CN117216728B - Excavator movable arm stability detection method - Google Patents
Excavator movable arm stability detection method Download PDFInfo
- Publication number
- CN117216728B CN117216728B CN202311481983.0A CN202311481983A CN117216728B CN 117216728 B CN117216728 B CN 117216728B CN 202311481983 A CN202311481983 A CN 202311481983A CN 117216728 B CN117216728 B CN 117216728B
- Authority
- CN
- China
- Prior art keywords
- sampling point
- displacement
- time image
- included angle
- sampling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims description 11
- 238000005070 sampling Methods 0.000 claims abstract description 364
- 238000006073 displacement reaction Methods 0.000 claims abstract description 136
- 238000000034 method Methods 0.000 claims abstract description 59
- 230000002159 abnormal effect Effects 0.000 claims abstract description 38
- 230000002776 aggregation Effects 0.000 claims abstract description 27
- 238000004220 aggregation Methods 0.000 claims abstract description 27
- 230000005856 abnormality Effects 0.000 claims description 49
- 238000012417 linear regression Methods 0.000 claims description 33
- 238000012886 linear function Methods 0.000 claims description 24
- 238000006116 polymerization reaction Methods 0.000 claims description 17
- 238000005314 correlation function Methods 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 description 17
- 230000008859 change Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 3
- 230000008602 contraction Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Abstract
The invention relates to the technical field of data processing, and discloses a method for detecting stability of an excavator movable arm, which comprises the steps of acquiring a plurality of displacement data of the excavator movable arm, which is close to the end part of a bucket, a plurality of included angle data of the movable arm and an extending arm and a plurality of stress data of the two ends of the movable arm at each time node in the same period; respectively constructing a displacement-time image, an included angle-time image and a stress-time image, weighting an LOF algorithm by using a weighting factor pair obtained by the displacement linear aggregation degree, the included angle linear aggregation degree and the fluctuation anomaly degree of each sampling point in the displacement-time image, the included angle-time image and the stress-time image, and detecting each sampling point by using the weighted LOF algorithm; the stability is judged by adding the multidimensional data of the excavator to weight whether the data of the excavator is abnormal or not, and the accuracy of stability judgment is effectively improved.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to a method for detecting stability of a movable arm of an excavator.
Background
The excavator is a large mechanical device which is frequently used, particularly in the construction engineering industry, such as bridges, highways, underground engineering and the like, and has many applications in living scenes, such as repairing highways, barren and the like. With the rapid development of engineering large-scale machines, the excavator plays an important role in more and more engineering. In the running process of the excavator, the higher the stability is, the higher the safety coefficient is, and the operability is better. In the working process of the traditional excavator, as the movable arm is a main part of the excavator, the stability of the movable arm plays a vital role in the construction of the whole excavator, at present, stability indexes of the movable arm of the excavator under different working conditions are analyzed by an analysis method based on a mechanical principle to conduct guidance engineering practice, but when the excavator works, the stability of the movable arm is affected by various factors, the stability of the excavator is analyzed by the force on the movable arm, the stability of the excavator cannot be accurately fed back, and the quality inspection and performance judgment of the excavator are affected.
Disclosure of Invention
The invention is used for solving the problem of inaccurate analysis results caused by analyzing the stability of an excavator through a single mechanical principle at present, and provides a method for detecting the stability of a movable arm of the excavator, which comprises the following steps of:
acquiring a plurality of displacement data of the movable arm of the excavator, which is close to the end part of the bucket, a plurality of included angle data of the movable arm and the extending arm and a plurality of stress data of the two ends of the movable arm at each time node in a plurality of periods; respectively constructing a displacement-time image, an included angle-time image and a stress-time image by using the displacement data, the included angle data and the stress data;
obtaining displacement linear polymerization degree by utilizing fitting data of the displacement-time image and the number of displacement data contained in a sampling area of each sampling point;
obtaining the linear aggregation degree of the included angles by using the fitting data of the included angle-time images and the number of included angle data contained in the sampling area of each sampling point;
performing cross-correlation function fitting according to stress data of each sampling point in the stress-time image to obtain a cross-correlation value; the fluctuation anomaly degree of each sampling point is obtained by utilizing the distance from the sampling point to the origin of coordinates in the stress-time image;
respectively utilizing the displacement linear aggregation degree, the included angle linear aggregation degree and the fluctuation anomaly degree of each sampling point to obtain the weighting factors of the sampling points;
and weighting the LOF algorithm by using a weighting factor of the sampling point, detecting the sampling point by using the weighted LOF algorithm, and judging the stability of the movable arm of the excavator in a set period according to a detection result.
