Machine tool assembly precision detection method and system

Abstract

The present application relates to the technical field of laser measurement, in particular to a machine tool assembly precision detection method and system, comprising the following steps: collecting the displacement of measuring points arranged along the guide rail and calculating the offset relationship relative to the laser reference, determining the rigid transmission direction in combination with the bed structure and the connection sequence, sequentially rearranging and sectioning the measuring point data, analyzing the consistency of the offset directions of adjacent measuring points and forming a continuous cumulative relationship, eliminating abnormal measuring points through section continuity discrimination, on this basis, continuing the effective offset cumulative result to complete straightness calculation, and outputting the machine tool assembly precision detection conclusion, in the present application, the data is made to conform to the rigid transmission direction through the corresponding association of the measuring point offset sequence and the bed structure, the cumulative analysis of the consistency of the offset directions of adjacent measuring points is carried out, the real structural offset is reflected, abnormal measuring points are eliminated in combination with the section continuity discrimination, local interference affecting the overall judgment is avoided, the stability and consistency of the assembly precision evaluation are improved, and effective quantitative judgment of the complex bed state is realized.

Description

Technical Field

[0001] This invention relates to the field of laser measurement technology, and in particular to a method and system for detecting the assembly accuracy of machine tools. Background Technology

[0002] Laser measurement technology is a field that utilizes the linearity, coherence, and stable wavelength characteristics of lasers to measure geometric quantities such as length, displacement, angle, and form and position errors of objects. Its core aspects include the emission and reception of laser light sources, the propagation and reflection of the beam in space, the acquisition of interference or displacement changes, and the establishment of measurement benchmarks. It is widely used in precision manufacturing, equipment assembly, geometric accuracy inspection, and online inspection scenarios to perform high-precision quantitative evaluation of mechanical structures and assembly conditions.

[0003] Traditional machine tool assembly accuracy testing methods and systems refer to the use of laser interferometers, laser displacement sensors, or laser tracking devices to detect the straightness, parallelism, perpendicularity, and positional deviation of key components such as the machine tool bed, guide rails, and spindle after or during the assembly of machine tool parts. Typically, a fixed laser emitting end is used as the measurement reference, and the laser beam is propagated along the measured axis or reference direction. A reflector or target is set on the surface of the measured component, and assembly error information is obtained by collecting changes in laser echo displacement or interference fringes. The overall assembly accuracy of the machine tool is judged by combining multiple measurement points and repeated measurements.

[0004] Current laser inspection of machine tool assembly accuracy relies on a fixed laser reference and changes in echoes from discrete measuring points for judgment. During the measurement process, multiple measuring points are assumed to be independent and equivalent, lacking an overall consideration of the rigidity transmission relationship of the machine bed structure and the assembly sequence. The inspection results are mainly obtained based on repeated comparisons of multiple points, which makes it difficult to reflect the transmission and superposition characteristics of offset in the structure. In cases of uneven assembly rigidity or complex connection sequence, the overall offset caused by structural transmission is easily misjudged as a local error, resulting in discrete fluctuations in straightness and position assessment results. At the same time, abnormal measuring points or local interference can directly affect the overall judgment, increasing the frequency of manual review and adjustment, thus limiting the inspection efficiency and stability. Summary of the Invention

[0005] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for detecting the assembly accuracy of machine tools, comprising the following steps:

[0006] S1: The displacement signal of the measuring points arranged along the length of the guide rail under the machine tool bed guide rail assembly is collected by the laser detection instrument and the analog-to-digital conversion is performed to calculate the spatial offset between the measuring points and the laser reference line and generate the measuring point offset data.

[0007] S2: Read the bed structure information and identify the connection sequence of the stiffener and guide rail, analyze the rigid topology of the bed assembly, determine the rigidity transmission direction, mark the rigidity attribute and section mapping of the measurement point offset data, and map it with the corresponding bed structure to generate measurement point rigidity data.

[0008] S3: Based on the rigidity data of the measuring points, the offset of adjacent measuring points arranged along the rigidity transmission direction is used to determine the consistency of the direction. If the direction is consistent, the offset vector is accumulated. If they are inconsistent, the rigid deformation inflection point is marked and segmented to generate rigidity transmission data.

[0009] S4: Based on the rigid transmission data, perform smoothness analysis on the cumulative offset change relationship of the measuring points in the multi-bed structure section, filter out measuring points that exceed the preset smooth change threshold, and generate configuration offset data;

[0010] S5: Based on the configured offset data, analyze the consistency of the offset direction between the end of the adjacent bed structure section and the starting measuring point. If they are consistent, continue the cumulative offset result and perform straightness offset calculation to generate machine tool assembly accuracy test results.

[0011] As a further embodiment of the present invention, the measuring point offset data includes measuring point spatial coordinates, laser reference line offset vector, and guide rail length sequence index; the measuring point rigidity data includes rigid section identifier, measuring point spatial sequence index, and section structure mapping relationship; the rigid transmission data includes continuously accumulated offset vector, segment position index, and rigid transmission direction identifier; the configuration offset data includes continuous measuring point index set, cumulative offset sequence within the section, and continuity judgment mark; and the machine tool assembly accuracy detection result includes bed straightness offset sequence and section splicing offset residual.

[0012] As a further aspect of the present invention, the specific steps of S1 are as follows:

[0013] S101: The displacement signals of multiple measuring points arranged along the length of the guide rail under the machine tool bed guide rail assembly are collected by a laser detection instrument. The continuous displacement signals are sampled at equal intervals with a position sensor. The sampled analog voltage is mapped into discrete digital quantities to generate a digital sequence of measuring point displacements.

[0014] S102: Based on the digital sequence of displacement of the measuring points, call the spatial parameters of the laser reference line, perform numerical difference calculation on the displacement values ​​of multiple measuring points and the theoretical positions of the laser reference line in the same coordinate system, and vectorize the difference results to generate a set of spatial offset of measuring points.

[0015] S103: Based on the set of spatial offsets of the measuring points, establish a serialized sort according to the physical location coordinates of the measuring points, maintain the spatial distribution relationship of the guide rail shape, and integrate the sorting results with the corresponding measuring point offsets to generate measuring point offset data.

[0016] As a further aspect of the present invention, the specific steps of S2 are as follows:

[0017] S201: Based on the measured point offset data, obtain the bed structure design information, align the stiffener plate number, stiffener plate spatial position parameters and guide rail installation position parameters, and perform sequence determination according to the structural connection sequence to transform the connection relationship between the stiffener plate and the guide rail, generating a stiffener plate and guide rail connection sequence.

[0018] S202: Based on the stiffener guide rail connection sequence, call the connection topology of multi-stiffener nodes and guide rail nodes in the bed structure design information, judge the constraint relationship between adjacent connection nodes, and perform direction determination according to the constraint transmission path to generate a rigid transmission direction sequence.

[0019] S203: For the rigid transmission direction sequence, the original spatial order of the measurement point offset data is maintained, and the segments are divided according to the transmission direction change nodes. The corresponding segments of the measurement point sequence are mapped to the corresponding bed structure segments to generate measurement point rigid data.

[0020] As a further aspect of the present invention, the specific steps of S3 are as follows:

[0021] S301: Based on the rigid data of the measuring points, detect the offset direction vectors corresponding to multiple measuring points, perform vector consistency judgment on the offset directions of adjacent measuring points, compare the direction angle with the preset offset direction consistency threshold, write the judgment result into the measuring point sequence identifier sequence, and generate an offset direction consistency tag set.

[0022] S302: Based on the offset direction consistency mark set, call the offset data of the measurement points marked with consistent directions, perform numerical accumulation operation on the offsets within the same rigid transmission path, and record the stacking results sequentially during the accumulation process to obtain the accumulated offset sequence.

