Industrial equipment operation analysis method and system based on machine vision

Abstract

The invention relates to the technical field of machine vision, in particular to an industrial equipment operation analysis method and system based on machine vision. According to the method, the track information of the guide rail of the numerical control machine tool is obtained, and the three-dimensional coordinates and the timestamp sequence of the track points are established, so that the dynamic change of the track of the guide rail can be accurately described, the key track points are extracted through density analysis, and it is ensured that the perception of the track form change is more targeted. The time window division of the trajectory data is combined with the form change amplitude calculation, so that the quantification of the trajectory deviation trend is realized, and the trajectory anomaly can be quickly identified. In combination with stress analysis of a track deviation overrun area, an expansion path of fatigue damage can be determined, and a visual description of a damage propagation process is formed. Through extraction and analysis of a guide rail surface visual image, features of local cracks, wear and deformation can be matched with track anomalies, a visual segmentation boundary is further optimized, and it is ensured that details of structure changes are accurately extracted.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision, and particularly to an industrial equipment operation analysis method and system based on machine vision. Background Art

[0002] The technical field of machine vision involves using computers and image processing technologies to collect, analyze, and process image or video data to extract useful information and perform corresponding control or decision-making. The core content of this technical field includes links such as image acquisition, preprocessing, feature extraction, target recognition, classification, and measurement, and is widely applied in scenarios such as industrial automation, quality inspection, target tracking, and intelligent monitoring.

[0003] Among them, the industrial equipment operation analysis method refers to detecting, identifying, and analyzing the operation state of industrial equipment based on machine vision technology to extract key operation parameters and perform fault early warning or state assessment. This method addresses problems such as abnormal motion trajectories, changes in vibration characteristics, component wear, and temperature deviations that may occur during the operation of industrial equipment. It obtains image or video data of the equipment's working state through high-speed cameras or industrial cameras, and analyzes the operation state of the equipment using methods such as image segmentation, edge detection, and feature matching.

[0004] The operation analysis of existing industrial equipment mainly relies on image processing means to detect motion trajectories, vibration characteristics, and component wear conditions, but lacks time series analysis of trajectory data, resulting in a lag in the judgment of trajectory change trends. It can only rely on static images to analyze a single state and cannot establish the dynamic evolution process of trajectory deviation. The force condition of the equipment is not combined with the trajectory deviation, making it difficult to infer the formation mechanism of fatigue damage. It can only be speculated based on surface wear characteristics, and it is difficult to accurately track the propagation path of damage. The detection of structural changes such as cracks and wear lacks analysis based on the logic of damage propagation, resulting in the setting of visual segmentation boundaries being easily affected by image noise, making it difficult to accurately identify tiny cracks or minor wear, and affecting the accuracy of fault detection. Since trajectory analysis only stays at the comparison of static forms, it is impossible to comprehensively analyze the connections between trajectory anomalies, fatigue damage, and structural changes, making the screening of abnormal states rely on single-point features, which may lead to misjudgment or omission of potential risks and affect the reliability of early warning effects. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose an industrial equipment operation analysis method and system based on machine vision.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: An industrial equipment operation analysis method based on machine vision, comprising the following steps:

[0007] S1: Obtain the trajectory of the numerical control machine tool guide rail under the current machining task, construct the three-dimensional coordinates and timestamps of each trajectory point, extract the key trajectory points during the operation of the guide rail, and obtain the guide rail trajectory distribution data;

[0008] S2: Divide the guide rail trajectory distribution data according to a fixed time window, calculate the morphological change amplitude of the trajectory points within adjacent time windows, judge the offset trend of the trajectory points in space, and obtain the trajectory deviation analysis result;

[0009] S3: Identify the trajectory regions in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtain the force condition of the guide rail within the region, judge the propagation path of the guide rail fatigue damage, and obtain the damage path analysis result;

[0010] S4: Based on the damage path analysis result, extract the corresponding visual images of the guide rail surface on the fatigue damage propagation path, analyze the local crack, wear and deformation characteristics of the guide rail surface and adjust the visual segmentation boundary, and generate the guide rail structure change analysis result;

[0011] S5: Analyze the correlation between the guide rail trajectory deviation, fatigue damage propagation path and structure change region in the guide rail structure change analysis result, screen the regions where the guide rail operation state is unstable, and obtain the guide rail abnormal operation analysis result.

[0012] As a further solution of the present invention, the guide rail trajectory distribution data includes a set of three-dimensional coordinates of trajectory points, a sequence of trajectory point timestamps, trajectory point density distribution information, and a set of key trajectory points. The trajectory deviation analysis result includes time window division information, trajectory morphological change amplitude, trajectory point spatial offset trend, and global trajectory morphological deviation. The damage path analysis result includes trajectory deviation overrun regions, guide rail force distribution information, stress concentration regions, and fatigue damage propagation paths. The guide rail structure change analysis result includes guide rail surface crack characteristics, guide rail surface wear regions, guide rail deformation characteristics, and visual segmentation boundary adjustment information. The guide rail abnormal operation analysis result includes trajectory deviation abnormal regions, fatigue damage expansion regions, structure change risk regions, and warning region marking information.

[0013] As a further solution of the present invention, the obtaining steps of the guide rail trajectory distribution data are specifically as follows:

[0014] S111: Obtain the trajectory of the numerical control machine tool guide rail under the current machining task, collect the spatial position information of the trajectory points, including three-dimensional coordinates and timestamps, analyze the collected trajectory point sequence, and determine the arrangement trend of the trajectory points in the time series to obtain the trajectory point time series sorting result;

[0015] S112: Based on the trajectory point time series sorting result, use the formula:

[0016]

[0017] Calculate the distribution density D of the trajectory point i i , and obtain the trajectory point density information;

[0018] Among them, P i represents the spatial position of the trajectory point i, and P j represents the spatial position of the trajectory point j, represents the spatial distribution influence factor between the trajectory point i and the trajectory point j, represents the time influence factor between the trajectory point i and the trajectory point j, and σ p represents the spatial coordinate normalization factor, and σ t represents the time normalization factor, and n represents the total number of trajectory points within the selected window;

[0019] S113: Based on the trajectory point density information, screen the trajectory points with distribution density values higher than the set density threshold, extract the key trajectory points during the operation of the guide rail, and count the spatial distribution range of the key trajectory points to generate the guide rail trajectory distribution data.