Preferably, the method for obtaining the displacement linear polymerization degree comprises the following steps:
performing linear function fitting by using the displacement-time image to obtain displacement linear offset;
obtaining the displacement abnormality degree of each sampling point in the displacement-time image by using the number of the displacement data contained in the sampling area of each sampling point in the displacement-time image;
and obtaining the displacement linear polymerization degree of each sampling point in the displacement-time image by using the displacement abnormality degree and the displacement linear offset degree of each sampling point in the displacement-time image.
Preferably, the method for obtaining the displacement linear offset comprises the following steps:
fitting by using displacement data contained in the displacement-time image to obtain a displacement linear regression function;
and obtaining the displacement linear offset by utilizing the slope and intercept of the displacement linear regression function obtained by fitting and the theoretical displacement linear regression function.
Preferably, the method for determining the sampling area of each sampling point in the displacement-time image comprises the following steps:
acquiring Euclidean distance between each sampling point and the nearest sampling point in the displacement-time image;
obtaining the sampling radius of each sampling point according to the Euclidean distance between each sampling point and the nearest sampling point in the displacement-time image on each time node;
and acquiring a sampling area of each sampling point in the displacement-time image by using the sampling radius of each sampling point in the displacement-time image.
Preferably, the method for determining the sampling area of each sampling point in the included angle-time image comprises the following steps:
acquiring Euclidean distance between each sampling point and the nearest sampling point in the included angle-time image;
obtaining the sampling radius of each sampling point according to the included angle on each time node, namely the Euclidean distance between each sampling point and the nearest sampling point in the time image;
and acquiring a sampling area of each sampling point in the included angle-time image by using the sampling radius of each sampling point in the included angle-time image.
Preferably, the method for performing cross-correlation function fitting to obtain a cross-correlation value according to stress data of each sampling point in the stress-time image comprises the following steps:
fitting by using stress data at two ends of a movable arm on each sampling point to obtain a stress linear function;
carrying out resultant force analysis on the stress data at the two ends of the movable arm on each sampling point, and taking the stress data after the resultant force analysis on each sampling point as ideal stress data of the movable arm on the sampling point;
fitting the obtained ideal stress data of the movable arm on each sampling point to obtain an ideal stress linear function;
constructing a stress cross-correlation function by using the obtained stress linear function and the ideal linear function;
and obtaining a cross-correlation value by using the stress cross-correlation function and the ideal cross-correlation function.
Preferably, the method for obtaining the fluctuation anomaly degree of each sampling point by using the distance from the sampling point to the origin of coordinates in the stress-time image comprises the following steps:
obtaining the stress abnormality degree of each sampling point in the stress-time image by utilizing the distance from each sampling point in the stress-time image to the origin of coordinates; and obtaining the fluctuation anomaly degree of each sampling point in the stress-time image by using the stress anomaly degree of each sampling point in the stress-time image.
Preferably, the method for obtaining the fluctuation anomaly degree of each sampling point in the stress-time image by using the stress anomaly degree of each sampling point in the stress-time image comprises the following steps:
and obtaining the fluctuation anomaly degree of each sampling point in each stress-time image by using the mean value of the Euclidean distance between each sampling point and other sampling points on each time node and the anomaly degree of the sampling point.
Preferably, the method for obtaining the weighting factor of the sampling point comprises the following steps:
obtaining displacement dimension weight, included angle dimension weight and stress dimension weight of each sampling point by utilizing the displacement linear aggregation degree, included angle linear aggregation degree and fluctuation anomaly degree of each sampling point respectively;
obtaining a weighting factor of each sampling point by using the displacement dimension weight, the included angle dimension weight and the stress dimension weight of each sampling point;
the expression of the weighting factor is as follows:
wherein P is the weighting factor of each sampling point;a displacement dimension weight for each sampling point; />The included angle dimension weight of each sampling point; />Stress dimension weight for each sampling point, +.>Is constant.