[0023] S303: For the accumulated offset sequence, call the index position of the measurement point marked as inconsistent in direction, mark the state segment of the corresponding measurement point in the rigid transmission path, and restart the accumulation record after the marked measurement point with the current point as the reference to generate rigid transmission data.

[0024] As a further aspect of the present invention, the offset direction consistency threshold is determined by statistically analyzing the angle difference between the offset direction vectors of adjacent measuring points to obtain the distribution sequence of the angle difference, calculating the median and standard deviation of the distribution sequence of the angle difference, and taking the sum of the median and a preset three times the standard deviation.

[0025] As a further aspect of the present invention, the specific steps of S4 are as follows:

[0026] S401: Based on the rigid transmission data, obtain the cumulative offset of multiple measuring points within the same structural segment, calculate the adjacent difference of the cumulative offset according to the spatial order of the measuring points, combine them into a change sequence and judge the smoothness of the change trend, filter out measuring points that exceed the preset smooth change threshold, and generate a continuous index set of cumulative offset.

[0027] S402: Based on the cumulative offset continuous index set, call the cumulative offset change sequence of the corresponding measurement point, mark the measurement point index whose change amplitude exceeds the preset continuous change judgment threshold as an abnormal index, and remove the abnormal index from the continuous index set to generate a continuous offset index set;

[0028] S403: For the continuous offset index set, call the corresponding cumulative offset data of the measurement points, call the corresponding cumulative offset of the measurement points, perform re-aggregation and interpolation connection on the data items corresponding to the index, map the aggregated offset data with the measurement point identifier, and generate configuration offset data.

[0029] As a further aspect of the present invention, the continuous change determination threshold is determined by statistically analyzing the adjacent change amplitude sequences of the cumulative offset of the measurement points within the same structural segment, obtaining the distribution characteristics of the change amplitude, calculating the median and dispersion index of the distribution characteristics, and then weighted summing the median and dispersion index to determine the threshold.

[0030] As a further aspect of the present invention, the specific steps of S5 are as follows:

[0031] S501: Based on the configured offset data, extract the cumulative offset vector components between the end measuring point and the starting measuring point of the adjacent bed structure section, perform sign consistency judgment on the multi-vector components, and summarize the judgment results to generate a section connection direction consistency mark set.

[0032] S502: Based on the segment connection direction consistency mark set, perform sequence splicing on the adjacent segment measurement point indices that are marked as consistent, and perform order mapping and index rearrangement on the spliced ​​cumulative offset sequence to generate a cross-segment cumulative offset continuous sequence;

[0033] S503: For the continuous sequence of cumulative offset across sections, perform linear fitting residual calculation on the coordinates of the corresponding measuring points and the cumulative offset within the multi-bed structure section, and serialize and aggregate the residual values ​​of multiple measuring points to generate machine tool assembly accuracy detection results.

[0034] Machine tool assembly accuracy testing system, including:

[0035] The measurement point analysis module uses a laser detection instrument to collect displacement signals of measurement points arranged along the length of the guide rail under the machine tool bed guide rail assembly, performs analog-to-digital conversion, calculates the spatial offset between the measurement point and the laser reference line, generates measurement point offset data, and transmits it to the rigid mapping module.

[0036] The rigid mapping module reads the bed structure information and identifies the connection sequence of the stiffeners and guide rails, analyzes the rigid topology of the bed assembly, determines the rigid transmission direction, marks the rigid attributes and section mapping of the measurement point offset data, maps it with the corresponding bed structure, generates the measurement point rigid data, and transmits it to the transmission accumulation module.

[0037] The transmission and accumulation module, based on the rigid data of the measurement points, performs a direction consistency judgment on the offset of adjacent measurement points arranged along the rigid transmission direction. If the directions are consistent, the offset vector is accumulated. If they are inconsistent, the rigid deformation inflection point is marked and segmented, generating rigid transmission data and transmitting it to the continuous filtering module.

[0038] The continuous screening module, based on the rigid transmission data, performs a smoothness analysis on the cumulative offset change relationship of the measuring points in the multi-bed structure section, filters out measuring points that exceed the preset smooth change threshold, generates configuration offset data, and transmits it to the straight line detection module.

[0039] The straight line detection module analyzes the consistency of the offset direction between the end of the adjacent bed structure section and the starting measuring point based on the configured offset data. If they are consistent, the cumulative offset result is continued and the straightness offset is calculated to generate the machine tool assembly accuracy detection result.

[0040] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0041] In this invention, by sequentially identifying the offset data of measuring points along the guide rail and correspondingly associating it with the bed structure information, the measured data is kept consistent with the actual rigidity transmission direction of the assembly. On this basis, the consistency analysis of the offset direction of adjacent measuring points is performed and a continuous cumulative relationship is formed, so that the structural offset can be truly reflected. In addition, abnormal measuring points are screened out by judging the continuity of offset in multiple sections, avoiding the amplification of local interference into overall error. This makes the straightness calculation based on the actual structural transmission, thereby improving the stability and consistency of assembly accuracy assessment, reducing reliance on repeated measurements and manual experience, and realizing effective quantitative judgment of complex bed assembly conditions. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the steps of the present invention;

[0044] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0045] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0046] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0047] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0048] Figure 6 This is a detailed schematic diagram of S5 of the present invention;

[0049] Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0050] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0051] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0052] Please see Figure 1 This invention provides a method for detecting the assembly accuracy of machine tools, comprising the following steps:

[0053] S1: The displacement signal of the measuring points arranged along the length of the guide rail under the machine tool bed guide rail assembly is collected by the laser detection instrument and the analog-to-digital conversion is performed to calculate the spatial offset between the measuring points and the laser reference line and generate the measuring point offset data.

[0054] S2: Read the bed structure information and identify the connection sequence of stiffeners and guide rails, analyze the rigid topology of bed assembly, determine the rigidity transmission direction, mark the rigidity attributes and section mapping of the measurement point offset data, and map it with the corresponding bed structure to generate measurement point rigidity data.

[0055] S3: Based on the rigidity data of the measuring points, the offset of adjacent measuring points arranged along the rigidity transmission direction is used to determine the consistency of the direction. If the direction is consistent, the offset vector is accumulated. If they are inconsistent, the rigid deformation inflection point is marked and segmented to generate rigidity transmission data.

[0056] S4: Based on rigid transfer data, perform smoothness analysis on the cumulative offset change relationship of measuring points in the multi-bed structure section, filter out measuring points that exceed the preset smooth change threshold, and generate configuration offset data;

[0057] S5: Based on the configuration offset data, analyze the consistency of the offset direction between the end of the adjacent bed structure section and the starting measuring point. If they are consistent, continue the cumulative offset result and perform straightness offset calculation to generate the machine tool assembly accuracy test result.

[0058] The measurement point offset data includes the spatial coordinates of the measurement point, the offset vector of the laser reference line, and the guide rail length sequence index. The measurement point rigidity data includes the rigid section identifier, the spatial sequence index of the measurement point, and the section structure mapping relationship. The rigid transfer data includes the continuously accumulated offset vector, the segment position index, and the rigid transfer direction identifier. The configuration offset data includes the continuous measurement point index set, the cumulative offset sequence within the section, and the continuity judgment mark. The machine tool assembly accuracy detection results include the bed straightness offset sequence and the section splicing offset residual.

[0059] Please see Figure 2 The specific steps of S1 are as follows:

[0060] S101: The displacement signals of multiple measuring points arranged along the length of the guide rail under the machine tool bed guide rail assembly are collected by a laser detection instrument. The continuous displacement signals are sampled at equal intervals with a position sensor. The sampled analog voltage is mapped into discrete digital quantities to generate a digital sequence of measuring point displacements.

