Tile leveling method applicable to irregular floors

Abstract

The present invention discloses a tile leveling method applicable to special-shaped floors, specifically relating to the technical field of floor paving construction, including: obtaining ground curvature and slope data through on-site measurement, and selecting no less than ten spatial control points to form a construction parameter set; generating a leveling reference path based on the comprehensive data as the height comparison baseline for subsequent laying; collecting the pre-laid elevation of each tile and comparing it with the target elevation of the reference path to form a height difference data sequence; outputting the leveling adjustment amount of each tile and controlling the slope continuity; when local height difference anomalies are found, interpolation calculation is performed based on adjacent control points to obtain correction values, and adjustments are only made within a local range to maintain the overall leveling continuity; through constructing a leveling reference path, generating a height difference data sequence and introducing priority determination, the present invention realizes precise height control, real-time deviation identification and local efficient correction during tile laying, improves construction accuracy and continuity, reduces the rework rate, and meets the complex construction requirements of special-shaped floors.

Description

Technical Field

[0001] The present invention relates to the technical field of floor paving construction, and more specifically, to a tile leveling method applicable to irregular floors. Background Art

[0002] At present, tile leveling construction methods are generally designed and applied based on standardized flat floor scenarios. Typical methods mainly include using a laser level for auxiliary leveling, manually pre-laying a leveling layer (such as cement mortar, tile adhesive, etc.), and using conventional linear tools (such as a spirit level, steel straightedge) for auxiliary positioning. These methods can basically meet the leveling requirements in building scenarios with simple structures and flat floors, and the construction technology is relatively mature. However, in irregular floor environments with continuously changing curvature characteristics, such as the curved ramps in urban ancient building protection areas, traditional leveling methods expose serious inapplicability, mainly reflected in the following aspects:

[0003] Precise curvature leveling cannot be achieved. Since a curved ramp is a typical spatially continuous surface, the ground elevation and curvature change dynamically with position. Traditional methods based on the plane assumption cannot generate a leveling reference line consistent with the actual curvature trend. The applicability of conventional leveling methods on such non-linear and non-rigid bases is extremely poor, easily causing local mutations, linear disorder, and uneven tile joints in tile laying, thus affecting the structural continuity and overall visual aesthetics.

[0004] There are natural defects in the adaptability of leveling materials. Common low-consistency materials (such as mortar, self-leveling glue) are prone to accumulation and offset due to gravity flow on the slope structure, resulting in uneven leveling layer thickness, sliding and hollowing, etc.; while high-consistency materials are difficult to fit complex curvatures due to poor ductility, and the construction effect depends on a large amount of manual experience, with uncontrollable errors and poor repeatability. Especially in cultural relic protection scenarios, it often leads to structural disturbances or destructive rework, lacking reliability.

[0005] The efficiency and cost of local damage repair are difficult to control. Since the irregular laying structure is usually non-linear and asymmetric, and the tile arrangement is non-uniform, conventional replacement methods need to damage multiple adjacent components to repair one damaged tile, which is not only inefficient but also extremely easy to damage the overall pattern arrangement, paving rhythm, and original process traces.

[0006] The existing technology is difficult to meet the special requirements of cultural relic protection scenarios. Traditional leveling processes lack the ability to model the curvature characteristics of the original appearance of ancient buildings and do not consider basic principles of cultural relic restoration such as "low invasiveness, reversibility, and local reparability". In actual operation, structural damages such as tile surface cracking, bonding layer disturbance, and base layer exposure are often caused by uncontrollable operations, greatly increasing the risks and costs of cultural heritage protection. Therefore, the present invention proposes a tile leveling method applicable to irregular floors in order to solve the above problems. Summary of the Invention

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A tile leveling method applicable to irregular ground, comprising the following steps:

[0009] Through on-site measurement, obtain the curvature change value and slope distribution data of the target ground within a continuous space range, and select no less than ten representative space control points within the area to be paved, and measure their elevation values one by one to form an initial construction parameter set;

[0010] Comprehensively process the obtained curvature change value, slope distribution data and elevation values of the space control points to generate a leveling reference path arranged in a space sequence. This leveling reference path is a positioning reference curve reflecting the original terrain undulation characteristics and is used as a unified baseline for positioning and height comparison during subsequent manual paving;

[0011] According to this leveling reference path, during actual tile laying, collect the elevation of the pre-laying state of each tile to be laid at the current construction point, and compare its numerical value with the target elevation on the corresponding leveling reference path, calculate the difference between the two, and form a height difference data sequence reflecting the leveling deviation;

[0012] According to the height difference data sequence, output the leveling adjustment amount required for each tile according to the construction sequence, and control the slope continuity during the leveling adjustment process by continuously comparing the change trend of the adjustment values of adjacent tiles to ensure that the overall laying surface forms a unified and non-jumping elevation transition relationship;

[0013] When there are single-point height differences or mutation values in a local area, immediately start the independent adjustment process for this area, calculate the correction value through the interpolation algorithm of adjacent front and rear control points, and limit the adjustment range not to extend beyond the adjacent area to ensure local leveling correction operation without destroying the logical continuity of the overall leveling;

[0014] In a preferred embodiment, the process of obtaining the curvature change value and slope distribution data of the target ground within a continuous space range includes two consecutive measurement steps:

[0015] In the first step, establish multiple space measurement lines along the vertical and horizontal directions respectively within the predetermined laying area through a handheld measurement tool or surveying and mapping equipment to form an intersecting measurement grid line. In the second step, obtain the three-dimensional space coordinate values of each intersection point, and perform curvature fitting and slope calculation on all measured points. The curvature fitting can adopt any one of the quadratic curve fitting method, hyperboloid fitting method or polynomial regression method;

[0016] The slope is calculated numerically based on the ratio of elevation change to horizontal distance, and its average change value is taken as the slope parameter. The curvature change value and the slope distribution data serve as the data support sources for subsequent path generation and comparison benchmarks.

[0017] In a preferred embodiment, the process of selecting no less than ten representative spatial control points includes the following two steps:

[0018] In the first step, the area to be paved is subjected to a standard rectangular projection according to its actual boundary dimensions. Based on the projection range, it is evenly divided into multiple regular grid units at intervals not greater than two meters along the horizontal and vertical directions, and a spatial control point is initially arranged at the center point of each grid unit as a basic control point.

[0019] In the second step, after the initial layout is completed, whether the elevation difference between the control points between three adjacent grid units exceeds the threshold of five millimeters is used as the judgment basis. If there are more than two consecutive positions that meet this difference condition, it is determined that the terrain undulation in this area is severe, and an additional control point is added at each edge position within the adjacent edge grids of this area to form a high-density control point area.

[0020] The control point set formed through these two layout steps constitutes a gradually changing distribution structure of "low density in the center and high density at the edges" in space, and the distribution of grid center points maintains matrix uniformity. The minimum distance between all control points is not less than 0.5 meters, and the maximum distance does not exceed two meters, ensuring the formation of an elevation point set that is both evenly covered and has a gradually changing characteristic in space, providing the basic data support for subsequent leveling path fitting and error control.

[0021] In a preferred embodiment, the generated leveling reference path is a curve in space. The position of the curve is determined after fitting the spatial control points. The generation of the path point sequence follows the sequential construction method from the starting control point to the ending control point. In the path generation process, a forward point-by-point interpolation algorithm is used, and no less than three intermediate transition points are inserted between every two control points to ensure the continuity and smoothness of the path curvature within each micro-segment.

[0022] The change in the distance between the path point and its corresponding control point is limited to no more than 10%. At the same time, the target elevation of each path point is obtained by calculating the weighted average of the elevations of its adjacent control points. The weighting method calculates the weight factor based on the spatial distance between the point and the control point and normalizes it. This process ensures the fitting degree of the path point in the spatial surface, thereby improving the visual predictability and actual executability during the manual leveling operation.