[0020] As a further solution of the present invention, the steps for obtaining the trajectory deviation analysis result are specifically as follows:

[0021] S211: Divide the guide rail trajectory distribution data according to a fixed time window, and use the formula:

[0022]

[0023] Calculate the average Euclidean distance change amount Δd of the trajectory points within the time window h h , and the average direction angle change amount Δθ of the trajectory points within the time window h h , to obtain the morphological change distribution data;

[0024] Among them, represents the Euclidean distance of the j1th trajectory point relative to the reference point within the time window h, represents the Euclidean distance of the j1th trajectory point relative to the reference point within the time window h + 1, represents the direction angle of the j1th trajectory point within the time window h, represents the direction angle of the j1th trajectory point within the time window h + 1, and n1 represents the total number of trajectory points within the time window;

[0025] S212: Analyze the spatial distribution pattern of the trajectory points based on the morphological change distribution data, determine the local offset direction of the trajectory points in adjacent time windows, and judge the overall offset trend of the trajectory points according to the spatial distribution density of the trajectory points and the continuous change of the local offset direction. Screen the trajectory points with abnormal morphological change rates to obtain abnormal trajectory point information;

[0026] S213: Based on the abnormal trajectory point information, use the formula:

[0027]

[0028] Calculate the global morphological deviation D of all abnormal trajectory points global , and obtain the trajectory deviation analysis result;

[0029] where m1 represents the total number of all abnormal trajectory points, represents the spatial coordinates of the k1-th abnormal trajectory point, X mean , Y mean , Z mean represent the spatial mean coordinates of all trajectory points.

[0030] As a further solution of the present invention, the steps for obtaining the damage path analysis result are specifically as follows:

[0031] S311: Identify the trajectory area in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtain the force condition of the guide rail in the area, and use the formula;

[0032]

[0033] Calculate the stress of the k2-th contact point in the trajectory area to obtain the guide rail stress distribution data;

[0034] where, represents the force on the k2-th contact point in the trajectory area, represents the contact area of the k2-th contact point in the trajectory area;

[0035] S312: Based on the guide rail stress distribution data, obtain the area where the stress is concentrated, as well as the average stress value in the contact point area where the stress is concentrated, analyze the spatial distribution of the stress concentration area, and judge the propagation path of the guide rail fatigue damage and the direction and range of damage expansion to obtain the damage path analysis result.

[0036] As a further solution of the present invention, the steps for obtaining the guide rail structure change analysis result are specifically as follows:

[0037] S411: Based on the damage propagation data in the damage path analysis results, extract the trajectory coordinates of the damaged area and set the visual image acquisition range. According to the acquisition range, collect images of the guide rail surface, extract the continuous surface visual images of the damaged area, and obtain the visual image data of the guide rail surface;

[0038] S412: Based on the visual image data of the guide rail surface, use the formula:

[0039]

[0040] Calculate the overall structural integrity index S of the target area c ;

[0041] where, represents the gray value of the i1-th pixel point, represents the edge gradient value of the i1-th pixel point, with the unit being dimensionless, and N represents the total number of pixel points in the target area;

[0042] S413: Based on the structural integrity data of the guide rail surface, analyze the local crack, wear, and deformation characteristics of the guide rail surface. According to the morphological damage amount of the deformation characteristics, adjust the visual segmentation boundary and optimize the boundary position to obtain the analysis result of the guide rail structure change.

[0043] As a further solution of the present invention, the steps for obtaining the analysis result of the abnormal operation of the guide rail are specifically as follows:

[0044] S511: Analyze the correlation between the guide rail trajectory deviation, fatigue damage propagation path, and structural change area in the analysis result of the guide rail structure change to obtain the trajectory damage correlation data;

[0045] S512: Based on the trajectory damage correlation data, screen the areas where the operation state of the guide rail is unstable and mark them as warning areas to obtain the analysis result of the abnormal operation of the guide rail.

[0046] An industrial equipment operation analysis system based on machine vision, the industrial equipment operation analysis system based on machine vision is used to execute the above-mentioned industrial equipment operation analysis method based on machine vision, and the system includes:

[0047] The trajectory data construction module obtains the trajectory of the numerical control machine tool guide rail under the current machining task, constructs the three-dimensional coordinates and timestamps of each trajectory point, and extracts the key trajectory points during the operation of the guide rail to obtain the guide rail trajectory distribution data;

[0048] The trajectory deviation calculation module divides the guide rail trajectory distribution data according to a fixed time window, calculates the morphological change amplitude of the trajectory points within adjacent time windows, and judges the offset trend of the trajectory points in space to obtain the trajectory deviation analysis result;

[0049] The fatigue damage identification module identifies the trajectory regions in the trajectory deviation analysis results where the deviation exceeds the deviation threshold, obtains the force conditions of the guide rails within the regions, determines the propagation path of the fatigue damage of the guide rails, and obtains the damage path analysis results;

[0050] Based on the damage path analysis results, the guide rail structure change detection module extracts the visual images of the corresponding guide rail surfaces on the fatigue damage propagation path, analyzes the characteristics of local cracks, wear, and deformation on the guide rail surfaces, and adjusts the visual segmentation boundaries to generate the guide rail structure change analysis results;

[0051] The abnormal operation region identification module analyzes the correlation between the guide rail trajectory deviation, the fatigue damage propagation path, and the structure change region in the guide rail structure change analysis results, screens out the regions where the operation state of the guide rails is unstable, and obtains the guide rail abnormal operation analysis results.