[0061] A laser interferometer receiver is deployed on the moving part of the machine tool bed guideway, and a laser emitter is deployed on the fixed end of the guideway. The laser detection instrument is activated, and as the moving part moves at a constant speed along the entire length of the guideway, the continuous displacement analog voltage signal generated by the change in optical path difference is acquired in real time. A sampling trigger mechanism with equal time intervals is set, and the time step is calculated based on the feed speed of the machine tool moving part and the preset spatial resolution requirements. A high-frequency clock pulse triggers the sample-and-hold circuit to discretize the continuously changing analog voltage signal. Subsequently, a high-precision analog-to-digital converter maps the acquired analog voltage quantity into a discrete digital quantity. This mapping process uses linear proportional correspondence logic to align the full-scale range of the analog voltage signal with the maximum value of the digital quantization bits. By calculating the proportion of the current sampled voltage value within the full-scale range and multiplying this proportion by the maximum digital quantization value, the corresponding dimensionless digital code is obtained. Finally, the generated series of digital codes are arranged in timestamp order. For example, if the feed speed of the machine tool's moving parts is set to 50 mm / s, and the preset spatial sampling interval is 50 mm for rapid full-length straightness scanning, then the sampling time interval is calculated to be 1.0 second through division (50 / 50=1), i.e., the sampling frequency is 1 Hz. A laser displacement sensor with an output range of 0 volts to 10 volts and a 16-bit resolution analog-to-digital converter are selected, with a digital quantization range of 0 to 65535. If the acquired analog voltage signal is 4.5 volts at a certain sampling moment, the mapping operation is performed: first, the ratio of 4.5 volts to the full-scale range of 10 volts is calculated (resulting in 0.45), then this ratio is multiplied by 65535, i.e., 0.45 * 65535 = 29490.75, which is rounded to the nearest integer to obtain the digital value 29491. The digital value 29491 is the displacement digital code corresponding to the measuring point at the current moment. All such codes of the sampling points are arranged in sequence to form the displacement digital sequence of the measuring points.

[0062] S102: Based on the digital sequence of displacement of the measuring points, call the spatial parameters of the laser reference line, perform numerical difference calculation on the displacement values ​​of multiple measuring points and the theoretical positions of the laser reference line in the same coordinate system, and vectorize the difference results to generate a set of spatial offset of measuring points.

[0063] Based on the generated digital sequence of measurement point displacements, the pre-calibrated laser reference line spatial parameters are first retrieved from non-volatile memory. These parameters include the linear equation coefficients of the laser beam in the machine tool coordinate system, namely the slope and intercept parameters. Next, the dimensionless digital quantities in the measurement point displacement sequence are converted into actual displacement values ​​with physical units, and the actual coordinates of each measurement point in the machine tool coordinate system are calculated by combining the kinematic parameters at the time of sampling. Subsequently, a numerical difference operation is performed, subtracting the actual displacement value of each measurement point from the theoretical position value of the laser reference line at the same longitudinal coordinate position. That is, subtracting the theoretical reference value from the actual measured value yields the single-point deviation value of the measurement point relative to the reference line. Then, the difference result is vectorized, using the longitudinal position coordinates of the measurement point as the index dimension of the vector, and the calculated single-point deviation value as the numerical dimension of the vector. The direction attribute of the vector is determined based on the sign of the deviation value. For example, in the retrieved laser reference line spatial parameters, the slope is set to 0 (ideal horizontal), and the intercept is 0. When processing the 5th measurement point, based on the set 50 mm sampling interval, the corresponding longitudinal coordinate position of the machine tool at this measurement point is 200 mm (i.e., 0 mm for the 1st point and 200 mm for the 5th point). First, the digital value (set to 31784) of this measurement point is converted into a physical displacement value. If the sensor's full scale corresponds to 0 to 1000 micrometers, the converted value is 31784 / 65535*1000=485 micrometers. Based on the baseline parameters, the theoretical position value at this location (200 mm) is calculated to be 450 micrometers (the baseline is set with a preset offset of 450 micrometers during installation and debugging to keep the optical path centered). At this point, a differential operation is performed, subtracting the theoretical baseline value of 450 micrometers from the actual displacement value of 485 micrometers, resulting in a difference of +35 micrometers. Subsequently, a spatial offset vector for this measurement point is generated, represented as (200 mm, 35 micrometers, positive direction). The set of such vectors for all measurement points constitutes the spatial offset set of the measurement points.

[0064] S103: Based on the spatial offset set of measurement points, establish a serialized sort according to the physical location coordinates of the measurement points, maintain the spatial distribution relationship of the guide rail morphology, and integrate the sorting results with the corresponding measurement point offsets to generate measurement point offset data;

[0065] Based on the generated set of spatial offsets of the measurement points, the data is first serialized and sorted. Using the vertical position coordinates in the vector as the key, the offsets of multiple measurement points are arranged in ascending order from the start end to the end end of the guide rail, reconstructing the physical topology of the entire guide rail length. Subsequently, a sequence determination is performed based on the magnitude and spatial distribution of the offset values. This determination process includes two layers of logic: the first layer compares the absolute value of the offset of each measurement point with a preset tolerance threshold for a single measurement point; if it exceeds the threshold, it is marked as a single point out of tolerance. The second layer analyzes the rate of change between adjacent measurement points, calculates the difference between the current measurement point offset and the offsets of the adjacent measurement points, and compares this difference with a preset local curvature threshold to determine if there are abrupt changes or wavy deformations. Finally, the above determination results are associated and integrated with the corresponding measurement point offsets and position coordinates to generate structured measurement point offset data. In this process, the tolerance threshold for a single measuring point is set using statistical methods: 1000 sets of guideway straightness measurement data under normal conditions are selected from the machine tool's historical operating data, the standard deviation of its offset is calculated, and the threshold is set to three times the standard deviation. For example, if the calculated standard deviation is 10 micrometers, then the tolerance threshold for a single measuring point is set to 30 micrometers. During the judgment process, the specific data are as follows: For measuring point number 1, its longitudinal coordinate position is 0 mm, the actual displacement value is 455 micrometers, the theoretical reference value is 450 micrometers, and the calculated offset value is +5 micrometers, resulting in a qualified judgment; For measuring point number 2, its longitudinal coordinate position is 50 mm, the actual displacement value is 462 micrometers, the theoretical reference value is 450 micrometers, and the calculated offset value is +12 micrometers. Since it is less than the 30-micrometer threshold, the judgment result is qualified; For measuring point number 3, its longitudinal coordinate position is 100 mm, and the actual displacement value is 478 micrometers. The theoretical baseline value is 450 micrometers, and the calculated offset value is +28 micrometers, which is considered acceptable. For measuring point 4, its longitudinal coordinate position is 150 millimeters, and the actual displacement value is 475 micrometers. The theoretical baseline value is 450 micrometers, and the calculated offset value is +25 micrometers, which is also considered acceptable. For measuring point 5, its longitudinal coordinate position is 200 millimeters, and the actual displacement value is 485 micrometers. The theoretical baseline value is 450 micrometers, and the calculated offset value is +35 micrometers. Comparison shows that 35 micrometers is greater than the 30 micrometer threshold, and the judgment result is a single point exceeding the tolerance. Regarding the spatial distribution, the difference between measuring point 2 and its adjacent measuring point 1 is calculated to be 12-5=7 micrometers. If the local curvature threshold is set to 20 micrometers, then the change in this neighborhood is within the acceptable range.

[0066] Please see Figure 3 The specific steps of S2 are as follows:

[0067] S201: Based on the measurement point offset data, obtain the bed structure design information, align the stiffener plate number, stiffener plate spatial position parameters and guide rail installation position parameters, and perform sequence determination according to the structural connection sequence to transform the connection relationship between the stiffener plate and the guide rail, generating a stiffener plate and guide rail connection sequence.