[0023] In a preferred embodiment, during the process of numerically comparing the elevation of the tiles in their initial laid state with the target elevation on the leveling reference path, the current elevation of the tiles is collected by digital ranging, and the height difference value is calculated by combining the target elevation values at the corresponding positions of the path points. This height difference is denoted as the actual height difference. When the actual height difference exceeds the acceptable error range, this point is recorded as a key adjustment point. At the same time, the height difference between the current tile and the previous tile is recorded, and the section with three or more consecutive key adjustment points is counted and marked as a high-risk area for leveling. In subsequent manual operations, the leveling adjustment is preferentially performed on the high-risk area.

[0024] In a preferred embodiment, after the height difference data sequence reflecting the leveling deviation is formed, before outputting the leveling adjustment amount, a global trend judgment process is first performed on this data sequence. The judgment criterion is whether there is a trend of continuous increase or continuous decrease on one side in the spatial sequence of the height difference values. If the continuous change exceeds five tile positions, it is considered that there may be an abnormal path point or the accumulation of manual laying errors, and the error backtracking logic is entered.

[0025] The elevation of the previous section of path points is inversely calculated according to the adjustment direction of adjacent tiles, and a corrected path point is attempted to be generated. The generation of the corrected path point is based on the trend of the original path point as the initial vector, and the local curve segment is re-fitted by combining the elevation of the control points near the error point to smoothly replace the original path segment, improving the global leveling coherence and preventing the phenomenon of slope jumping caused by error accumulation.

[0026] In a preferred embodiment, the method of controlling the slope continuity during the leveling adjustment process includes two parts: height difference smoothing processing and multi-segment linear consistency detection. Height difference smoothing processing means taking the difference between the leveling adjustment amount of each tile and the adjustment amount of the adjacent tile, calculating the difference sequence, and marking the points where the difference exceeds the fixed threshold as warning nodes during the construction process.

[0027] Multi-segment linear consistency detection is to perform a linear regression analysis on the leveling adjustment amounts of five consecutive tiles and compare the included angle with the tangent direction of the overall fitting curve in this section. If the included angle exceeds the preset value, it is determined that there is a risk of linear mutation in this section, and the construction personnel need to review whether the actual operation trajectory deviates from the reference path to improve the overall visual coherence and structural smoothness after laying.

[0028] In a preferred embodiment, when there is a single-point height difference abnormality or mutation value in a local area, the independent adjustment process includes three steps:

[0029] First, starting from the mutation point, extend two tile positions forward and backward along the sequence of path points, and collect the elevation values of five points for local analysis. Secondly, use any one of the linear interpolation method, cubic spline interpolation method, or weighted average method to generate a local correction curve segment, and calculate the new elevation value to which the mutation point should be adjusted.

[0030] Finally, calculate the difference between the new elevation value and the actual elevation of the current tile, output the local height difference adjustment value, and set the leveling operation to be only carried out within this five-point interval, without affecting the laying benchmarks of the completed areas before and after it, ensuring that the correction operation is limited to the necessary range and avoiding the destruction of the overall continuity.

[0031] In a preferred embodiment, a priority determination rule is introduced during the process of outputting the height difference adjustment plan for the local range, and the determination rule is established based on the following criteria:

[0032] One is whether the absolute value of the height difference at the mutation point exceeds the set upper limit, the second is whether the change rate of the height difference between the mutation point and the adjacent point is the maximum value in this interval, and the third is whether the point is located in the edge area or the structural turning area. If any two of them are satisfied, set this point as a high-priority correction point and list it as the primary operation object in the height difference adjustment plan.

[0033] The technical effects and advantages of the present invention:

[0034] By constructing a leveling reference path and generating a high-density spatial path point in the early stage of laying, the present invention enables clear reference for height control during the tile laying process, avoiding the problem of cumulative leveling errors caused by traditional reliance on manual experience. Especially under complex terrain conditions such as special-shaped floors like arc ramps, inclined platforms, and non-uniform foundations, the curve fitted by the spatial control points in the present invention can truly reflect the terrain continuity. With the setting of the target elevation of the path points, precise height positioning of each tile can be achieved. Compared with the "visual leveling" or "using hard templates for positioning" methods in the prior art, the present invention provides a data-driven leveling guidance logic, improving construction consistency and control accuracy, and effectively reducing the mislaying rate and rework cost.

[0035] The present invention realizes real-time monitoring and process control of leveling deviation by dynamically generating height difference data sequences and automatically identifying high-risk areas, greatly enhancing the risk identification ability and local processing efficiency during manual construction. By comparing the actual laying elevation of each tile with the target elevation of the corresponding path point, constructing the error as a spatial sequence, and combining the continuity judgment and mutation detection logic, it can timely detect height anomalies caused by uneven materials, manual deviation, or unreasonable path settings. The present invention can not only locate the height difference mutation points, but also timely trace back path anomalies through height difference trend analysis to achieve early intervention. Compared with the traditional post-facto repair method, the present invention provides a technical path of "identifying during construction + real-time correction", preventing error accumulation from causing structural slope jumps or uneven paving, and ensuring the aesthetics and safety of the structure.

[0036] The present invention realizes a precise repair strategy under limited resources by setting priority judgment rules and restricting the local scope of adjustment operations, enabling manual leveling to have higher operation efficiency and structural stability; when local mutation points are identified, the present invention introduces three-dimensional space judgment criteria (height difference value, change rate, spatial position) to grade and judge the processing priority, clarify which points need to be corrected first, and combine local re-fitting of the correction curve to only perform fine-tuning of the leveling of tiles within a small range to avoid overall disturbance. This approach enables rapid correction through a minimum intervention strategy even in the presence of sudden deviations under complex ground conditions, and is particularly suitable for scenarios with tight construction periods, limited resources, or partially completed laying paths, significantly improving construction efficiency and continuity control ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the accompanying drawings;

[0038] Figure 1 It is the schematic diagram of the tile leveling method for special-shaped floors in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0040] Refer to Figure 1 The following embodiments are obtained:

[0041] Embodiment 1: In view of the difficult problem of laying tiles on irregular ground, the present invention constructs a leveling control method based on spatial geometric measurement and digital logic judgment. First, through on-site multi-point measurement, the curvature change and slope distribution information in the laying area are comprehensively collected, and on this basis, an elevation point set with a gradually changing density is established to form a basic data set containing the terrain undulation characteristics. By performing spatial fitting processing on these control points, a continuous and reference leveling reference path is generated, which is used as the basis for unified height comparison in subsequent manual laying.

[0042] During the laying process, the pre-laying state elevation of each tile is collected in real time, and its numerical comparison is made with the target elevation of the corresponding path point to calculate the height difference data sequence. Based on this sequence, the adjustment amount of each tile is further output, and the adjustment trend of adjacent tiles is analyzed to determine whether it meets the requirement of elevation continuity. When sudden changes or anomalies are found in a local area, a local correction curve will be generated according to the front and rear control points to limit the correction range and avoid damage to the overall path logic.

[0043] The overall method realizes the dynamic adjustment and local fine repair of the leveling error during the laying process of complex curved slopes through a combination of data-driven and rule control, ensuring an effective balance between visual consistency and structural continuity during the laying process, and at the same time having strong versatility and manual executability.

[0044] Specifically, it includes the following steps:

[0045] Through on-site measurement, obtain the curvature change value and slope distribution data of the target ground within a continuous spatial range, and select no less than ten representative spatial control points in the required laying area, and measure their elevation values one by one to form an initial construction parameter set; The significance of the first step is to construct a comprehensive and reliable terrain data foundation. Through on-site actual measurement, collect the curvature change value and slope distribution data of the laying area within a continuous spatial range, and set no less than ten representative spatial control points, and record the elevation of each point one by one to form an initial construction parameter set reflecting the true undulation state of the terrain. This process ensures that all subsequent judgment and adjustment operations are based on accurate terrain perception, avoiding error accumulation caused by experience or visual estimation, and laying a quantitative basis for the scientific fitting of the leveling path.

