformity to the structure to be monitored, the qualification of conformity being chosen between conform and non-conform.

12. The geometry monitoring method (100) as claimed in any one of claims 1 to 11, in which the plurality of first measurements (50) comes from an acquisition of measurements (1001) by means of a measuring instrument in the structure to be monitored and in which the measuring instrument includes a laser detection and ranging device (91) disposed in an internal space (11) of the structure to be monitored.

13. The geometry monitoring method (100) as claimed in any one of claims 1 to 12, in which the structure to be monitored is a supporting structure for a liquefied gas storage installation (1) and / or in which the structure to be monitored is made of concrete.

14. A computer program comprising instructions which, when the computer program is implemented by a computer, cause the computer to execute the geometry monitoring method (100) as claimed in any one of claims 1 to 13.

15. A geometry monitoring device (3000) comprising at least one processor (3001) and at least one memory (3002) containing a computer program, in which said at least one memory (3002) and the computer program are configured, with said at least one processor (3001), to cause execution of the geometry monitoring method (100) as claimed in any one of claims 1 to 13 by the geometry monitoring device (3000).