[0068] First, a communication connection is established with the machine tool bed's 3D design database to extract the structural design information of the bed casting. This information includes the geometric dimensions, spatial distribution coordinates, and positioning parameters of the guide rail mounting base surfaces for all internal stiffeners. The spatial position parameters of the stiffeners are defined as the linear coordinate values ​​of the stiffener's center plane along the guide rail length, and the guide rail mounting position parameters are defined as the linear coordinate values ​​of the mounting holes on the bottom surface of the guide rail. Then, an alignment operation is performed to map these two sets of parameters to a single linear coordinate system with the guide rail's starting end as the origin. Next, a sequence determination is performed based on the physical order of the structural connections. This determination logic uses a position overlap calculation method: the linear coordinate values ​​of each stiffener are subtracted from the coordinate values ​​of all guide rail mounting holes one by one, and the absolute value of the distance between them is calculated. This absolute value is then compared with a preset structural association threshold. If the absolute value of the distance is less than or equal to the structural association threshold, it is determined that the current stiffener and the corresponding guide rail mounting position have a direct physical support connection, and an association pair containing the connection point coordinates, stiffener number, and connection attributes is generated. Finally, all the verified association pairs are arranged in ascending order along the guide rail length to generate a rib-guide rail connection sequence reflecting the internal support structure of the bed. In this process, the structural association threshold is set based on the tolerance range of the bed casting process and the machining accuracy of the guide rail mounting holes. For example, if the retrieved casting position tolerance is 2.0 mm and the machining position tolerance is 0.5 mm, then the structural association threshold is set to 2.5 mm through a summation operation (2.0 + 0.5). The bed design information shows that the center position coordinates of rib 1 are 200 mm and rib 2 are 400 mm; the guide rail mounting hole sequence contains holes with coordinates of 199.5 mm and 402.0 mm. The difference between the rib 1 coordinates of 200 mm and the guide rail hole coordinates of 199.5 mm is calculated, yielding an absolute distance of 0.5 mm. Since 0.5 mm is less than 2.5 mm, a connection is established between the two. Similarly, the difference between the coordinates of the second stiffening plate (400 mm) and the guide rail hole (402.0 mm) is 2.0 mm, which is less than 2.5 mm, thus a connection is established. The final generated stiffening plate-guide rail connection sequence clearly shows the existence of rigid support points at 200 mm and 400 mm.

[0069] S202: Based on the stiffener guide rail connection sequence, call the connection topology of multi-stiffener nodes and guide rail nodes in the bed structure design information, judge the constraint relationship between adjacent connection nodes, and perform direction determination according to the constraint transmission path to generate a rigid transmission direction sequence.

[0070] Based on the generated stiffener guide rail connection sequence, the connection topology map of multi-stiffener nodes and guide rail nodes in the bed structure design information is further invoked. This map defines the mechanical transmission paths between various structural components. First, the constraint relationship between adjacent connection nodes is judged, and the degree of freedom attributes at the connection nodes are identified. The specific judgment logic is as follows: if the node attribute is marked as "stiffener support", then the node is assigned the "fully constrained" attribute, representing high rigidity characteristics; if the node is located between two stiffeners, it is marked as "suspended span" and assigned the "elastic constraint" attribute. Subsequently, the direction of constraint transmission path is determined. This judgment logic adopts the shortest force flow path principle: for each guide rail position point, the shortest path length of its vertical downward transmission to the bed foot is calculated. If the path passes directly through the stiffener vertically downward, the rigidity transmission direction is determined to be "vertical main load"; if the path needs to pass through the guide rail crossbeam to both sides before downward, it is determined to be "distributed transmission". Finally, the transmission attributes of all position points along the guide rail length are combined in spatial order to generate a rigidity transmission direction sequence describing the rigidity distribution characteristics of the entire guide rail length. In this determination process, a "rigid influence zone coefficient" is introduced to quantify the constraint capacity of the stiffener plate on the surrounding area. This coefficient is set to 1.5 times the thickness of the stiffener plate. For example, if the stiffener plate thickness is 20 mm, 20 * 1.5 = 30, resulting in a rigid influence zone coefficient of 30 mm. This means that within a 30 mm range before and after the center of the stiffener plate, it is determined to be a high-rigidity area with "vertical main load". Continuing the previous example, for the first stiffener plate at 200 mm, its node attribute is determined to be "fully constrained", and its rigid influence range is calculated to cover the interval from 170 mm to 230 mm (i.e., 200 minus 30 to 200 plus 30). The transmission direction within this interval is marked as "vertical main load". As for the position at 300 mm, since it is located in the middle zone between the first and second stiffener plates and exceeds the rigid influence zone, its constraint relationship is determined to be "elastic constraint", and the rigid transmission direction is "distributed transmission". The resulting rigidity transmission direction sequence clearly divides the entire length of the guide rail into several alternating high-rigidity and low-rigidity segments.

[0071] S203: For the rigid transmission direction sequence, the original spatial order of the measurement point offset data is maintained, and the segments are divided according to the nodes of the transmission direction change. The corresponding segments of the measurement point sequence are mapped to the corresponding bed structure segments to generate the rigid data of the measurement points.

[0072] For the generated rigid transmission direction sequence, the acquired measurement point offset data is rearranged to ensure that the index of the measurement point sequence is strictly aligned with the spatial coordinates of the rigid transmission direction sequence. Next, based on the nodes where the transmission direction changes, the entire guide rail length is divided into "rigid support sections" and "suspended deformation sections." The specific division logic is as follows: Coordinate points in the rigid transmission direction sequence where the attribute is flipped (i.e., the critical point changing from "vertical main load" to "distributed transmission") are retrieved and defined as section boundary points. Based on this, the rearranged measurement point sequence is cut into corresponding subsets. Subsequently, the measurement point subsets are mapped to the corresponding bed structure sections, and a structural rigidity label is attached to each measurement point, generating measurement point rigidity data containing displacement values, coordinate values, and structural rigidity attributes. In this process, a "rigidity association confidence score" parameter is introduced to verify the effectiveness of the mapping. This parameter is obtained by calculating the ratio of measurement point density to section length. If a section is 100 mm long and contains 2 measurement points, the confidence score is 0.02 points / mm. To ensure data validity, a confidence threshold of 0.01 points / mm was preset. Taking the implementation data shown in Table 1 as an example, the aforementioned calculation results were integrated. Measuring point 5, located at 200 mm, falls within the rigid support zone (170 mm to 230 mm) defined by the first stiffener plate, and is therefore marked as the "rigid support zone." Measuring point 3, located at 100 mm, falls within the suspended deformation zone and is marked as the "suspended zone." It was found that the offset of measuring point 5 is as high as 35 micrometers and is located in the high rigidity zone, indicating that the deviation at this point is not caused by stress deformation, but is highly likely due to machining errors on the guide rail mounting base or unevenness of the assembly pad. Conversely, if a large deviation occurs at a measuring point in the suspended zone, it may be attributed to stress deformation of the bed.

[0073] Table 1. Mapping Table of Rigid Data for Measuring Points

[0074]

[0075] As shown in Table 1, this table reveals the physical mechanism behind the numerical values ​​by integrating the measurement point offset with the rigidity characteristics of the bed structure.