[0046] The obtained curvature change values, slope distribution data, and elevation values of the spatial control points are comprehensively processed to generate a leveling reference path arranged in a spatial sequence. This leveling reference path is a positioning reference curve reflecting the undulation characteristics of the original terrain and serves as a unified baseline for positioning and height comparison during subsequent manual laying. The significance of the second step lies in forming a unified leveling reference through data modeling means. By comprehensively processing the curvature, slope, and elevation data obtained in the previous step and fitting a leveling reference path according to the principle of spatial continuity. This path is not a simple linear connection but a spatial reference curve reflecting the undulation characteristics of the original ground through mathematical processing of the terrain change trend. Its role is to provide a unified and highly perceptible laying reference for construction workers, enabling the laying of each tile to no longer rely on subjective judgment but to use the target elevation at the path points as an objective reference, thus realizing standardized construction guidance.

[0047] Based on this leveling reference path, during the actual tile laying, the elevation of the pre-laying state of each tile to be laid at the current construction point is collected, and its numerical comparison is made with the target elevation on the corresponding leveling reference path. The difference between the two is calculated to form a height difference data sequence reflecting the leveling deviation. The significance of the third step lies in realizing dynamic feedback and real-time comparison during construction. Before laying each tile, the elevation in its actual pre-laying state is collected on-site and numerically compared point by point with the target elevation at the corresponding position in the leveling reference path. The height difference between the two is calculated. This height difference reflects the deviation degree between the laying site and the theoretical reference, forming a complete height difference data sequence, which serves as an important basis for identifying leveling deviation and construction accuracy. Through this step, the transformation of leveling control from static guidance to dynamic verification is realized, enhancing the real-time performance and adjustment ability of the construction process.

[0048] According to the height difference data sequence, the leveling adjustment amount required for each tile is output in the construction order, and by continuously comparing the change trend of the adjustment values of adjacent tiles, the slope continuity during the leveling adjustment process is controlled to ensure that a unified and seamless elevation transition relationship is formed on the overall laying surface. The significance of the fourth step lies in ensuring the overall continuity of the slope during the laying process. By sequentially reading the height difference data sequence, the leveling adjustment amount for each tile is output one by one in the construction order, and combined with the change trend of the adjustment values between the front and rear tiles, it is judged whether there are unreasonable height mutations or fault phenomena. If a significant jump is found in the adjustment amount of adjacent tiles, local leveling anomalies can be predicted and intervened in advance. This process not only realizes the control of individual tiles but also establishes a set of control logics for the smoothness of the entire paving surface, ensuring that the construction results are physically coherent and visually consistent.

[0049] When there is an abnormal or sudden change in the height difference of a single point in a local area, the independent adjustment process of the area is immediately started, and the correction value is calculated by the interpolation algorithm of the adjacent control points before and after, and the adjustment range is limited to not extend beyond the adjacent area, so as to ensure that the local leveling correction operation is achieved without destroying the continuity of the overall leveling logic. The significance of the fifth step is to deal with the inevitable local errors or abnormal mutations in construction. When an obvious single-point height difference is found in a specific area, by activating the local independent adjustment mechanism, the data of the front and rear control points are called for interpolation operations to generate a new elevation correction value for the point, and the adjustment operation is limited to only act on the abnormal point and its minimum adjacent area, without affecting the overall path structure. The core value of this mechanism lies in its fast and controllable deviation correction capability, which can not only achieve local correction without destroying the continuity of the overall leveling logic, but also prevent large-scale redoing due to minor errors, greatly improving construction efficiency and repair reliability.

[0050] The process of obtaining the curvature change value and slope distribution data of the target ground in a continuous spatial range includes two continuous measurement steps:

[0051] The first step is to establish multiple spatial measurement lines in the vertical and horizontal directions in the predetermined paving area by handheld measurement tools or surveying equipment to form interlaced measurement grid lines. The second step is to obtain the three-dimensional spatial coordinate value of each intersection point, and perform curvature fitting and slope calculation on all measured points. The curvature fitting adopts any one of the quadratic curve fitting method, hyperbolic surface fitting method or polynomial regression method.

[0052] The slope is calculated numerically based on the ratio of elevation change to horizontal distance, and the average change value is taken as the slope parameter. The curvature change value and slope distribution data are used as data support sources for subsequent path generation and comparison benchmarks.

[0053] In the first step, construction workers can use handheld laser rangefinders, 3D laser scanners or total stations and other tools with spatial coordinate measurement capabilities to perform wiring measurements on the scheduled paving area. The specific operation is: first, project the paving area onto a horizontal reference plane, set vertical and horizontal measurement lines along the plane, and set the distance between the two to no more than two meters and one meter respectively; the measurement lines are laid out in an orthogonal manner to form a regular measurement grid. At each intersection of the measurement grid, spatial measurement points are collected, and the three-dimensional coordinate data of each point is recorded, including the horizontal coordinate, vertical coordinate and elevation value, and the same reference plane is used for elevation measurement to prevent data deviation.

[0054] In the second step, the collected spatial points need to be subjected to curvature fitting and slope estimation. The curvature fitting process selects one of the following three mathematical models for processing, including:

[0055] Quadratic curve fitting method: applicable to the situation where the local area changes gently. This method regards the elevation of control points in a certain direction as variables and solves the fitting function in the following form by the least squares method:

[0056] The vertical height is equal to the constant term plus the linear term plus the quadratic term, that is, the height is equal to the constant coefficient plus the coefficient of the first-order term multiplied by the horizontal coordinate, plus the coefficient of the quadratic term multiplied by the square of the horizontal coordinate.

[0057] Hyperbolic surface fitting method: applicable to the situation where there is a spatial surface transition region. This method establishes a two-dimensional function in the following form: the height is equal to a combined function of a set of coefficients and the transverse coordinate, the longitudinal coordinate, their product terms, and square terms. For example: the height is equal to (constant) + (transverse coordinate coefficient) × transverse coordinate + (longitudinal coordinate coefficient) × longitudinal coordinate + (product coefficient) × product of transverse and longitudinal coordinates + (square term coefficient) × square of transverse coordinate + (square term coefficient) × square of longitudinal coordinate. Determine all coefficients through the principle of minimum error and evaluate the root mean square of the residuals to judge the quality of the surface fitting.

[0058] Polynomial regression method: applicable to the situation where the overall change in the entire area is relatively complex. This method constructs a higher-order polynomial function form, takes the coordinates of multiple control points as inputs, and the independent variables include the first-order, second-order, cross terms, cubic terms, etc., and solves multiple coefficients through least squares regression to finally form a global function for representing the ground form.

[0059] The selection criteria for the above three fitting methods can be determined by the actual terrain complexity at the construction site or can be optimized through the fitting residual evaluation index. When the residual index such as the average fitting error is less than 5 mm and the standard deviation of the residuals is less than 2 mm, it is considered to meet the construction requirements.

[0060] After the above point set is completed, the process of slope calculation is carried out by calculating the ratio of the elevation difference between adjacent control points to the horizontal distance point by point. Let the numbers of two adjacent control points be the front point and the rear point, and their corresponding elevations are denoted as the front elevation and the rear elevation respectively, and the corresponding horizontal projection distance is the Euclidean distance of the two points in the horizontal and vertical coordinates. Then the slope value is equal to the elevation difference divided by the horizontal projection distance, and the calculation formula is:

[0061] First, select two adjacent control points in space, one as the front point and one as the rear point, and record the elevation values of these two points, that is, their vertical heights relative to the unified reference plane. Then, determine the position differences in the lateral direction and the longitudinal direction of these two points respectively. Next, square the lateral difference, square the longitudinal difference as well, add these two squared values, and then calculate the square root to obtain the straight-line distance between these two points on the horizontal projection plane. Finally, take the elevation difference between these two points, that is, the elevation of the rear point minus the elevation of the front point, as the vertical height change value, and divide this vertical height change value by the horizontal distance calculated previously. The resulting value is the slope value between these two points.

[0062] This calculation is performed between all pairs of adjacent points, and the arithmetic mean of the results is taken as the average slope parameter for the entire area to be paved. To ensure the representativeness of this parameter, the maximum slope value and the minimum slope value can be calculated simultaneously and their fluctuation range can be evaluated to determine whether the paving area needs to be further segmented.

[0063] The finally obtained curvature change value and slope distribution data are used both to generate the surface function in the leveling reference path and as an important reference standard for comparing the target elevation during the laying of each subsequent tile. This basic data set must have high consistency and coverage to support real-time deviation calculation and dynamic adjustment control during the entire leveling process.