[0052] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0053] In the present invention, by obtaining the trajectory information of the guide rails of a numerically controlled machine tool, establishing the three-dimensional coordinates and timestamp sequences of the trajectory points, the dynamic changes of the guide rail trajectories can be accurately depicted, and key trajectory points are extracted through density analysis to ensure more targeted perception of the changes in the trajectory form. The combination of the time window division of the trajectory data and the calculation of the morphological change amplitude realizes the quantification of the trajectory deviation trend, enabling the rapid identification of trajectory anomalies. Combining the force analysis of the regions where the trajectory deviation exceeds the limit can clarify the propagation path of the fatigue damage and form an intuitive description of the damage propagation process. The extraction and analysis of the visual images of the guide rail surfaces enable the characteristics of local cracks, wear, and deformation to be matched with the trajectory anomalies, further optimizing the visual segmentation boundaries to ensure the accurate extraction of the details of the structure changes. Based on the correlation between the trajectory deviation, the fatigue damage propagation path, and the structure change region, the regions with unstable operation states can be screened out earlier, providing early warning support for potential faults. Through this series of analysis means, the operation state of the guide rails, from trajectory deviation to force change, then to fatigue damage propagation, and finally to surface morphological changes, is refined and disassembled, making the identification of abnormal states more comprehensive and avoiding misjudgments that may be caused by single-parameter analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a schematic diagram of the working process of the present invention;

[0055] Figure 2 is a flowchart of step S1 of the present invention;

[0056] Figure 3 is a flowchart of step S2 of the present invention;

[0057] Figure 4 is a flowchart of step S3 of the present invention;

[0058] Figure 5 This is the flowchart of step S4 of the present invention;

[0059] Figure 6 This is the flowchart of step S5 of the present invention. Specific embodiments

[0060] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0061] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0062] Please refer to Figure 1 , the present invention provides a technical solution: an industrial equipment operation analysis method based on machine vision, including the following steps:

[0063] S1: Obtain the trajectory of the numerical control machine tool guide rail under the current processing task through machine vision, construct the three-dimensional coordinates and timestamps of each trajectory point, analyze the arrangement trend of the trajectory points in the time series, extract the key trajectory points during the operation of the guide rail by calculating the distribution density of the trajectory points, and obtain the guide rail trajectory distribution data;

[0064] S2: Divide the guide rail trajectory distribution data according to a fixed time window, calculate the morphological change amplitude of the trajectory points in adjacent time windows, analyze the spatial distribution pattern of the trajectory points, judge the offset trend of the trajectory points in space according to the local offset direction of the trajectory points, screen out the trajectory points with abnormal morphological change rates and calculate the global morphological deviation to obtain the trajectory deviation analysis result;

[0065] S3: Identify the trajectory regions in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtain the force condition of the guide rail in the region, calculate the stress distribution on the guide rail contact surface, obtain the region where the stress distribution is concentrated, and judge the propagation path of the guide rail fatigue damage to obtain the damage path analysis result;

[0066] S4: Based on the results of the damage path analysis, extract the corresponding visual images of the guide rail surface on the fatigue damage propagation path, calculate the structural integrity of the guide rail surface within the target area, analyze the characteristics of local cracks, wear, and deformation on the guide rail surface, and adjust the visual segmentation boundary to generate the analysis results of the guide rail structure changes;

[0067] S5: Analyze the correlation between the guide rail trajectory deviation, fatigue damage propagation path, and structure change area in the analysis results of the guide rail structure changes, screen the areas with unstable guide rail operation states, and mark them as warning areas to obtain the analysis results of the abnormal guide rail operation;

[0068] The guide rail trajectory distribution data includes the set of three-dimensional coordinates of trajectory points, the sequence of trajectory point timestamps, the density distribution information of trajectory points, and the set of key trajectory points. The trajectory deviation analysis results include time window division information, the amplitude of trajectory shape changes, the spatial offset trend of trajectory points, and the global trajectory shape deviation. The damage path analysis results include areas with excessive trajectory deviation, guide rail force distribution information, stress concentration areas, and fatigue damage propagation paths. The guide rail structure change analysis results include guide rail surface crack characteristics, guide rail surface wear areas, guide rail deformation characteristics, and visual segmentation boundary adjustment information. The abnormal guide rail operation analysis results include areas with abnormal trajectory deviation, fatigue damage propagation areas, structure change risk areas, and warning area marking information.

[0069] Please refer to Figure 2 , and the specific steps for obtaining the guide rail trajectory distribution data are as follows:

[0070] S111: Obtain the trajectory of the CNC machine tool guide rail under the current machining task, collect the spatial position information of the trajectory points, including three-dimensional coordinates and timestamps, analyze the collected sequence of trajectory points, and determine the arrangement trend of the trajectory points in the time series to obtain the sorting result of the trajectory point time series;

[0071] To obtain the trajectory of the CNC machine tool guide rail under the current machining task, it is necessary to use a high-speed camera to collect trajectory points. The camera should have a high frame rate (at least 1000 fps) to ensure accurate recording of trajectory data. At the same time, it is necessary to ensure that the optical resolution of the camera is within 0.01 mm to reduce sampling errors. After the camera collects the trajectory, the system needs to preprocess the trajectory points, including removing abnormal points, noise reduction, and interpolation and completion. Abnormal points can be screened by the rate of change of the distance between trajectory points. If the rate of change between trajectory points exceeds the set threshold of 10%, then this point is excluded. For example, the rate of change between the coordinates of trajectory point 1 (10.23, 5.12, 0.34) and trajectory point 2 (10.50, 5.30, 0.40) is:

[0072]

[0073] Among them, R 1,2Represents the rate of change of the distance between trajectory point 1 and trajectory point 2, expressed as a percentage, d 1,2 Represents the Euclidean distance between trajectory point 1 and trajectory point 2, in millimeters (mm), d 0,1 Represents the Euclidean distance between trajectory point 0 and trajectory point 1, in millimeters (mm), |d 1,2 -d 0,1 | Represents the change in Euclidean distance between two adjacent trajectory points, in millimeters (mm).

[0074] Assume the distance d between trajectory point 0 and trajectory point 1 0,1 = 0.20, and calculate the rate of change as follows:

[0075]

[0076] Since 40% is greater than the 10% threshold, trajectory point 2 is excluded. Median filtering is used for noise reduction, with a window size of 5 points as the benchmark. After denoising, missing trajectory points need to be interpolated and filled. Cubic spline interpolation can be used to fill the missing data between trajectory points. The trajectory points are sorted by timestamp to form a time series data table. For example, the timestamp of trajectory point 1 is 0 ms, the timestamp of trajectory point 2 is 5 ms, and the timestamp of trajectory point 3 is 10 ms. Ensure that the trajectory points are arranged in chronological order to obtain: the sorted result of the trajectory point time series.