[0076] Please see Figure 4 The specific steps of S3 are as follows:

[0077] S301: Based on the rigid data of the measuring points, detect the offset direction vectors corresponding to multiple measuring points, perform vector consistency judgment on the offset direction of adjacent measuring points, compare the direction angle with the preset offset direction consistency threshold, write the judgment result into the measuring point sequence identifier sequence, and generate an offset direction consistency tag set;

[0078] Based on the generated rigidity data of the measurement points, a local feature vector is constructed using the spatial coordinates and offset values ​​of two adjacent measurement points. The difference in longitudinal coordinates between the current measurement point and the previous measurement point is extracted as the horizontal component of the vector, and the difference in offset between the two is extracted as the vertical component of the vector, thus synthesizing a two-dimensional offset direction vector characterizing the local deformation trend. To eliminate the calculation distortion caused by the difference in dimensions, a scaling factor is introduced to normalize the vertical component, mapping the micrometer-level offset to a space of the same order of magnitude as the millimeter-level coordinates. Subsequently, a vector consistency judgment is performed, calculating the cosine value of the angle between two adjacent offset direction vectors. This calculation is obtained by dividing the dot product by the product of the magnitudes. The calculated angle value is compared with a preset offset direction consistency threshold. If the angle is less than the threshold, it indicates that the deformation trend of the adjacent segment remains smooth and is marked as "consistent in direction"; if the angle is greater than or equal to the threshold, it indicates that an inflection point or abrupt change has occurred, and is marked as "inconsistent in direction". Finally, the judgment result of each measurement point is written into the measurement point sequence identifier sequence in order, generating an offset direction consistency marker set containing Boolean value markers. In this process, the scaling factor is set to 1000, meaning that micrometer values ​​are directly treated as millimeter values ​​to calculate geometric angles. The offset direction consistency threshold is set using the tolerance zone envelope method to determine the angle corresponding to the maximum allowable rate of change: the vector deflection angle generated by the maximum allowable local curvature of the guide rail is set to 15 degrees. Considering measurement noise, a 5-degree redundancy is added, so the threshold is set to 20 degrees. Taking the data in step S203 as an example, for measuring point 3 (100 mm, +28 μm) relative to measuring point 2 (50 mm, +12 μm), its vector horizontal component is 50, and its vertical component is 16 (i.e., 28 minus 12). After scaling, the vector is recorded as (50, 16), and its tilt angle relative to the horizontal axis is calculated to be approximately 17.7 degrees. For measuring point 4 (150 mm, +25 μm) relative to measuring point 3, the vector is (50, -3), and the tilt angle is approximately -3.4 degrees. The absolute value difference between the angles of the two vectors is calculated to be 21.1 degrees (17.7 minus -3.4). Comparing 21.1 degrees with the threshold of 20 degrees, the result exceeds the threshold and is judged as "inconsistent direction". The advantage of this calculation logic is that, through vectorized angle analysis, it can keenly capture the subtle trend reversal from measuring point 3 to measuring point 4, and identify the discontinuous changes within the suspended deformation section.

[0079] S302: Based on the offset direction consistency mark set, call the offset data of the measurement points marked with consistent direction, perform numerical accumulation operation on the offset within the same rigid transmission path, and record the result of each superposition sequentially during the accumulation process to obtain the accumulated offset sequence.

[0080] Based on the generated offset direction consistency marker set, the numerical accumulation engine is activated. First, it retrieves consecutive measurement point segments marked as "directionally consistent" from the sequence. For consecutive consistent measurement points within the same rigid transmission path (such as a suspended section or a supported section), a numerical accumulation operation is performed. This operation is not a simple positional superposition but an integral processing of the "deformation increment." The offset difference (i.e., increment) of each measurement point relative to the previous measurement point is extracted and superimposed into the current accumulation register in both temporal and spatial order. During the accumulation process, the intermediate results of each step are recorded in real time, forming a dynamically changing sequence of accumulated offsets. This sequence reflects the pure cumulative deformation shape of the guide rail within this specific rigid section due to gravity or internal stress after excluding abrupt disturbances. If a node marked as "directionally inconsistent" is encountered, accumulation is paused at the current node, and the current accumulated value is archived for subsequent truncation processing. Measurement points 1 to 3 are determined to be directionally consistent. The increment for measuring point 1 is +5 micrometers (relative to 0), the increment for measuring point 2 is +7 micrometers (12 minus 5), and the increment for measuring point 3 is +16 micrometers (28 minus 12). Accumulation is performed: first step records 5, second step records 12 (5 plus 7), and third step records 28 (12 plus 16). The result of 28 micrometers is the geometric cumulative deformation under the current path. The advantage of this calculation logic is that, through incremental integral reconstruction, the true physical contour of the guide rail under continuous stress is accurately reconstructed, providing a distortion-free geometric data foundation for subsequent error analysis.

[0081] S303: For the accumulated offset sequence, call the index position of the measurement point marked as inconsistent in direction, mark the state segment of the corresponding measurement point in the rigid transfer path, and restart the accumulation of the accumulated record after the marked measurement point with the current point as the reference to generate rigid transfer data.

[0082] For the generated cumulative offset sequence, the offset direction consistency marker set is called again to accurately locate the index positions of all measuring points marked as "inconsistent direction". These index positions are defined as "rigid break points", that is, deformation mode switching points in a physical sense. For each rigid break point, the state of the corresponding measuring point in the rigid transfer path is canceled, that is, the measuring point is determined to no longer belong to the previous continuous deformation band, and it and all subsequent cumulative records are removed from the current band, and truncation processing is performed. The truncated data segments are repackaged, retaining only the data segments that conform to the characteristics of continuous physical deformation, generating the final rigid transfer data. This data removes high-frequency noise caused by machining errors (such as single-point deviation of measuring point 5) or assembly abrupt changes, and retains only the low-frequency deformation components that reflect the structural characteristics of the bed. A comprehensive analysis is performed in conjunction with the implementation data shown in Table 2. Measuring point 4 is identified as a rigid break point because the direction change with measuring point 3 exceeds the threshold (21.1 degrees is greater than 20 degrees). A cutoff was performed at measuring point 4, and the cumulative sequence consisting of measuring points 1 to 3 (with a final value of 28 micrometers) was determined to be a valid "suspended sagging deformation segment". The data at measuring point 4 and its subsequent data were isolated and processed, and a new rigid segment analysis needs to be started.

[0083] Table 2. Analysis of the Consistency of Direction and Rigidity Transmission of Guide Rail Measuring Points

[0084]

[0085] As shown in Table 2, the table clearly illustrates the logical flow from vector-based calculation to the final truncation determination. The "inconsistent direction" determination at measurement point 4 directly triggers the truncation operation in S303, effectively distinguishing the continuous sagging deformation before measurement point 3 from the subsequent fluctuations. This ensures that the rigid transfer data only contains structural deformation information with physical continuity, providing a clean data sample for subsequent error tracing.

[0086] Please see Figure 5 The specific steps of S4 are as follows:

[0087] Based on rigid transfer data, the cumulative offset of multiple measuring points within the same structural segment is obtained. The adjacent differences of the cumulative offset are calculated according to the spatial order of the measuring points, combined into a change sequence, and the smoothness of the change trend is judged. Measuring points exceeding the preset smooth change threshold are filtered out, and a continuous index set of cumulative offset is generated.