[0064] For example, in a special-shaped slope area with a side length of ten meters, if grid lines with a spacing of one meter are laid out, one hundred intersection measurement points can be formed. The overall curvature model is calculated through hyperbolic surface fitting, and the average slope value is calculated to be 3.2% through slope calculation. This result will be used in the generation of all leveling paths and the height difference judgment process for subsequent tile laying to ensure that manual construction is carried out under precise control.

[0065] The process of selecting no less than ten representative spatial control points includes the following two steps:

[0066] In the first step, project the area to be paved onto a standard rectangle according to its actual boundary dimensions, and evenly divide it into multiple regular grid units along the horizontal and vertical directions at intervals not greater than two meters. Then, preliminarily arrange a spatial control point at the center point of each grid unit as a basic control point.

[0067] In the second step, after this preliminary arrangement, based on whether the elevation difference between the control points in three adjacent grid units exceeds the threshold of 5 mm as the judgment basis, if there are more than two consecutive positions that meet this difference condition, it is determined that the terrain in this area has a drastic undulation change, and an additional control point is added at each edge position in the adjacent edge grids of this area to form a high-density control point area.

[0068] The set of control points formed by this two-step layout constitutes a gradient distribution structure of "low density in the center and high density at the edges" in space, and the distribution of the grid center points maintains matrix uniformity. The minimum distance between all control points is not less than 0.5 meters, and the maximum distance does not exceed 2 meters, ensuring the formation of an elevation point set that is both uniformly covered and has gradient characteristics in space, providing the basic data support for subsequent leveling path fitting and error control.

[0069] Specifically, in the first step, the area to be paved is projected into a standard rectangle according to its actual boundary dimensions, which means that an irregular ground area that may originally have arcs, slopes, or curved edges is uniformly projected into the outer frame of the smallest enclosing rectangle on a two-dimensional plane. This rectangle serves as the boundary basis for the layout of control points. Within this boundary range, grid division is carried out separately in the horizontal and vertical directions. When dividing, it is uniformly separated according to a fixed interval not greater than 2 meters, forming multiple regular grid units. For example, if the paved area is 12 meters long and 8 meters wide, it can be divided into six segments in the horizontal direction and four segments in the vertical direction, forming a total of 24 grid units. A spatial control point is arranged at the center point of each grid unit, and this control point records its spatial position and elevation information, serving as the source of the basic points for subsequent leveling path fitting. Since each control point is evenly distributed at the grid center and can cover the overall shape inside the area, it is a basic uniform sampling strategy.

[0070] In the second step, after this preliminary layout is completed, in order to further improve the control accuracy of the terrain change area, a density enhancement mechanism is introduced. The judgment condition of this mechanism is: among the control points of three consecutive adjacent grid units, if the elevation difference between its adjacent control points is greater than 5 mm, and such differences are satisfied at two or more consecutive positions, it indicates that the elevation change in this area is not an accidental fluctuation but has an obvious undulating trend. At this time, this area is identified as "a severely undulating terrain change area". In order to increase the sampling density of control points in such areas to improve the fitting accuracy and error response ability, an additional control point will be added at each edge position of its adjacent edge grid, for example, added at the middle or corner of the four boundaries respectively, forming a high-density control point distribution structure in this area.

[0071] Through the above two-step layout, a gradient control point set of "low density in the center and high density at the edges" is finally formed in space. The advantage of this layout method lies in the reasonable allocation of resources: maintaining the basic sampling density for the terrain-stable area and increasing the sampling density for the area with severe changes, taking into account both calculation efficiency and control accuracy. At the same time, since the basic grid center points are regularly divided and have matrix uniformity in spatial distribution, it avoids the over-concentration or dispersion of control points, ensuring that there will be no error diffusion caused by unreasonable point distribution during the path fitting process.

[0072] In addition, to ensure that construction workers can effectively obtain the measurement point information during actual operation, the minimum distance between any two points in this control point set shall not be less than 0.5 meters to avoid construction conflicts caused by excessive density. At the same time, the maximum distance shall not exceed 2 meters to ensure that the ability to describe the surface trend will not be lost due to sparse points. The finally formed point set is reasonably distributed and clearly structured in space, which not only provides data input for the fitting of the leveling path, but also establishes a stable elevation reference for subsequent height difference data comparison and local error correction.

[0073] For example, in a curved slope laying area of 9 meters by 6 meters, if the basic grid is set at 2 meters per grid, twelve basic units will be formed and twelve basic control points will be arranged. In a certain horizontal survey line, the elevation differences of three consecutive grid control points are 6 mm, 7 mm, and 5.5 mm respectively, meeting the judgment condition of drastic elevation change. Then, one control point is added to each of the four points at the edge position of this section, so that the total number of control points in this area reaches sixteen, realizing area densification. The entire laying area may finally contain more than twenty control points, forming a uniform and gradually changing elevation sampling network.

[0074] The generated leveling reference path is a curve in space. The position of this curve is determined after the fitting process of the space control points. The generation of the path point sequence follows the construction method of starting from the starting control point to the ending control point. During the path generation process, the forward point-by-point interpolation algorithm is used, and no less than three intermediate transition points are inserted between every two control points to ensure the continuity and smoothness of the path curvature in each micro-segment.

[0075] The change in the distance between the path point and its corresponding control point is limited to no more than 10%. At the same time, the target elevation of each path point is obtained by calculating the weighted average of the elevations of its adjacent control points. This weighting method calculates the weight factor based on the spatial distance between the point and the control point and performs normalization processing. This process ensures the fitting degree of the path point in the space surface, thereby improving the visual predictability and actual executability during the manual leveling operation.

[0076] Specifically, during the fitting process, all the selected space control points are numbered in the order of their distribution in the construction area first, and the starting control point and the ending control point are used as the boundary starting point and ending point of the path. The generation of the path point sequence follows the construction method of starting from the starting control point to the ending control point, and the spatial positions of the path points are determined one by one along this direction. The fitting method uses the segmented interpolation method to avoid excessive local error accumulation caused by overall fitting.

[0077] During the path generation process, a forward point-by-point interpolation algorithm is adopted. This algorithm means that after the positions and corresponding elevations of two endpoints (i.e., adjacent control points) in a certain section are known, starting from the first point, multiple intermediate points are inserted between it and the second point at a fixed interval or ratio. During the interpolation process, at least no less than three intermediate transition points are inserted between every two control points, distributed at equal intervals or in proportion, so that the overall path that might originally have only a dozen or so control points is refined into a sequence of multiple high-density continuous path points, ensuring the continuity and smoothness of the path curvature within each micro-segment. This refined curve can more accurately fit the actual terrain surface, avoiding skipping segments or sudden changes, which is conducive to manual visual inspection and the operation of leveling tools.

[0078] The change in the distance between a path point and its corresponding control point is limited to no more than ten percent. This means that during the interpolation process, the horizontal projection distance from any generated path point to its upstream or downstream control point shall not be greater than ten percent of the control point spacing. For example, if the horizontal distance between two control points is two meters, the deviation of the intermediate path point from any control point shall not exceed twenty centimeters. This rule is used to constrain the path points from deviating from their original control structure, ensuring the geometric continuity and spatial stability of the generated curve.

[0079] The target elevation of each path point is obtained by calculating the weighted average of the elevations of its adjacent control points. The specific operation is as follows: Using the spatial distance from the path point to each adjacent control point as a reference, a weight factor is constructed. The setting of the weight factor follows the principle that the closer the distance, the higher the weight. For example, if the distance from a path point to the previous control point is one meter and the distance to the next control point is two meters, then unnormalized weights of one-half and one-quarter are assigned to the front and back points respectively. Then, after summing up all the weights, each factor is normalized, that is, each weight is divided by the sum, so that the final weighted result is only determined by the elevations of the two control points and their relative distances. This weighted average method can effectively filter out the elevation discreteness error and improve the rationality and spatial smoothness of the path point height.