[0077] S112: Based on the sorted result of the trajectory point time series, use the formula:

[0078]

[0079] Calculate the distribution density D of trajectory point i i , to obtain the trajectory point density information;

[0080] Among them, P i Represents the spatial position of trajectory point i, equivalent to the coordinate value of the trajectory point (x i , y i , z i ), in millimeters (mm), P j Represents the spatial position of trajectory point j, equivalent to the coordinate value of the trajectory point (x j , y j , z j ), in millimeters (mm), Represents the spatial distribution influence factor between trajectory point i and trajectory point j, Represents the time influence factor between trajectory point i and trajectory point j, σ p Represents the spatial coordinate normalization factor, usually set according to the spatial change range of the trajectory points, in millimeters (mm), σ tRepresents the time normalization factor, usually set according to the distribution range of the time interval data, with the unit of millisecond (ms), and n represents the total number of trajectory points within the selected window.

[0081] First, calculate the distribution density of each trajectory point. Select the calculation window size n = 10 as the statistical range of the number of trajectory points, and calculate the Euclidean distance between the trajectory points. The formula is:

[0082]

[0083] Among them, d i,j Represents the Euclidean distance between trajectory point i and trajectory point j, with the unit of millimeter (mm), x i , y i , z i Represents the spatial coordinate value of trajectory point i, with the unit of millimeter (mm), x j , y j , z j Represents the spatial coordinate value of trajectory point j, with the unit of millimeter (mm).

[0084] If the coordinates of trajectory point 1 are (10.23, 5.12, 0.34), the coordinates of trajectory point 2 are (10.25, 5.15, 0.35), and the coordinates of trajectory point 3 are (10.28, 5.18, 0.36), calculate the Euclidean distance between trajectory point 1 and trajectory point 2:

[0085]

[0086] Calculate the Euclidean distance between trajectory point 1 and trajectory point 3:

[0087]

[0088] The time interval normalization processing method is:

[0089] Among them, t i,j Represents the time normalization value between trajectory point i and trajectory point j, T i Represents the timestamp of trajectory point i, with the unit of millisecond (ms), T j Represents the timestamp of trajectory point j, with the unit of millisecond (ms), σ t Represents the time normalization factor, usually taking the standard deviation of the average time interval of the trajectory points, with the unit of millisecond (ms).

[0090] Assume σ t = 5ms, calculate the time normalization value between trajectory point 1 and trajectory point 2:

[0091] Calculate the time normalization value between trajectory point 1 and trajectory point 3:

[0092] Substitute into the distribution density calculation formula:

[0093] Assume σ p = 0.05, and calculate the distribution density of trajectory point 1:

[0094]

[0095] Calculate the distribution density of trajectory point 2:

[0096]

[0097] The results show that the distribution density of trajectory point 1 is 0.025, and the distribution density of trajectory point 2 is 0.035.

[0098] S113: Based on the trajectory point density information, screen the trajectory points with distribution density values higher than the set density threshold, extract the key trajectory points during the operation of the guide rail, and count the spatial distribution range of the key trajectory points to generate the guide rail trajectory distribution data;

[0099] According to the trajectory point distribution density matrix, it is necessary to screen the key trajectory points and set the threshold δ of the distribution density D = 0.03. If the density value D of the trajectory point i is higher than δ D , then this trajectory point is regarded as a key trajectory point. For example:

[0100] The density of trajectory point 1 is D1 = 0.025, which is less than δ D = 0.03 and is not used as a key trajectory point;

[0101] The density of trajectory point 2 is D2 = 0.035, which is greater than δ D = 0.03 and is used as a key trajectory point;

[0102] The density of trajectory point 3 is D3 = 0.030, which is equal to δ D = 0.03 and is used as a key trajectory point.

[0103] Finally, a set of key trajectory points is obtained, and trajectory point 2 and trajectory point 3 are selected. Next, calculate the spatial bounding box of the key trajectory points to obtain the motion boundary of the trajectory points. The calculation method is as follows:

[0104] X min = min(x k ), X max = max(x k );

[0105] Y min = min(y k ), Y max= max(y k );

[0106] Z min = min(z k ), Z max = max(z k );

[0107] Wherein, X min represents the minimum coordinate value of the key trajectory point in the x direction, in millimeters (mm), X max represents the maximum coordinate value of the key trajectory point in the x direction, in millimeters (mm), Y min represents the minimum coordinate value of the key trajectory point in the y direction, in millimeters (mm), Y max represents the maximum coordinate value of the key trajectory point in the y direction, in millimeters (mm), Z min represents the minimum coordinate value of the key trajectory point in the z direction, in millimeters (mm), Z max represents the maximum coordinate value of the key trajectory point in the z direction, in millimeters (mm), x k , y k , z k represent the coordinate values of all key trajectory points, in millimeters (mm).

[0108] Assume the coordinates of key trajectory point 2 and trajectory point 3 are as follows: Trajectory point 2: (10.25, 5.15, 0.35) Trajectory point 3: (10.28, 5.18, 0.36).

[0109] It is calculated that:

[0110] X min = min(10.25, 10.28) = 10.25, X max = max(10.25, 10.28) = 10.28;

[0111] Y min = min(5.15, 5.18) = 5.15, Y max = max(5.15, 5.18) = 5.18;

[0112] Z min = min(0.35, 0.36) = 0.35, Z max = max(0.35, 0.36) = 0.36;

[0113] The result shows that X min = 10.25 mm represents the leftmost boundary of the key trajectory point in the x direction, X max = 10.28 mm represents the rightmost boundary of the key trajectory point in the x direction, Ymin = 5.15 mm represents the lowest boundary of the key trajectory point in the y - direction, Y max = 5.18 mm represents the highest boundary of the key trajectory point in the y - direction, Z min = 0.35 mm represents the position closest to zero of the key trajectory point in the z - direction, Z max = 0.36 mm represents the farthest position of the key trajectory point in the z - direction. Finally, establish the spatial envelope range of the key trajectory point and generate the guide - rail trajectory distribution data.