[0088] Based on the generated rigidity transfer data, a set of valid measuring points within the same structural segment (e.g., the "suspended sagging deformation segment") is identified. The cumulative offset values ​​of these measuring points in the spatial sequence are extracted, and adjacent difference calculations are performed. This involves subtracting the cumulative offset of the previous measuring point from the cumulative offset of the subsequent measuring point to obtain the adjacent difference value characterizing the local deformation slope. All adjacent difference values ​​are arranged in spatial order to form a change sequence, and then a sequence consistency judgment logic is initiated. This judgment logic aims to verify the monotonicity of the deformation trend. Specifically, it checks whether the signs of consecutive terms in the change sequence are the same (i.e., all positive or all negative). If the signs of adjacent difference values ​​of three or more consecutive measuring points are consistent, these measuring points are determined to be in a stable monotonic deformation state, and their index numbers are recorded. If the signs are reversed, a monotonicity tolerance judgment is entered. In this process, a "monotonicity tolerance counter" is introduced to handle measurement noise. If a single small negative fluctuation occurs within a continuous positive change, and the absolute value of this fluctuation is less than 10% of the mean of the preceding differences, the counter ignores the reversal and maintains the monotonicity determination; otherwise, it is determined as a valid non-monotonic change and the continuous indexing is terminated. Measurement points 1 to 5 are set within the "suspended section," with accumulated offsets of 0 micrometers, 4 micrometers, 9 micrometers, 13 micrometers, and 12 micrometers, respectively. Adjacent differences are calculated: measurement point 2 is 4 micrometers relative to measurement point 1 (4 minus 0), measurement point 3 is 5 micrometers relative to measurement point 2 (9 minus 4), measurement point 4 is 4 micrometers relative to measurement point 3 (13 minus 9), and measurement point 5 is -1 micrometer relative to measurement point 4 (12 minus 13). The first three difference values ​​(4, 5, 4) are all positive, conforming to the monotonically increasing law of drooping deformation; therefore, the indices of measurement points 1, 2, 3, and 4 are included in the set. Regarding the -1 micrometer change observed at measurement point 5, the mean of the preceding difference was calculated to be 4.33 micrometers (i.e., 4 plus 5 plus 4 divided by 3), and 10% of this is 0.433 micrometers. Since the absolute value of -1 micrometer is greater than 0.433 micrometers, this reversal is determined to exceed the fault tolerance range, therefore the continuous indexing is terminated. The final generated cumulative offset continuous index set is {1, 2, 3, 4}.

[0089] S402: Based on the cumulative offset continuous index set, call the cumulative offset change sequence of the corresponding measurement point, mark the measurement point index whose change amplitude exceeds the preset continuous change judgment threshold as abnormal index, and remove the abnormal index from the continuous index set to generate a continuous offset index set;

[0090] Based on the generated cumulative offset continuous index set, the cumulative offset change sequence of the corresponding measuring points (i.e., the previously calculated adjacent difference values) is called to perform a safety check on the deformation rate. Each adjacent difference value (representing the deformation amplitude within a unit spacing) is numerically compared with a preset continuous change judgment threshold. If the adjacent difference value of a measuring point is less than or equal to the threshold, it is determined that the deformation of that point is within the elastic range of the material, and its index is retained; if the adjacent difference value is greater than the threshold, it is determined that an unnatural abrupt change has occurred at that point (such as foreign object elevation or sensor jump), it is marked as an abnormal index, and a removal operation is performed from the continuous index set. By traversing all indices, the cleaning work is completed, and a clean continuous offset index set is generated. In this process, the setting of the continuous change judgment threshold is determined based on the allowable bending strain rate of the bed material. The ultimate elastic strain value of the cast iron material is obtained as 0.0002. Combined with the measuring point spacing of 50 mm, the theoretical maximum deformation is calculated to be 0.01 mm (i.e., 10 micrometers) through multiplication (0.0002 multiplied by 50). Considering a safety factor of 1.5 for actual working conditions, the threshold for continuous variation is set to 15 micrometers. Continuing the previous example, in the index set {1, 2, 3, 4}, the corresponding adjacent difference value sequence is 4 micrometers, 5 micrometers, and 4 micrometers. Assume there is an artifact data measurement point X with an adjacent difference value as high as 18 micrometers. Verification is performed: 4 micrometers, 5 micrometers, and 4 micrometers are all less than 15 micrometers, so the judgment is qualified. However, for measurement point X with a value of 18 micrometers, because it is greater than 15 micrometers, it is immediately identified as an abnormal index and removed. This logic ensures that the final retained data is not only monotonic in direction but also conforms to the material mechanical properties in amplitude. The final generated continuous offset index set only contains smooth deformation points that conform to physical laws.

[0091] S403: For a continuous offset index set, call the corresponding cumulative offset data of the measurement points, call the corresponding cumulative offset of the measurement points, perform re-aggregation and interpolation connection on the data items corresponding to the index, map the aggregated offset data with the measurement point identifier, and generate configuration offset data.

[0092] For the generated continuous offset index set, final data reconstruction and encapsulation are performed. First, based on the cleaned index list, the corresponding cumulative offset values ​​and spatial coordinates of the measurement points are extracted from the original database. Then, these scattered data items are re-aggregated and rearranged according to the spatial order of the guide rail from beginning to end, eliminating gaps caused by the removal operation and ensuring the continuity of the storage structure. Finally, a permanent mapping relationship is established between the aggregated offset data and the physical identifier of the measurement point (such as a QR code ID or laser engraving number), generating configuration offset data that can be directly read by CNC. The implementation data shown in Table 3 is used for illustration. The effective data after S401 monotonicity filtering and S402 amplitude cleaning are integrated. For example, the adjacent difference value of measurement point 2 is 4.0 micrometers; after superimposing the baseline value of measurement point 1 (0), its cumulative configuration offset is 4.0 micrometers; the adjacent difference value of measurement point 3 is 5.0 micrometers; after superimposing the value of measurement point 2, its cumulative configuration offset is 9.0 micrometers. Measurement point 5 is removed because it is judged to have abnormal monotonicity in S401. The final generated configuration data is shown in Table 3.

[0093] Table 3 Final Table of Guide Rail Configuration Offset Data

[0094]

[0095] Please see Figure 6 The specific steps of S5 are as follows:

[0096] S501: Based on the configuration offset data, extract the cumulative offset vector components between the end measuring point and the starting measuring point of adjacent bed structure sections, perform sign consistency judgment on the multi-vector components, and summarize the judgment results to generate a section connection direction consistency mark set;

[0097] Based on the generated configuration offset data, the boundary nodes of different bed structure sections (such as the main guide rail section and auxiliary guide rail section of the machine tool, or the connection point of a spliced ​​bed) are identified. The set of end measurement points of the preceding structural section (e.g., the last two measurement points) and the set of starting measurement points of the following structural section (e.g., the first two measurement points) are extracted to construct a cross-section boundary vector set. Vector component calculation is performed on each pair of adjacent measurement points, extracting the longitudinal offset difference as the vertical component and the transverse coordinate spacing as the horizontal component, and calculating the ratio of the two to obtain the local rate of change. Subsequently, a sign consistency judgment is performed to check whether the local rates of change on both sides of the boundary have the same positive or negative sign. If the rate of change at the end of the preceding section is positive (representing an upward trend), and the rate of change at the beginning of the following section is also positive, the deformation trend at the connection point is determined to be continuous and marked as "connection direction consistent"; if opposite signs appear (e.g., one positive and one negative, forming a "V" shape or an inverted "V" shape inflection point), it is determined as "connection direction inconsistent". Finally, the judgment results for all connections are summarized. In this process, a "trend dead zone threshold" is set to filter out minor fluctuations; this threshold is set to 2 micrometers per meter. If the absolute value of the calculated rate of change is less than this threshold, it is considered a horizontal extension regardless of the sign, and no inconsistency flag is triggered. Taking the splicing of two long guide rails as an example, the coordinates of the last two measuring points at the end of the preceding segment A are (1950 mm, +45 micrometers) and (2000 mm, +48 micrometers), respectively. The calculated vertical component is +3 micrometers, the horizontal component is 50 mm, and the rate of change is +0.06 micrometers per millimeter (i.e., 60 micrometers per meter). The coordinates of the starting measuring points of the following segment B (recounted relative to the splicing point) are (2000 mm, 0 micrometers) and (2050 mm, +2 micrometers). Note that the original data of segment B is usually zeroed out. If the calculated rate of change of the first segment vector of segment B is +0.04 micrometers per millimeter (i.e., 40 micrometers per meter). Comparing the two signs, both are positive and their absolute values ​​are much greater than 2 micrometers / meter, indicating that the splicing joint has a consistent trend and no abrupt folding has occurred.