[0080] This process ensures the fitting degree of the path points in the spatial surface, that is, the path point heights are continuously close to the elevation trend surface formed by the control points, without drifting or reverse mutation, thereby improving the visual predictability and actual executability during the manual leveling operation. Construction workers can directly judge the laying horizontal reference of each tile based on the position of the path points, without relying on experience judgment, improving the construction consistency.

[0081] For example, within an area where the control point spacing is two meters, ten control points are laid out, dividing a total of nine path segments. Three intermediate path points are inserted into each segment, resulting in a total of thirty-six path points. If the distances of a certain path point from the left and right control points are 0.8 meters and 1.2 meters respectively, and the elevations of the control points are 0.5 meters and 0.6 meters, then the weighting factors are 1 divided by 0.8 and 1 divided by 1.2. After normalization, they are 60% and 40% respectively. The target elevation of this path point is 0.5 multiplied by 60% plus 0.6 multiplied by 40%, resulting in a path point target elevation of 0.54 meters. The height setting of all path points is completed in this way to ensure a continuous curve and adapt to the actual site changes.

[0082] The process of numerically comparing the elevation of the initial laying state of the tile with the target elevation on the leveling reference path. The current elevation of the tile is collected through digital distance measurement, and the height difference value is calculated by combining the target elevation value at the corresponding position of the path point. This height difference is recorded as the actual height difference. When the actual height difference exceeds the acceptable error range, this point is recorded as a key adjustment point. At the same time, the height difference between the current tile and the previous tile is recorded, and the section with three or more consecutive key adjustment points is counted and marked as a leveling high-risk area. In subsequent manual operations, the leveling adjustment is preferentially performed on the high-risk area.

[0083] The specific operation steps are as follows: First, after the tiles are initially placed, use digital distance measurement to measure the surface of this tile. Common tools include laser rangefinders, digital elevation measuring rods, etc. This measurement is based on a unified reference plane, such as taking the ground starting elevation or the design elevation zero point as the zero position. The measured result is the actual laying elevation of the current tile. Subsequently, in the preset leveling reference path, according to the spatial position of this tile, find the corresponding path point and read the target elevation value of this path point. Subtract the actual elevation from the target elevation to obtain the height difference of this tile, which is recorded as the "actual height difference".

[0084] This actual height difference is used to reflect the deviation degree between the current laying state and the theoretical reference. If the absolute value of this height difference does not exceed the previously set error tolerance range, it is considered that the laying height is within the allowable range; if the absolute value of this actual height difference exceeds the acceptable error range, this point is recorded as a "key adjustment point". The acceptable error range is usually set between 3 millimeters and 5 millimeters, and the specific value can be determined by the on-site construction requirements. For example, in a high-precision scenario, it is set to 3 millimeters. That is, if the actual height difference of a certain tile is 6 millimeters, it is marked as a key point that needs to be corrected.

[0085] Meanwhile, after recording the height difference of this tile, it is also necessary to calculate the height difference between the current tile and the previous one, that is, the difference between their actual height differences. This value is used to identify whether the height change shows a sudden change trend. If the difference is too large, it indicates that there is a problem with the continuity of leveling. The construction recorder needs to synchronously digitalize and register this difference for each tile for continuous trend analysis.

[0086] Furthermore, during the construction process, if a section composed of three or more key adjustment points in a row is statistically analyzed, that is, the height differences of multiple adjacent tiles exceed the standard, it can be determined that there is a local overall leveling deviation in this area, and this section is marked as a "leveling high-risk area". This area may deviate from the theoretical benchmark as a whole due to insufficient control points, path point fitting errors, or preliminary laying method deviations. Identifying such high-risk areas has high technical value because it can be the object of priority allocation of construction resources, preventing the deviation from continuing to expand on the wrong benchmark in subsequent areas and causing structural jumps.

[0087] In subsequent manual operations, priority should be given to performing leveling adjustments on the identified leveling high-risk areas. Specific measures include: repositioning the laying starting point of the tiles in this area; using a more accurate leveling tool to compare the actual elevation with the path elevation again; if necessary, local reference points can be slightly adjusted according to the trends of the path points before and after this area to improve the overall laying smoothness.

[0088] For example, in a local path section, five tiles are laid in sequence. After the initial measurement, the actual height differences are 3 mm, 6 mm, 7 mm, 5.5 mm, and 2 mm respectively. Assuming the acceptable error is 5 mm, then the second to fourth tiles all exceed the tolerance, constituting three consecutive key adjustment points. Therefore, this section is automatically identified as a high-risk section, and the construction person in charge can immediately adjust the path or repair the initial laying method to avoid problems such as overall slope jumps or tile instability caused by continuous errors.

[0089] After the height difference data sequence reflecting the leveling deviation is formed, before outputting the leveling adjustment amount, a global trend judgment process is first performed on this data sequence. The judgment criterion is whether there is a trend of continuous increase or continuous decrease on one side in the spatial sequence of the height differences. If the continuous change exceeds five tile positions, it is considered that there may be path point anomalies or cumulative human laying errors, and the error backtracking logic is entered;

[0090] Reverse calculate the elevation of the previous path point according to the adjustment direction of adjacent tiles and try to generate corrected path points. The generation of corrected path points is based on the trend of the original path point as the initial vector, and combined with the elevation of the control points near the error points to re-fit the local curve section to smoothly replace the original path section, improving the global leveling coherence and preventing the slope jump phenomenon caused by error accumulation.

[0091] The core criterion for this global trend judgment process is as follows: Observe the changing trend of the height difference values in the spatial sequence. That is, in accordance with the tile laying order, arrange the height difference values between each tile and the leveling reference path in sequence according to their spatial positions to form a one-dimensional height difference sequence. If there is a trend of continuous increase or continuous decrease on one side in a certain continuous position of this height difference sequence, it indicates that there may be an accumulation of height offsets in this area. The specific judgment condition is: in at least five consecutive tile positions, if the height difference values continuously increase or continuously decrease without any reverse fluctuations or interruptions in the change, it is determined as "significant unilateral trend".

[0092] When such a trend appears, it is regarded as possibly having path point anomalies or an accumulation of manual laying errors. Path point anomalies refer to the situation where the originally fitted leveling reference path fails to accurately reflect the true terrain trend in this area, which may be caused by insufficient control point density or distorted interpolation points; manual errors may be due to deviations from the reference during the manual laying process, gradually accumulating to cause an overall offset.

[0093] At this time, the error backtracking logic is entered. The backtracking logic is as follows: Starting from the end point of the current abnormal trend segment, backtrack forward, and reverse calculate the elevation of the previous path point according to the adjustment direction of adjacent tiles to evaluate whether the elevation of the original path point is set too high or too low, and try to generate a new sequence of corrected path points. This process is not equivalent to overall refitting, but rather replaces local curves, with the characteristics of controlled scope and high correction accuracy.

[0094] The generation of corrected path points is carried out according to the following principles: Taking the trend of the original path point as the initial vector, determine the correction direction and approximate change rate; combining the elevation data of spatial control points near the error point, select two or three control points before and after the corresponding point, and refit a local curve segment. Fitting methods can include linear interpolation, polynomial fitting, or least squares curve adjustment, etc. Select an appropriate model according to the number of control points and the error curvature judgment. The newly generated corrected path points will replace the path points at the same section position in the original path and be used for the calculation of subsequent adjustment amounts.

[0095] The purpose of this operation is to smoothly replace the original path segment with an abnormal trend, improve the global leveling continuity, and prevent the phenomenon of stepped slopes caused by error accumulation. The stepped slope phenomenon refers to an obvious mutation or sudden jump in the height of the tile laying surface in a certain area, which seriously affects the structural continuity and visual flatness.

[0096] For example, in a certain section, starting from the eleventh tile to the fifteenth tile, the height differences from the path are successively 1 mm, 2 mm, 3.5 mm, 5 mm, and 6.5 mm, showing a continuously increasing and unbroken trend, meeting the condition of continuous unilateral change of five tiles. The system marks this section as an abnormal section, traces back to the ninth tile, and refits this section of the path by combining the elevations of the eighth, ninth, and thirteenth control points, and regenerates the target elevation values of the tenth to fifteenth path points. The new path obtained by this adjustment will be used for the next round of leveling adjustment amount calculation to correct the error orientation that the original path may bring.