[0114] Please refer to Figure 3 , and the steps for obtaining the trajectory deviation analysis result are specifically as follows:

[0115] S211: Divide the guide - rail trajectory distribution data according to a fixed time window, and use the formula:

[0116]

[0117] Calculate the average Euclidean distance change amount Δd of the trajectory points within the time window h h , and the average direction - angle change amount Δθ of the trajectory points within the time window h h , to obtain the morphological change distribution data;

[0118] Among them, represents the Euclidean distance of the j1 - th trajectory point relative to the reference point within the time window h, with the unit of millimeter (mm), represents the Euclidean distance of the j1 - th trajectory point relative to the reference point within the time window h + 1, with the unit of millimeter (mm), represents the direction angle of the j1 - th trajectory point within the time window h, with the unit of degree (°), represents the direction angle of the j1 - th trajectory point within the time window h + 1, with the unit of degree (°), and n1 represents the total number of trajectory points within the time window.

[0119] First, based on the trajectory - point data divided by the time window, extract the three - dimensional coordinates of all trajectory points within the time windows h and h + 1, and calculate the relative motion parameters between the trajectory points. Set the number of trajectory points n1 to 10 to ensure that the trajectory points are evenly distributed in each time window. When calculating the morphological change amplitude of the trajectory points, it is necessary to evaluate the Euclidean distance change amount and the direction - angle change amount respectively. The Euclidean distance change amount represents the spatial - position change of the trajectory points within the time windows h and h + 1, and use:

[0120] Among them, Represents the Euclidean distance of trajectory point j1 within time window h. Assume that the Euclidean distance of a certain trajectory point within time window 1 is 10.25 mm, and the Euclidean distance of the same trajectory point within time window 2 is 10.60 mm. Then the change in the Euclidean distance of this point is calculated as follows: Δd1 = |10.60 - 10.25| = 0.35 mm.

[0121] This calculation result indicates that the spatial displacement amplitude of this trajectory point between adjacent time windows is 0.35 mm. If this value is large, it means that there is a significant displacement of the trajectory point between time windows. Conversely, it means that the spatial stability of the trajectory point is strong.

[0122] When calculating the change in the direction angle, it is necessary to analyze the angle change of the trajectory point within consecutive time windows, using:

[0123] where Represents the direction angle of trajectory point j1 within time window h. Assume that the direction angle of a certain trajectory point within time window 1 is 32.5°, and the direction angle within time window 2 is 30.0°. Then the change in the direction angle of this point is calculated as follows: Δθ1 = |30.0 - 32.5| = 2.5°

[0124] This calculation result indicates that the change in the direction angle of this trajectory point within adjacent time windows is 2.5°. If this value is large, it means that the direction of the trajectory point changes greatly in a short time, indicating that the motion state of the trajectory point is unstable. If this value is small, it means that the direction change of the trajectory point is relatively stable.

[0125] S212: Based on the morphological change distribution data, analyze the spatial distribution pattern of the trajectory points, judge the local offset direction of the trajectory points in adjacent time windows, and judge the overall offset trend of the trajectory points according to the spatial distribution density of the trajectory points and the continuous change of the local offset direction. Screen the trajectory points with abnormal morphological change rates to obtain abnormal trajectory point information;

[0126] Analyze the spatial distribution pattern of the trajectory points, calculate the spatial density of the trajectory points within each time window h, calculate the projection distribution of the trajectory points on the x - y plane and the x - z plane, obtain the local offset direction of the trajectory points, and evaluate the offset trend of the trajectory points within consecutive time windows. If the offset direction of the trajectory point remains consistent within multiple time windows, that is, the movement amplitude of the trajectory point in the x - direction is always greater than 0.2 mm, while the movement amplitudes in the y - direction and z - direction are kept within 0.1 mm, then it is determined that the trajectory point has a stable offset trend. If the direction - angle change rate of the trajectory point in multiple consecutive time windows is greater than 5° / s, it is considered that the trajectory point has an abnormal offset. Screen out the points with abnormal morphological change rates of the trajectory points to obtain: trajectory points with abnormal morphological change rates.

[0127] S213: Based on the abnormal trajectory point information, use the formula:

[0128]

[0129] Calculate the global shape deviation D of all abnormal trajectory points global , and obtain the trajectory deviation analysis result;

[0130] where m1 represents the total number of all abnormal trajectory points, represents the spatial coordinate of the k1-th abnormal trajectory point, with the unit of millimeter (mm), X mean 、Y mean 、Z mean represent the spatial mean coordinates of all trajectory points, with the unit of millimeter (mm).

[0131] If the calculated trajectory mean coordinate is (10.5, 5.2, 0.6) mm and the coordinate of a certain abnormal trajectory point is (11.0, 5.4, 0.9) mm, then the global shape deviation of this point is calculated as follows:

[0132]

[0133] The calculation result shows that the offset of this abnormal trajectory point relative to the mean position of all trajectory points is 0.583 mm. If this value is large, it indicates that the movement trajectory of this abnormal trajectory point has a significant deviation from the overall trajectory, which may indicate abnormal movement during the processing. If this value is small, it indicates that the deviation degree of this trajectory point is low and is relatively consistent with the overall trajectory distribution.

[0134] Please refer to Figure 4 , and the specific steps for obtaining the damage path analysis result are as follows:

[0135] S311: Identify the trajectory area in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtain the force condition of the guide rail within the area, and use the formula;

[0136]

[0137] Calculate the stress of the k2-th contact point within the trajectory area to obtain the guide rail stress distribution data;

[0138] where, represents the force on the k2-th contact point within the trajectory area, with the unit of Newton (N), represents the contact area of the k2-th contact point within the trajectory area, with the unit of square millimeter (mm2).

[0139] Set the deviation threshold T d , traverse the trajectory data, and filter out the trajectory point deviation value exceeding the threshold Td area. Assume that the guide rail trajectory data contains 1000 trajectory points, and each trajectory point records the three-dimensional coordinate position and deviation value under different time windows h. First, count the deviation range of all trajectory points and set a deviation threshold, for example, T d = 0.5 mm, that is, all trajectory points with a deviation greater than 0.5 mm are regarded as abnormal trajectory points. For each trajectory point k2, if its deviation satisfies then this point is classified into the area with excessive deviation. For example, if the deviation of a certain trajectory point meets the condition, it is included in the deviation area.