[0098] S502: Based on the segment connection direction consistency mark set, perform sequence splicing on the adjacent segment measurement point indices that are marked as consistent, and perform order mapping and index rearrangement on the spliced ​​cumulative offset sequence to generate a continuous sequence of cross-segment cumulative offset;

[0099] Based on the generated segment connection direction consistency marker set, the sequence splicing logic is initiated to process the data of adjacent segments marked as "connected consistently". First, the final cumulative offset value of the preceding segment is locked and used as the "baseline offset value". Then, all measurement point offset data of the subsequent segment are read, and a baseline superposition operation is performed on each data item, that is, the "baseline offset value" is added to each original offset of the subsequent segment, thereby achieving physical unification of the coordinate system and eliminating zero-point inconsistencies caused by segmented measurements. After numerical alignment is completed, a rearrangement operation is performed on all measurement point indices, adding the index value of the subsequent segment to the total number of measurement points in the preceding segment to generate a globally unique continuous index sequence. The aligned offset values ​​are bound to the rearranged indices to generate a continuous sequence of cumulative offsets across segments. The preceding segment A is set to contain 40 measurement points, and the cumulative offset of the 40th measurement point (located at 2000 mm) is +48 micrometers. The subsequent segment B contains 40 measuring points. The original reading of its first measuring point (physically adjacent to segment A) is 0 micrometers, and the original reading of the second measuring point is +2 micrometers. The splicing process is as follows: the offset of the first measuring point in segment B is corrected to 48 micrometers (0 plus 48), the second measuring point is corrected to 50 micrometers (2 plus 48), and so on. Simultaneously, the index of the first measuring point in segment B is remapped to number 41, and the index of the second measuring point is remapped to number 42.

[0100] S503: For the continuous sequence of cumulative offset across sections, perform linear fitting residual calculation on the coordinates of the corresponding measuring points and the cumulative offset within the multi-bed structure section, and serialize and aggregate the residual values ​​of multiple measuring points to generate machine tool assembly accuracy detection results.

[0101] For the generated continuous sequence of cumulative offsets across segments, the core calculation for global straightness evaluation—least squares line fitting—is performed. The global coordinates of all measurement points in the sequence are extracted as the independent variable sequence, and the corresponding aligned cumulative offsets are extracted as the dependent variable sequence. First, the arithmetic mean of the independent and dependent variable sequences are calculated to determine the geometric center of the data. Then, the following operations are performed: the difference between the independent variable value and the mean of the independent variables at each measurement point (independent variable deviation), and the difference between the dependent variable value and the mean of the dependent variables (dependent variable deviation); the sum of the products of the independent and dependent variable deviations (sum of deviation products); and the sum of the squares of the independent variable deviations (sum of squared deviations). Dividing the sum of deviation products by the sum of squared deviations yields the slope parameter of the fitted line; subtracting the product of the slope and the mean of the independent variables from the mean of the dependent variable yields the intercept parameter of the fitted line. After obtaining the fitted line equation, each measurement point is iterated through, and the difference between its actual offset and the corresponding theoretical value of the fitted line is calculated; this difference is the "residual". All residual values ​​are serialized and aggregated, and the largest positive residual and the smallest negative residual are identified. The difference between the two is calculated as the final evaluation index of the machine tool assembly accuracy, generating the machine tool assembly accuracy inspection result. The calculation is explained using the implementation data shown in Table 4. The merged sequence contains 5 key measurement points, with coordinates X as {0, 500, 1000, 1500, 2000} (unit: mm) and cumulative offset Y as {0, 10, 25, 38, 48} (unit: micrometers). First, the arithmetic mean of X is calculated to be 1000, and the arithmetic mean of Y is 24.2. Calculate the sum of the products of deviations: (0-1000)*(0-24.2)+(500-1000)*(10-24.2)+(1000-1000)*(25-24.2)+(1500-1000)*(38-24.2)+(2000-1000)*(48-24.2)=24200+7100+0+6900+23800=62000. Calculate the sum of squares of the deviations of the independent variable: the square of (-1000) plus the square of (-500) plus the square of 0 plus the square of 500 plus the square of 1000, the result is 2500000. Calculate the slope: 62000 / 2500000=0.0248. Calculate the intercept: 24.2 - (0.0248 * 1000) = -0.6. The fitted equation is: y = 0.0248x − 0.6. Calculate the residuals: Point 1 (0 mm): 0 - (-0.6) = +0.6. Point 2 (500 mm): 10 - (0.0248 * 500 - 0.6) = 10 - 11.8 = -1.8. Point 3 (1000 mm): 25 - (24.8 - 0.6) = +0.8. Point 4 (1500 mm): 38 - (37.2 - 0.6) = +1.4. Point 5 (2000 mm): 48 - (49.6 - 0.6) = -1.0.The maximum positive residual is +1.4, and the minimum negative residual is -1.8. The final accuracy index (range) is 1.4 - (-1.8) = 3.2 micrometers. The experimental results show that the straightness error of the 2-meter-long guide rail of the machine tool is 3.2 micrometers, which is better than the 5-micrometer accuracy level specified in the national standard GB / T17421.1, and the assembly is judged to be "qualified".

[0102] Table 4. Detailed Results of Machine Tool Assembly Accuracy Inspection

[0103]

[0104] Please see Figure 7 Machine tool assembly precision testing system, including:

[0105] The measurement point analysis module uses a laser detection instrument to collect displacement signals of measurement points arranged along the length of the guide rail under the machine tool bed guide rail assembly, performs analog-to-digital conversion, calculates the spatial offset between the measurement point and the laser reference line, generates measurement point offset data, and transmits it to the rigid mapping module.

[0106] The rigid mapping module reads the bed structure information and identifies the connection sequence of the stiffeners and guide rails, analyzes the rigid topology of the bed assembly, determines the rigid transmission direction, marks the rigid attributes and sections of the measurement point offset data, maps them to the corresponding bed structure, generates the measurement point rigid data, and transmits it to the transmission accumulation module.

[0107] The transmission and accumulation module, based on the rigid data of the measuring points, judges the direction consistency of the offset of adjacent measuring points arranged along the rigid transmission direction. If the directions are consistent, the offset vector is accumulated. If they are inconsistent, the rigid deformation inflection point is marked and segmented, generating rigid transmission data and transmitting it to the continuous filtering module.

[0108] The continuous screening module, based on rigid transmission data, performs smoothness analysis on the cumulative offset change relationship of measuring points in the multi-bed structure section, filters out measuring points that exceed the preset smooth change threshold, generates configuration offset data and transmits it to the straight line detection module.

[0109] The straight line detection module analyzes the consistency of the offset direction between the end of the adjacent bed structure section and the starting measuring point based on the configured offset data. If they are consistent, the cumulative offset result is continued and the straightness offset is calculated to generate the machine tool assembly accuracy detection result.

[0110] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting the assembly accuracy of machine tools, characterized in that, Includes the following steps: S1: The displacement signal of the measuring points arranged along the length of the guide rail under the machine tool bed guide rail assembly is collected by the laser detection instrument and the analog-to-digital conversion is performed to calculate the spatial offset between the measuring points and the laser reference line and generate the measuring point offset data. S2: Read the bed structure information and identify the connection sequence of the stiffener and guide rail, analyze the rigid topology of the bed assembly, determine the rigidity transmission direction, mark the rigidity attribute and section mapping of the measurement point offset data, and map it with the corresponding bed structure to generate measurement point rigidity data. S3: Based on the rigidity data of the measuring points, the offset of adjacent measuring points arranged along the rigidity transmission direction is used to determine the consistency of the direction. If the direction is consistent, the offset vector is accumulated. If they are inconsistent, the rigid deformation inflection point is marked and segmented to generate rigidity transmission data. S4: Based on the rigid transmission data, perform smoothness analysis on the cumulative offset change relationship of the measuring points in the multi-bed structure section, filter out measuring points that exceed the preset smooth change threshold, and generate configuration offset data; S5: Based on the configured offset data, analyze the consistency of the offset direction between the end of the adjacent bed structure section and the starting measuring point. If they are consistent, continue the cumulative offset result and perform straightness offset calculation to generate machine tool assembly accuracy test results.