[0097] Through this processing mechanism, structural problems can be actively identified and repaired in a timely manner before the leveling process is executed, avoiding relying on repeated manual rework in the later stage, and greatly improving the laying quality and continuity.

[0098] The methods for controlling the slope continuity in the leveling adjustment process include two parts: smooth processing of height difference values and detection of multi-segment linear consistency. Smooth processing of height difference values means taking the difference between the leveling adjustment amount of each tile and the adjustment amount of the adjacent tile, calculating the difference sequence, and marking the points where the difference exceeds a fixed threshold as warning nodes during the construction process;

[0099] Detection of multi-segment linear consistency is to perform linear regression analysis on the leveling adjustment amounts of five consecutive tiles, and compare the included angle with the tangent direction of the overall fitting curve in this section. If the included angle exceeds a preset value, such as 15 degrees, it is determined that there is a risk of linear mutation in this section, and the construction personnel need to review whether the actual operation trajectory deviates from the reference path and make manual fine-tuning if necessary to improve the overall visual coherence and structural smoothness after laying.

[0100] Smooth processing of height difference values refers to performing a difference operation between the leveling adjustment amount required for each tile and the leveling adjustment amount of the previous adjacent tile. This difference reflects the degree of mutation in the height adjustment requirements between two tiles. If the difference is too large, it indicates a tendency of local unevenness of the slope. The specific operation is as follows: starting from the first tile, successively read its corresponding leveling adjustment amount, and then subtract the adjustment amount of the previous tile to obtain a new difference sequence. Each difference represents the adjustment difference between a group of adjacent tiles.

[0101] After the differential sequence is generated, analyze each item of the sequence to determine whether there are points where the value exceeds a fixed threshold. This threshold is set according to the construction accuracy level, and the common range is from three millimeters to five millimeters, depending on the site requirements and tile specifications. For example, if a set of differences is six millimeters, which exceeds the set threshold of five millimeters, it is considered a non-smooth node, and this position is marked as a "warning node" during construction. The marking method can be through visual marking on the construction drawings or by prompting the construction personnel through the measurement data report, reminding them to focus on the leveling quality of this node during actual operation to prevent inconsistent phenomena from occurring on the subsequent laying surface.

[0102] For the multi-segment linear consistency detection, it is to conduct an overall trend analysis on the leveling adjustment amount of the tiles in a continuous laying area. Its core purpose is to determine whether there is a sudden deflection or linear bend in the slope of this laying surface. The operation method is as follows: Select the leveling adjustment amounts of five consecutive tiles. Using their spatial positions as independent variables and the height adjustment amounts as dependent variables, construct five data points and perform a linear regression analysis to fit the linear trend of this section. Then, extract the path tangent direction corresponding to this section from the original leveling reference path, and use the theoretical trend of this path section as the comparison benchmark.

[0103] Next, calculate the angle between the fitted result line and the path tangent. The angle calculation is based on the difference of the arctangent function of the slopes of the two lines and is finally converted into an angle value. If this angle exceeds the preset value, such as fifteen degrees, it is considered a "risk section of linear mutation". The setting of this threshold is based on the manual construction ability and the tolerance of visual error, usually fluctuating between ten degrees and twenty degrees. The larger the angle value, the more obvious the deviation between the actual adjustment surface and the reference path, which may cause the tile surface to show a zigzag fluctuation visually or lead to local stress concentration.

[0104] When the detection result shows a risk of mutation, the construction personnel should review the operation trajectory of this section of the area, including reconfirming factors such as the tile placement position, whether the leveling reference point is offset, and whether there is a material thickness error. After identifying the source of the problem, according to the on-site conditions, take manual fine-tuning measures to appropriately raise or lower the elevation of the relevant tiles to make them conform to the continuity requirements of the reference curve again, so as to improve the overall visual coherence and structural smoothness after laying.

[0105] For example, in a certain ramp area, the leveling adjustment of five consecutive tiles was 3 mm, 6 mm, 9 mm, 8 mm, and 12 mm respectively. The fitting slope was large, and the angle between the original tangent line of the path was calculated to be 17 degrees, which exceeded the set upper limit of 15 degrees. Therefore, the section was marked as a linear mutation section. After receiving the prompt, the construction personnel reviewed the original measurement records and the actual laying situation and found that the elevation data of the third block was affected by the abnormally high value of the previous block. Therefore, the height was readjusted to eliminate the linear mutation and restore the path to smoothness.

[0106] After confirming the source of the problem, manual fine-tuning is performed according to the on-site conditions to appropriately raise or lower the elevation of the relevant tiles so that they are close to the continuity requirements of the benchmark curve. The fine-tuning includes the following specific methods:

[0107] Initial loosening and positioning: First, at the position of the tile that has been initially laid but not yet finally fixed, use a rubber hammer, flat scraper or light prying tool to gently tap or lift the four corners or edges of the target tile to loosen it between the underlying leveling material (such as mortar layer, dry-laid leveling pad) to leave adjustment space for subsequent lifting and lowering.

[0108] Manual padding or extraction of materials: If the elevation of the point is lower than the target elevation after measurement, a thin layer of leveling material can be added manually under the tile, such as scraping a small amount of dry hard mortar, adding plastic leveling pads, and locally spreading fine-grained sand to raise the tile as a whole to close to the target value; if the elevation is too high, a thin scraper can be used to extract or compact part of the leveling material under the tile to lower the laying height.

[0109] Synchronous control of level ruler and target point comparison: During the lifting process, by placing a level ruler or laser marking device, the height difference between the fine-tuning target tile and the adjacent tiles is compared to ensure that the height of the fine-tuned tile is linearly connected with the two ends or multi-point reference tiles to prevent local "bulge" or "depression" and improve linear continuity.

[0110] Final position locking and secondary measurement: After determining the target height, re-tap the tile to fix it in place, then use a digital distance measuring tool or laser elevation meter to measure the actual elevation of the point again to verify whether it meets the error tolerance with the leveling reference path. If the error is still not within the range, repeat the above steps for a second round of fine-tuning until the accuracy requirements are met.

[0111] For example, after discovering that the tenth tile was six millimeters higher than the target elevation, the construction worker first used a rubber hammer to tap the edges and corners of the tile to loosen it, and then used a scraper to scrape out some mortar from the bottom of the tile, causing it to sink three millimeters. After comparing it with the spirit level, it was found that it was still three millimeters higher than the target elevation. Finally, manual pressure was used to press the center of the tile to the ideal position and the tile was remeasured. The final error was controlled within two millimeters, which was within the construction tolerance.

[0112] By combining the smoothing of elevation differences with the detection of multi-segment linear consistency, the present invention ensures dynamic management of slope continuity at both the macroscopic control and local response levels, effectively suppressing overall disharmony caused by the spread of local errors and improving the final laying quality.

[0113] When there are single-point elevation anomalies or mutation values in a local area, the independent adjustment process performed includes three steps:

[0114] First, starting from the mutation point, extend two tile positions forward and backward along the path point sequence to collect five-point elevation values for local analysis; secondly, use any one of linear interpolation method, cubic spline interpolation method or weighted average method to generate a local correction curve segment and calculate the new elevation value to which the mutation point should be adjusted;

[0115] Finally, calculate the difference between the new elevation value and the actual elevation of the current tile, output the local elevation adjustment value, and set the leveling operation to be only performed within this five-point interval, without affecting the laying benchmarks of the completed areas before and after it, ensuring that the correction operation is limited to the necessary range and avoiding damage to the overall coherence.

[0116] "Elevation anomaly" refers to the situation where the difference between the actual elevation and the target elevation exceeds the tolerance standard; if this anomaly shows a sudden fluctuation in the spatial sequence and the difference from adjacent points is significant, then this difference is defined as a "mutation value", and its occurrence position is the "mutation point", which is the trigger basis for local correction operations and the interpolation input reference point; for example, during the leveling process, the actual elevation difference of the twelfth tile is +10 mm, exceeding the preset tolerance (±5 mm), constituting an elevation anomaly; at the same time, its elevation difference from adjacent tiles is only ±2 mm, forming a mutation value of 8 mm, so this point is identified as a mutation point and enters the local interpolation correction process.