[0140] After determining the trajectory deviation area, further obtain the force condition of the guide rail in the area. First, calculate the contact pressure corresponding to each trajectory point. Assume that there are 50 deviation points in the trajectory area, and the force of each point is determined by external load and internal stress. Extract the external force at the position of the trajectory point and calculate its corresponding contact area Subsequently, calculate the stress distribution on the contact surface of the guide rail in the trajectory area, using:

[0141]

[0142] Set Then calculate the stress of the first contact point:

[0143]

[0144] Repeat the calculation for all deviation trajectory points to obtain the stress values of all contact points in the trajectory area, and obtain the guide rail stress distribution data.

[0145] S312: Based on the guide rail stress distribution data, obtain the area where the stress distribution is concentrated, as well as the average stress value in the area of the contact points with stress concentration, analyze the spatial distribution of the stress concentration area, judge the propagation path of the guide rail fatigue damage and the main direction and range of damage expansion, and obtain the damage path analysis result;

[0146] Set the stress concentration determination threshold T σ , and screen out the area where the stress is greater than T σ . For example, set T σ = 4 MPa, then all areas where the contact point stress are regarded as stress concentration characteristic areas. Assume that 5 out of 50 contact points meet this condition, then the stress concentration area is composed of these contact points.

[0147] Subsequently, calculate the average stress value in the area of the contact points with stress concentration, calculate the mean value of all stress concentration points, using the formula:

[0148] If the stresses of the 5 contact points are respectively Then calculate their mean value:

[0149]

[0150] Finally, analyze the spatial distribution of the stress concentration area, judge the propagation path of the guide rail fatigue damage, calculate the main direction and range of damage propagation, and obtain: the damage path analysis result.

[0151] Among them, σ avg represents the average stress of the contact points within the stress concentration area, with the unit of megapascal (MPa), m1 represents the number of contact points within the stress concentration area, represents the stress of the k2-th contact point within the trajectory area, with the unit of megapascal (MPa), T σ represents the stress concentration determination threshold, with the unit of megapascal (MPa).

[0152] Please refer to Figure 5 , and the specific steps for obtaining the analysis result of the guide rail structure change are as follows:

[0153] S411: Based on the damage propagation data in the damage path analysis result, extract the trajectory coordinates of the damage area and set the visual image acquisition range. According to the acquisition range, perform image acquisition on the guide rail surface, extract the continuous surface visual images of the damage area, and obtain the guide rail surface visual image data;

[0154] First, call the damage propagation path data obtained from the damage path analysis. This data includes the spatial coordinate information of the damage area, that is, the trajectory coordinates (x p , y p , z p ), and the propagation direction of the damage path. Determine the visual image acquisition range according to the trajectory coordinates. When setting the range, consider the trajectory point spacing d t and the damage area coverage radius R v . Set d t = 2mm, R v = 10mm to ensure that the acquisition area completely covers the damage area. Subsequently, use a high-precision optical camera device to perform image acquisition on the guide rail surface, and intercept images at fixed intervals d v along the damage path. Set d v = 5mm to make the adjacent images have partial overlap and enhance the regional integrity. After the image data is stored, perform preliminary screening to remove the images with abnormal lighting conditions, and obtain a series of continuous surface visual image data. Finally, obtain: the guide rail surface visual image data.

[0155] S412: Based on the guide rail surface visual image data, use the formula:

[0156]

[0157] Calculate the overall structural integrity index S of the target area c ;

[0158] Among them, represents the gray value of the i1-th pixel point, with the unit of 0 - 255, represents the edge gradient value of the i1-th pixel point, with the unit of dimensionless, and N represents the total number of pixel points in the target area.

[0159] First, extract the gray feature and edge gradient information of the surface visual image. The gray feature is represented by the gray value of the pixel point, and the gradient information is calculated using the Sobel operator. The gradient value of each pixel point is obtained through

[0160]

[0161] Among them, represents the gray change rate of the i1-th pixel point in the horizontal direction, with the unit of dimensionless, represents the gray change rate of the i1-th pixel point in the vertical direction, with the unit of dimensionless,

[0162] is calculated, where and respectively represent the gray change rates in the horizontal and vertical directions. Set the gray threshold G th = 50 to eliminate the low-contrast area. Subsequently, calculate the surface integrity index S c , using the formula:

[0163]

[0164] Set N = 10000 pixel points for calculation. Assume that part of the data of a damaged area is:

[0165] Substitute into the formula for calculation:

[0166]

[0167] Integrity index S c After calculation, compare it with the integrity threshold T s = 10. If S c < T s , then it is determined as a damaged area. In this example, S c = 15.17 is greater than the threshold, indicating that the structure of this area is complete. Finally, obtain: the surface structure integrity data of the guide rail.

[0168] S413: Analyze the local crack, wear and deformation characteristics of the guide rail surface based on the data of the guide rail surface structural integrity. According to the morphological damage amount of the deformation characteristics, adjust the visual segmentation boundary, optimize the boundary position, and obtain the analysis result of the guide rail structure change;

[0169] Analyze the local crack, wear and deformation characteristics of the guide rail surface. Calculate the morphological damage amount according to the deformation characteristics. Call the damage area information in the structural integrity data, select the judgment area, and set the crack width threshold W cr = 0.3 mm and the wear depth threshold D w = 0.5 mm to screen out the areas with severe crack propagation or wear. In the screened areas, calculate the boundary change of the damage area, extract the damage morphological characteristics, and judge whether the damage morphological characteristics exceed the set standard. If the morphological damage amount exceeds the threshold, adjust the visual segmentation boundary to ensure the recognition accuracy of the damage area. After extracting the damage morphological characteristics, optimize the boundary position to make the segmentation boundary in the visual image match the actual damage morphology. Finally, obtain: the analysis result of the guide rail structure change.