2. The machine tool assembly accuracy detection method according to claim 1, characterized in that, The measurement point offset data includes the measurement point spatial coordinates, the laser reference line offset vector, and the guide rail length sequence index. The measurement point rigidity data includes the rigid section identifier, the measurement point spatial sequence index, and the section structure mapping relationship. The rigid transmission data includes the continuously accumulated offset vector, the segment position index, and the rigid transmission direction identifier. The configuration offset data includes the continuous measurement point index set, the cumulative offset sequence within the section, and the continuity judgment mark. The machine tool assembly accuracy detection results include the bed straightness offset sequence and the section splicing offset residual.

3. The machine tool assembly accuracy detection method according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: The displacement signals of multiple measuring points arranged along the length of the guide rail under the machine tool bed guide rail assembly are collected by a laser detection instrument. The continuous displacement signals are sampled at equal intervals with a position sensor. The sampled analog voltage is mapped into discrete digital quantities to generate a digital sequence of measuring point displacements. S102: Based on the digital sequence of displacement of the measuring points, call the spatial parameters of the laser reference line, perform numerical difference calculation on the displacement values ​​of multiple measuring points and the theoretical positions of the laser reference line in the same coordinate system, and vectorize the difference results to generate a set of spatial offset of measuring points. S103: Based on the set of spatial offsets of the measuring points, establish a serialized sort according to the physical location coordinates of the measuring points, maintain the spatial distribution relationship of the guide rail shape, and integrate the sorting results with the corresponding measuring point offsets to generate measuring point offset data.

4. The machine tool assembly accuracy detection method according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Based on the measured point offset data, obtain the bed structure design information, align the stiffener plate number, stiffener plate spatial position parameters and guide rail installation position parameters, and perform sequence determination according to the structural connection sequence to transform the connection relationship between the stiffener plate and the guide rail, generating a stiffener plate and guide rail connection sequence. S202: Based on the stiffener guide rail connection sequence, call the connection topology of multi-stiffener nodes and guide rail nodes in the bed structure design information, judge the constraint relationship between adjacent connection nodes, and perform direction determination according to the constraint transmission path to generate a rigid transmission direction sequence. S203: For the rigid transmission direction sequence, the original spatial order of the measurement point offset data is maintained, and the segments are divided according to the transmission direction change nodes. The corresponding segments of the measurement point sequence are mapped to the corresponding bed structure segments to generate measurement point rigid data.

5. The machine tool assembly accuracy detection method according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Based on the rigid data of the measuring points, detect the offset direction vectors corresponding to multiple measuring points, perform vector consistency judgment on the offset directions of adjacent measuring points, compare the direction angle with the preset offset direction consistency threshold, write the judgment result into the measuring point sequence identifier sequence, and generate an offset direction consistency tag set. S302: Based on the offset direction consistency mark set, call the offset data of the measurement points marked with consistent directions, perform numerical accumulation operation on the offsets within the same rigid transmission path, and record the stacking results sequentially during the accumulation process to obtain the accumulated offset sequence. S303: For the accumulated offset sequence, call the index position of the measurement point marked as inconsistent in direction, mark the state segment of the corresponding measurement point in the rigid transmission path, and restart the accumulation record after the marked measurement point with the current point as the reference to generate rigid transmission data.

6. The machine tool assembly accuracy detection method according to claim 5, characterized in that, The offset direction consistency threshold is determined by statistically analyzing the angle difference between the offset direction vectors of adjacent measuring points, obtaining the distribution sequence of the angle difference, calculating the median and standard deviation of the distribution sequence of the angle difference, and taking the sum of the median and three times the preset standard deviation.

7. The machine tool assembly accuracy detection method according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Based on the rigid transmission data, obtain the cumulative offset of multiple measuring points within the same structural segment, calculate the adjacent difference of the cumulative offset according to the spatial order of the measuring points, combine them into a change sequence and judge the smoothness of the change trend, filter out measuring points that exceed the preset smooth change threshold, and generate a continuous index set of cumulative offset. S402: Based on the cumulative offset continuous index set, call the cumulative offset change sequence of the corresponding measurement point, mark the measurement point index whose change amplitude exceeds the preset continuous change judgment threshold as an abnormal index, and remove the abnormal index from the continuous index set to generate a continuous offset index set; S403: For the continuous offset index set, call the corresponding cumulative offset data of the measurement points, call the corresponding cumulative offset of the measurement points, perform re-aggregation and interpolation connection on the data items corresponding to the index, map the aggregated offset data with the measurement point identifier, and generate configuration offset data.

8. The machine tool assembly accuracy detection method according to claim 7, characterized in that, The continuous change judgment threshold is determined by statistically analyzing the sequence of adjacent changes in the cumulative offset of measurement points within the same structural segment, obtaining the distribution characteristics of the change amplitude, calculating the median and dispersion index of the distribution characteristics, and then weighting and summing the median and dispersion index.

9. The machine tool assembly accuracy detection method according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: Based on the configured offset data, extract the cumulative offset vector components between the end measuring point and the starting measuring point of the adjacent bed structure section, perform sign consistency judgment on the multi-vector components, and summarize the judgment results to generate a section connection direction consistency mark set. S502: Based on the segment connection direction consistency mark set, perform sequence splicing on the adjacent segment measurement point indices that are marked as consistent, and perform order mapping and index rearrangement on the spliced ​​cumulative offset sequence to generate a cross-segment cumulative offset continuous sequence; S503: For the continuous sequence of cumulative offset across sections, perform linear fitting residual calculation on the coordinates of the corresponding measuring points and the cumulative offset within the multi-bed structure section, and serialize and aggregate the residual values ​​of multiple measuring points to generate machine tool assembly accuracy detection results.

10. A machine tool assembly accuracy detection system, characterized in that, The system is used to implement the machine tool assembly accuracy detection method according to any one of claims 1-9, the system comprising: The measurement point analysis module uses a laser detection instrument to collect displacement signals of measurement points arranged along the length of the guide rail under the machine tool bed guide rail assembly, performs analog-to-digital conversion, calculates the spatial offset between the measurement point and the laser reference line, generates measurement point offset data, and transmits it to the rigid mapping module. The rigid mapping module reads the bed structure information and identifies the connection sequence of the stiffeners and guide rails, analyzes the rigid topology of the bed assembly, determines the rigid transmission direction, marks the rigid attributes and section mapping of the measurement point offset data, maps it with the corresponding bed structure, generates the measurement point rigid data, and transmits it to the transmission accumulation module. The transmission and accumulation module, based on the rigid data of the measurement points, performs a direction consistency judgment on the offset of adjacent measurement points arranged along the rigid transmission direction. If the directions are consistent, the offset vector is accumulated. If they are inconsistent, the rigid deformation inflection point is marked and segmented, generating rigid transmission data and transmitting it to the continuous filtering module. The continuous screening module, based on the rigid transmission data, performs a smoothness analysis on the cumulative offset change relationship of the measuring points in the multi-bed structure section, filters out measuring points that exceed the preset smooth change threshold, generates configuration offset data, and transmits it to the straight line detection module. The straight line detection module analyzes the consistency of the offset direction between the end of the adjacent bed structure section and the starting measuring point based on the configured offset data. If they are consistent, the cumulative offset result is continued and the straightness offset is calculated to generate the machine tool assembly accuracy detection result.

Images