[0117] The specific independent adjustment process performed is as follows: First, starting from the mutation point, extend two tile positions forward and backward along the path point sequence to collect five-point elevation values for local analysis. A mutation point refers to a point where the elevation difference between the actual elevation and the target elevation of a certain tile during the leveling adjustment process is much higher than the acceptable range, and there are no similar deviations in adjacent tiles, showing significant sudden characteristics. For example, if the current acceptable error is five millimeters, and the elevation difference at this point is ten millimeters, and the elevation differences of the two adjacent points are both within three millimeters, then this point can be defined as a mutation point. Subsequently, starting from this mutation point, count two tile positions forward and two tile positions backward, and collect the actual elevations and target elevations of a total of five tiles to construct a local height analysis data set. The five-point data range is sufficient to depict the height trend of the current paragraph while being limited to the local range to avoid large-area interference.

[0118] Next, use any one of linear interpolation method, cubic spline interpolation method or weighted average method to generate a local correction curve segment, and calculate the new elevation value to which the mutation point should be adjusted. The three algorithms are suitable for different scenarios: the linear interpolation method is applicable when the height change trend is basically uniform, and the calculation method is that the target elevation of the mutation point is equal to the average of the target elevations of the two adjacent points before and after; the cubic spline interpolation method is applicable when the surrounding elevation changes continuously but non-linearly. By constructing multiple cubic polynomials, the fitting curve is made continuous at each point, and the first and second derivatives are ensured to be continuous at the junction points, so as to generate a smooth curve; the weighted average method is applicable when the errors of adjacent points are large. Different weights are assigned to adjacent points according to the inverse distance principle. For example, the closer the mutation point is to a control point, the higher its weight. The final target elevation is equal to the sum of the target elevations of all reference points multiplied by their respective weights. In practical applications, construction workers can select a suitable fitting method according to the height distribution pattern of the current area, or set an automatic optimization logic in the on-site tools.

[0119] Finally, calculate the difference between the new elevation value and the actual elevation of the current tile, output the local elevation difference adjustment value, and set the leveling operation to be only carried out within this five-point interval, without affecting the laying benchmarks of the completed areas before and after it, ensuring that the correction operation is limited to the necessary range and avoiding the destruction of the overall coherence. The calculation method of the adjustment value is: subtract the actual elevation of the current tile from the target elevation after fitting, and the result is the leveling correction amount required; a positive value indicates that it needs to be lifted, and a negative value indicates that it needs to be lowered. The correction operation is only carried out on the tiles within the mutation point and its extended interval, strictly avoiding interference with the path, elevation or slope of the area outside this range, so as to ensure the continuity and stability of the entire leveling benchmark. This limitation principle is especially applicable to the areas where the leveling work has been completed or the sites with strict requirements for the laying sequence.

[0120] For example, on a curved slope, ten continuous tiles have been laid. The actual elevation of the sixth tile is measured to be 0.65 meters, the target elevation is 0.55 meters, and the actual elevation difference is 1 centimeter, far exceeding the 5-millimeter error limit, so it is judged as a mutation point. The target elevations of the fifth to seventh tiles are collected as 0.54, 0.55, 0.56, and the actual elevations are 0.55, 0.65, 0.565. Using the linear interpolation method, it is estimated that the sixth tile should be 0.55 meters. Comparing with the actual 0.65 meters, the difference is -1 centimeter, that is, it needs to be lowered by 1 centimeter. This operation is limited to between the fifth and seventh tiles, and no adjustment is made to the first four and the last four leveled areas.

[0121] Through the above steps, the independent adjustment process can solve local sudden elevation difference problems at the cost of minimum intervention, avoid misleading or rework of the entire section of construction caused by single-point errors, and maximize the maintenance of the linear continuity and elevation smoothness of the overall leveling structure.

[0122] In order to achieve a systematic expression of the local height difference correction results, the new elevation of the mutation point, the adjustment amount and the limited range are constructed as a local height difference adjustment plan. In the process of outputting the height difference adjustment plan within the local range, the priority judgment rule is introduced. The judgment rule is established according to the following standards:

[0123] The first is whether the absolute value of the height difference of the mutation point exceeds the set upper limit, the second is whether the rate of change of the height difference of the mutation point relative to the adjacent point is the maximum value in this interval, and the third is whether the point is located in the edge area or the structural turning area. If any two of them are met, the point is set as a high-priority correction point and listed as the primary operation object in the height difference adjustment plan; this process is used to reasonably allocate the focus of leveling work in construction sites with limited human resources, improve efficiency and reduce the possibility of structural deviation caused by ignoring abnormal points, forming a set of correction operation logic with guidance, judgment and limitation, and enhancing the actual executable ability of the overall leveling method on complex and irregular ground.

[0124] The "elevation adjustment plan" referred to in this description refers to the collective expression of the corrected target elevation value, the corresponding leveling adjustment amount and the local operation instructions generated based on the elevation analysis results for the mutation point and its local area. The judgment rules are established based on the following standards:

[0125] The first criterion is whether the absolute value of the height difference at the mutation point exceeds the set upper limit. The so-called mutation point refers to a single location in the continuous paving path where there is a significant deviation from the target elevation. The height difference is the difference between the actual elevation of the point and the target elevation of the path. The absolute value means that whether it is higher or lower than the benchmark, as long as the degree of deviation exceeds the threshold, it is considered abnormal. The upper limit is set according to the paving accuracy requirements, and the common range is three to eight millimeters. For example, when paving a curved ramp for cultural heritage, the accuracy requirements are high, so the upper limit is set to three millimeters; if the height difference at a certain point is measured to be negative ten millimeters, it is obviously over the limit and meets the first criterion.

[0126] The second criterion is whether the rate of change of the mutation point relative to the height difference of the adjacent points is the maximum value in this interval. The rate of change is calculated by taking the actual height difference between the mutation point and the adjacent tile on its left or right side as the difference, and dividing it by the spatial distance between the two to obtain the rate of change of the height difference per unit distance, and then comparing this value with the rate of change of other tiles in the same section. If it is the largest, it is determined to be the position with the most drastic height jump. For example, in a local interval composed of five tiles, the height difference change rates between the tiles are 1.5 mm / m, 2 mm / m, 4.5 mm / m, 1 mm / m, and 0.5 mm / m, respectively. The change between the third tile and the adjacent point is the most drastic, which meets the second criterion.

[0127] The third criterion is whether the point is located in the edge area or the structural transition area. The edge area refers to the control points within one meter at the very front or the very end of the laying range. The structural transition area refers to the positions where the path curvature changes, such as the position where the downhill slope turns into a flat slope or the position where the flat slope enters a turning section, i.e., the positions where the curve deflects. In these areas, due to the high requirement for path continuity and stronger sensitivity to errors, deviations are more likely to cause the overall structural offset or visual mutation. If a point is located in the end area at the bottom of the slope or exactly in the transition section of the curve, it meets the third criterion.

[0128] If any two of the above are met, set this point as a high-priority correction point and list it as the primary operation object in the height difference adjustment plan. Such points will be processed in advance and no longer processed according to the normal construction sequence. The purpose is to form a constraint and traction effect on the stability of the overall path benchmark through priority correction, and avoid the backward transmission of deviations.

[0129] This process is used to reasonably allocate the focus of the leveling work at the construction site with limited human resources, improve efficiency and reduce the possibility of structural offset caused by ignoring abnormal points. For example, on a ten-meter-long special-shaped ramp, the construction workers detected that the height difference of the third point was -9 millimeters, and the change rate was higher than the values of the points on both sides. Moreover, this position was exactly at the corner of the slope, meeting all three determination conditions. Based on this, the construction worker listed it as the primary adjustment point and completed the height correction and repositioning of this point prior to the subsequent leveling operation to ensure that the subsequent tiles are laid point by point according to the correct benchmark.

[0130] Ultimately, this priority determination rule forms a set of guiding, judgmental, and restrictive correction operation logics, strengthening the practical executable ability of the overall leveling method on complex special-shaped floors, taking into account the balance between local intervention and global control, and ensuring that manual construction always makes dynamic adjustments around the path surface benchmark in complex terrains.