[0170] Please refer to Figure 6 , and the steps for obtaining the analysis result of the abnormal operation of the guide rail are specifically as follows:

[0171] S511: Analyze the correlation between the guide rail trajectory deviation, fatigue damage propagation path and structural change area in the analysis result of the guide rail structure change to obtain the trajectory damage correlation data;

[0172] First, determine the displacement change amount between trajectory points, calculate the Euclidean distance of each adjacent trajectory point, and accumulate and sum the trajectory deviations within all time steps to calculate the overall trajectory deviation rate. If the trajectory deviation value exceeds the set reference value, determine that the trajectory point is an abnormal point. The reference value is set according to the historical data of the normal trajectory. For example, if the average deviation of the normal trajectory data is 2.5 mm, the reference value can be set to 2 times the standard deviation, that is, 5 mm. Screen out the trajectory points that exceed this threshold, and record their time and position. Subsequently, extract the fatigue damage propagation path data, calculate the stress change rate of the damage area, and select the stress values at multiple time steps for calculation. For example, if the damage propagation rate within 10 seconds is 15%, use this data to evaluate the damage propagation trend in the future time period, and set the damage propagation judgment threshold. For example, if the average value of the normal damage propagation rate is 10%, the threshold can be set to 12%. Screen out the damage propagation areas that exceed the threshold and record the relevant data. Finally, match the trajectory deviation area with the fatigue damage propagation path, analyze the spatial distribution of the two. If the overlap degree between the trajectory deviation area and the fatigue damage propagation path is more than 80%, it is determined that the two are strongly correlated.

[0173] S512: Based on the trajectory damage correlation data, screen the areas where the guide rail operation state is unstable, mark them as warning areas, and obtain the analysis results of abnormal guide rail operation;

[0174] Screen the areas where the guide rail operation state is unstable, mark them as warning areas, extract the areas with large trajectory deviations, calculate the cumulative change amount of the trajectory deviation, perform differential calculation on the trajectory points at different times, and sum them cumulatively. For example, if the cumulative change amount of the trajectory deviation within 30 seconds is 20 mm, which exceeds the set threshold of 15 mm, it is determined that this area is an unstable operation area. Subsequently, calculate the damage increase amplitude of the damage propagation path within the unstable area, select the stress data of this area, and calculate its time change rate. For example, if the damage increase amplitude of this area within 20 seconds is 18%, exceeding the judgment threshold of 15%, this area needs further attention. Finally, screen the areas that meet the warning conditions, mark them as warning areas, and record the trajectory deviation increase amplitude and damage propagation trend data, which are the analysis results of abnormal guide rail operation.

[0175] An industrial equipment operation analysis system based on machine vision, which is used to execute the above-mentioned industrial equipment operation analysis method based on machine vision. The system includes:

[0176] The trajectory data construction module obtains the trajectory of the numerical control machine tool guide rail under the current machining task, constructs the three-dimensional coordinates and timestamps of each trajectory point, and extracts the key trajectory points during the operation of the guide rail to obtain the guide rail trajectory distribution data;

[0177] The trajectory deviation calculation module divides the guide rail trajectory distribution data according to a fixed time window, calculates the morphological change amplitude of the trajectory points within adjacent time windows, and judges the offset trend of the trajectory points in space to obtain the trajectory deviation analysis result;

[0178] The fatigue damage identification module identifies the trajectory areas in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtains the force condition of the guide rail within the area, and judges the propagation path of the guide rail fatigue damage to obtain the damage path analysis result;

[0179] The guide rail structure change detection module, based on the damage path analysis result, extracts the corresponding visual images of the guide rail surface on the fatigue damage propagation path, analyzes the local crack, wear and deformation characteristics of the guide rail surface and adjusts the visual segmentation boundary to generate the guide rail structure change analysis result;

[0180] The abnormal operation area identification module analyzes the correlation between the guide rail trajectory deviation, fatigue damage propagation path and structure change area in the guide rail structure change analysis result, screens the areas where the guide rail operation state is unstable, and obtains the analysis results of abnormal guide rail operation.

[0181] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. An industrial equipment operation analysis method based on machine vision, characterized in that, Including the following steps: S1: Obtain the trajectory of the CNC machine tool guide rail under the current machining task, construct the three-dimensional coordinates and timestamps of each trajectory point, extract the key trajectory points during the operation of the guide rail, and obtain the guide rail trajectory distribution data; S2: Divide the guide rail trajectory distribution data according to a fixed time window, calculate the morphological change amplitude of the trajectory points within adjacent time windows, judge the offset trend of the trajectory points in space, and obtain the trajectory deviation analysis result; S3: Identify the trajectory regions in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtain the force condition of the guide rail within the region, judge the propagation path of the fatigue damage of the guide rail, and obtain the damage path analysis result; S4: Based on the damage path analysis result, extract the visual images of the corresponding guide rail surface on the fatigue damage propagation path, analyze the local crack, wear and deformation characteristics of the guide rail surface and adjust the visual segmentation boundary, and generate the guide rail structure change analysis result; S5: Analyze the correlation between the guide rail trajectory deviation, fatigue damage propagation path and structure change region in the guide rail structure change analysis result, screen the regions with unstable guide rail operation state, and obtain the guide rail abnormal operation analysis result.

2. The industrial equipment operation analysis method based on machine vision according to claim 1, wherein The guide rail trajectory distribution data includes the three-dimensional coordinate set of trajectory points, the trajectory point timestamp sequence, the trajectory point density distribution information, and the key trajectory point set. The trajectory deviation analysis result includes the time window division information, the trajectory morphological change amplitude, the trajectory point spatial offset trend, and the global trajectory morphological deviation. The damage path analysis result includes the trajectory deviation overrun region, the guide rail force distribution information, the stress concentration region, and the fatigue damage propagation path. The guide rail structure change analysis result includes the guide rail surface crack characteristics, the guide rail surface wear region, the guide rail deformation characteristics, and the visual segmentation boundary adjustment information. The guide rail abnormal operation analysis result includes the trajectory deviation abnormal region, the fatigue damage propagation region, the structure change risk region, and the warning region marking information.