[0131] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0132] It should be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the above processes do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0133] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0134] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0135] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A method for leveling tiles on special-shaped floors, characterized in that: The following steps are involved: Through on-site measurement, obtain the curvature change value and slope distribution data of the target ground in a continuous spatial range, and select no less than ten representative spatial control points in the required paving area, measure their elevation values ​​one by one, and form an initial construction parameter set; The obtained curvature change value, slope distribution data and elevation value of the spatial control point are processed comprehensively to generate a leveling reference path arranged in a spatial sequence. The leveling reference path is a positioning reference curve reflecting the undulating characteristics of the original terrain and is used as a unified baseline for positioning and height comparison in the subsequent manual paving process. According to the leveling reference path, when laying the actual tiles, the elevation of each tile to be laid at the current construction point is collected for the pre-placed state, and the elevation is numerically compared with the target elevation on the corresponding leveling reference path, and the difference between the two is calculated to form a height difference data sequence reflecting the leveling deviation; According to the height difference data sequence, the leveling adjustment amount required for each tile is output according to the construction order, and the slope continuity during the leveling adjustment process is controlled by continuously comparing the change trend of the adjustment values ​​of adjacent tiles to ensure that the overall paving surface forms a unified elevation transition relationship without jump points; When there is an abnormal or sudden change in height at a single point in a local area, the independent adjustment process of the area is immediately started. The correction value is calculated through the interpolation algorithm of the adjacent control points, and the adjustment range is limited to not extend beyond the adjacent area, ensuring that the local leveling correction operation is achieved without destroying the continuity of the overall leveling logic.

2. The method for leveling tiles on special-shaped floors according to claim 1, characterized in that: The process of obtaining the curvature change value and slope distribution data of the target ground in a continuous spatial range includes two continuous measurement steps: The first step is to establish multiple spatial measurement lines in the vertical and horizontal directions in the predetermined paving area by handheld measurement tools or surveying equipment to form interlaced measurement grid lines. The second step is to obtain the three-dimensional spatial coordinate value of each intersection point, and perform curvature fitting and slope calculation on all measured points. The curvature fitting adopts any one of the quadratic curve fitting method, hyperbolic surface fitting method or polynomial regression method. The slope is calculated numerically based on the ratio of elevation change to horizontal distance, and the average change value is taken as the slope parameter. The curvature change value and slope distribution data are used as data support sources for subsequent path generation and comparison benchmarks.

3. The method for leveling tiles on special-shaped floors according to claim 2, characterized in that: The process of selecting at least ten representative spatial control points includes the following two steps: The first step is to project the area to be paved into a standard rectangle according to its actual boundary size, and evenly divide it into multiple regular grid units with a spacing of no more than two meters in the horizontal and vertical directions according to the projection range, and preliminarily arrange a spatial control point at the center point of each grid unit as the basic control point; The second step is to determine whether the elevation difference of the control points between three adjacent grid cells exceeds a threshold of five millimeters based on the local terrain after the preliminary layout is completed. If there are two or more consecutive locations that meet the difference condition, it is determined that the terrain in the area is drastically changing, and an additional control point is added to each edge position in the adjacent edge grids of the area to form a high-density control point area. The control point set formed by the two-step layout forms a gradient distribution structure of "low density in the center and high density at the edge" in space, and the distribution of the grid center points maintains matrix uniformity. The minimum spacing between all control points is not less than 0.5 meters, and the maximum spacing is no more than two meters, ensuring the formation of an elevation point set with both uniform coverage and gradient characteristics in space, which is used as the basic data support for subsequent leveling path fitting and error control.

4. The method for leveling tiles on special-shaped floors according to claim 3, characterized in that: The generated leveling reference path is a curve in space. The position of the curve is determined after fitting the spatial control points. The generation of the path point sequence follows the sequential construction method from the starting control point to the ending control point. The forward point-by-point interpolation algorithm is used in the path generation process, and no less than three intermediate transition points are inserted between every two control points to ensure that the path curvature maintains continuity and smoothness in each micro-segment. The distance change between a path point and its corresponding control point is limited to no more than ten percent. At the same time, the target elevation of each path point is obtained by calculating the weighted average of the elevations of its adjacent control points. The weighting method calculates the weight factor based on the spatial distance between the point and the control point and normalizes it. This process ensures the fit of the path point in the spatial surface, thereby improving the visual predictability and practical feasibility during the manual leveling operation.

5. The method for leveling tiles on special-shaped floors according to claim 4, characterized in that: The process of numerically comparing the elevation of the tiles in their initial laying state with the target elevation on the leveling reference path, collecting the current elevation of the tiles through digital distance measurement, and calculating the height difference in combination with the target elevation value at the corresponding position of the path point. The height difference is recorded as the actual height difference. When the actual height difference exceeds the acceptable error range, the point is recorded as a key adjustment point. At the same time, the height difference between the current tile and the previous tile is recorded, and the section with three or more consecutive key adjustment points is counted. The section is marked as a high-risk area for leveling, and in subsequent manual operations, leveling adjustments are performed preferentially on high-risk areas.

6. The method for leveling tiles on special-shaped floors according to claim 5, characterized in that: After the height difference data sequence reflecting the leveling deviation is formed, before the leveling adjustment amount is output, a global trend judgment process is first performed on the data sequence. The judgment standard is whether there is a trend of continuous increase or decrease on one side in the height difference value in the spatial sequence. If the continuous change exceeds five tile positions, it is considered that there may be abnormal path points or accumulation of artificial laying errors, and the error backtracking logic is entered; According to the adjustment direction of the adjacent tiles, the elevation of the previous path point is reversely calculated and an attempt is made to generate a revised path point. The revised path point is generated based on the trend of the original path point as the initial vector, and the local curve segment is refitted in combination with the elevation of the control point near the error point to smoothly replace the original path segment, thereby improving the global leveling consistency and preventing the slope jump phenomenon caused by error accumulation.

7. The method for leveling tiles on special-shaped floors according to claim 6, characterized in that: The method of controlling the continuity of the slope during the leveling adjustment process includes two parts: height difference smoothing processing and multi-segment linear consistency detection. Height difference smoothing processing refers to making a difference between the leveling adjustment amount of each tile and the adjustment amount of the adjacent tile. After calculating the difference sequence, the points where the difference exceeds a fixed threshold are marked as early warning nodes during the construction process. The multi-segment linear consistency test performs a linear regression analysis on the leveling adjustment amounts of five consecutive tiles, and compares the angle with the tangent direction of the overall fitting curve in the segment. If the angle exceeds the preset value, it is determined that there is a risk of linear mutation in this segment, and the construction personnel are required to review whether the actual operation trajectory deviates from the reference path.

8. The method for leveling tiles on special-shaped floors according to claim 7, characterized in that: When there is a single point of high-difference abnormal or sudden change value in a local area, the independent adjustment process performed includes three steps: First, starting from the mutation point, two tile positions are extended forward and backward along the path point sequence, and five-point elevation values ​​are collected for local analysis; secondly, a local correction curve segment is generated using any one of the linear interpolation method, cubic spline interpolation method or weighted average method to calculate the new elevation value to which the mutation point should be adjusted; Finally, the difference between the new elevation value and the actual elevation of the current tiles is calculated, and the local height difference adjustment value is output. It is set that the leveling operation is only performed within the five-point range, without affecting the paving benchmark of the completed areas before and after, ensuring that the correction operation is limited to the necessary scope to avoid the overall continuity being destroyed.

9. The method for leveling tiles on special-shaped floors according to claim 8, characterized in that: Priority determination rules are introduced in the process of outputting the elevation adjustment plan within the local range. The determination rules are established based on the following standards: The first is whether the absolute value of the height difference of the mutation point exceeds the set upper limit; the second is whether the rate of change of the height difference of the mutation point relative to the adjacent point is the maximum value in this interval; the third is whether the point is located in the edge area or the structural turning area. If any two of them are met, the point is set as a high-priority correction point and listed as the primary operation object in the height difference adjustment plan.

Images