3. The method for analyzing the operation of industrial equipment based on machine vision according to claim 2, wherein The specific steps for obtaining the guide rail trajectory distribution data are as follows: S111: Obtain the trajectory of the CNC machine tool guide rail under the current machining task, collect the spatial position information of the trajectory points, including three-dimensional coordinates and timestamps, analyze the collected trajectory point sequence, and determine the arrangement trend of the trajectory points in the time series to obtain the trajectory point time series sorting result; S112: Based on the trajectory point time series sorting result, use the formula: Calculate the distribution density D of the trajectory point i i , and obtain the trajectory point density information; Among them, P i represents the spatial position of trajectory point i, P j represents the spatial position of trajectory point j, represents the spatial distribution influence factor between trajectory point i and trajectory point j, represents the time influence factor between trajectory point i and trajectory point j, σ p represents the spatial coordinate normalization factor, σ t represents the time normalization factor, and n represents the total number of trajectory points within the selected window; S113: Based on the trajectory point density information, screen the trajectory points with distribution density values higher than the set density threshold, extract the key trajectory points during the operation of the guide rail, and statistically analyze the spatial distribution range of the key trajectory points to generate the guide rail trajectory distribution data.

4. The industrial equipment operation analysis method based on machine vision according to claim 3, characterized in that, The specific steps for obtaining the trajectory deviation analysis result are as follows: S211: Divide the guide rail trajectory distribution data according to a fixed time window, use the formula: Calculate the average change in Euclidean distance Δd of the trajectory points within the time window h respectively h , and the average change in direction angle Δθ of the trajectory points within the time window h h , and obtain the morphological change distribution data; Among them, represents the Euclidean distance of the j1-th trajectory point relative to the reference point within the time window h, represents the Euclidean distance of the j1-th trajectory point relative to the reference point within the time window h + 1, represents the direction angle of the j1-th trajectory point within the time window h, represents the direction angle of the j1-th trajectory point within the time window h + 1, and n1 represents the total number of trajectory points within the time window; S212: Based on the morphological change distribution data, analyze the spatial distribution pattern of the trajectory points, judge the local offset direction of the trajectory points in adjacent time windows, and judge the overall offset trend of the trajectory points according to the continuity change of the spatial distribution density and local offset direction of the trajectory points, and screen the trajectory points with abnormal morphological change rates to obtain the abnormal trajectory point information; S213: Based on the abnormal trajectory point information, use the formula: Calculate the global shape deviation D of all abnormal trajectory points global , and obtain the trajectory deviation analysis result; Among them, m1 represents the total number of all abnormal trajectory points, represents the spatial coordinate of the k1-th abnormal trajectory point, X mean , Y mean , Z mean represents the spatial mean coordinate of all trajectory points.

5. The industrial equipment operation analysis method based on machine vision according to claim 4, characterized in that, The specific steps for obtaining the damage path analysis result are as follows: S311: Identify the trajectory area in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtain the force condition of the guide rail within the area, and use the formula; Calculate the stress of the k2-th contact point within the calculated trajectory area Obtain the stress distribution data of the guide rail Among them, represents the force on the k2-th contact point within the trajectory region, represents the contact area of the k2-th contact point within the trajectory region; S312: Based on the guide rail stress distribution data, obtain the area with concentrated stress distribution and the average stress value within the contact point area with stress concentration, analyze the spatial distribution of the stress concentration area, judge the propagation path of the guide rail fatigue damage and the direction and range of damage expansion, and obtain the damage path analysis result.

6. The method for analyzing the operation of industrial equipment based on machine vision according to claim 5, wherein, The specific steps for obtaining the guide rail structure change analysis result are as follows: S411: Based on the damage expansion data in the damage path analysis result, extract the trajectory coordinates of the damaged area and set the visual image acquisition range. According to the acquisition range, collect images of the guide rail surface, and extract the continuous surface visual images of the damaged area to obtain the guide rail surface visual image data; S412: Based on the guide rail surface visual image data, use the formula: Calculate the overall structural integrity index S of the target area c ; Among them, represents the gray value of the i1-th pixel point, represents the edge gradient value of the i1-th pixel point, with the unit being dimensionless, and N represents the total number of pixel points in the target area; S413: Based on the guide rail surface structure integrity data, analyze the local crack, wear and deformation characteristics of the guide rail surface, and adjust the visual segmentation boundary according to the morphological damage amount of the deformation characteristics to optimize the boundary position, and obtain the guide rail structure change analysis result.

7. The industrial equipment operation analysis method based on machine vision according to claim 6, characterized in that The specific steps for obtaining the guide rail abnormal operation analysis result are as follows: S511: Analyze the correlation between the guide rail trajectory deviation, fatigue damage propagation path and structure change area in the guide rail structure change analysis result to obtain the trajectory damage correlation data; S512: Based on the trajectory damage correlation data, screen the areas where the guide rail operation state is unstable and mark them as warning areas to obtain the guide rail abnormal operation analysis result.

8. An industrial equipment operation analysis system based on machine vision, characterized in that, According to the machine vision-based industrial equipment operation analysis method according to any one of claims 1-7, the system includes: The trajectory data construction module obtains the trajectory of the numerical control machine tool guide rail under the current processing task, constructs the three-dimensional coordinates and timestamps of each trajectory point, and extracts the key trajectory points during the operation of the guide rail to obtain the guide rail trajectory distribution data; The trajectory deviation calculation module divides the guide rail trajectory distribution data according to a fixed time window, calculates the morphological change amplitude of the trajectory points within adjacent time windows, and judges the offset trend of the trajectory points in space to obtain the trajectory deviation analysis result; The fatigue damage identification module identifies the trajectory area in the trajectory deviation analysis result where the deviation exceeds the deviation threshold, obtains the force condition of the guide rail within the area, and judges the propagation path of the guide rail fatigue damage to obtain the damage path analysis result; The guide rail structure change detection module extracts the corresponding guide rail surface visual images on the fatigue damage propagation path based on the damage path analysis result, analyzes the local crack, wear and deformation characteristics of the guide rail surface and adjusts the visual segmentation boundary to generate the guide rail structure change analysis result; The abnormal operation area identification module analyzes the correlation between the guide rail trajectory deviation, fatigue damage propagation path and structure change area in the guide rail structure change analysis result, and screens the areas where the guide rail operation state is unstable to obtain the guide rail abnormal operation analysis result.

Images