Preferably, the method for obtaining the linear aggregation degree of the included angle by using the fitting data of the included angle-time image and the number of included angle data contained in the sampling area of each sampling point comprises the following steps:
fitting by utilizing included angle data contained in the included angle-time image to obtain an included angle linear regression function;
obtaining the linear offset of the included angle by utilizing the slope and intercept of the linear regression function of the included angle obtained by fitting and the linear regression function of the theoretical included angle;
acquiring the abnormal degree of the included angle of each sampling point in the included angle-time image by using the number of included angle data contained in the sampling area of each sampling point in the included angle-time image;
and obtaining the included angle linear aggregation degree of each sampling point in the included angle-time image by using the included angle abnormality degree and the included angle linear offset degree of each sampling point in the included angle-time image.
The beneficial effects of the invention are as follows: detecting displacement, included angle and stress data of each time node of the excavator movable arm in a plurality of working periods to obtain displacement data, included angle data and stress data, obtaining weights in different dimensions by using the obtained multidimensional data, obtaining weighting factors by using the multidimensional weights, weighting an LOF algorithm by using the obtained weighting factors, detecting each sampling point by using the weighted LOF algorithm, and judging the stability of the excavator movable arm in a set period according to a detection result; the LOF algorithm is weighted through the multidimensional data, the weighted LOF algorithm is used for detecting the data of the sampling points, whether the data on the sampling points are abnormal or not can be accurately obtained, the stability of the excavator is judged through the number of the obtained data abnormality, the stability is judged through whether the data of the excavator are abnormal or not after the multidimensional data of the excavator are weighted, and the accuracy of stability judgment is effectively improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic flow chart of the method of the present invention;
FIG. 2 is a schematic view of an excavator according to the present invention;
FIG. 3 is a displacement-time image in an embodiment of the invention;
in the figure: 1. a movable arm; 2. extending arms; 3. a boom cylinder; 4. an arm extending oil cylinder; 5. a bucket; 6. bucket cylinder.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
As shown in fig. 1, the embodiment provides a method for detecting stability of a boom of an excavator, including:
the displacement sensor, the angle sensor and the force sensor are arranged on the excavator and are used for collecting data of displacement, angle and force of the movable arm in the working process of the excavator; since the boom 1 is driven to move by the boom cylinder 3 when the excavator works as shown in fig. 2, the boom 2 is driven to rotate along the end part of the boom 1, which is close to the bucket 5, by the boom cylinder 4, and the bucket 5 is driven to work by the bucket cylinder 6; when the movable arm 1 moves, the movable arm is pushed to move through the movable arm oil cylinder 3, so that the displacement of the movable arm 1 is equal to the expansion and contraction amount of the movable arm oil cylinder 3, the displacement sensor is arranged at the end part of the movable arm oil cylinder 3 and used for detecting the expansion and contraction amount of the movable arm oil cylinder 3, so that the displacement of the movable arm 1 can be effectively detected, the movable arm 1 and the movable arm 2 are rotationally linked through the rotating shaft, the movable arm 2 is driven to rotate along the end part of the movable arm 1 through the movable arm oil cylinder 4, an angle sensor is arranged between the movable arm 1 and the movable arm 2 and used for detecting the angle between the movable arm 1 and the movable arm 2 during operation, and meanwhile, a force sensor is arranged at two ends of the movable arm 1 and used for detecting the stress conditions of the two ends of the movable arm 1 and used for analyzing the force of the movable arm 1 through the stress conditions of the two ends of the movable arm 1;
the displacement, the included angle and the force of the movable arm are detected through a displacement sensor, an angle sensor and a force sensor which are respectively arranged on the excavator, specifically, the working period of the excavator is acquired, and the data of time nodes in each period are randomly acquired;
specifically, in the set period, the initial state of the operation period is set so that the distance at point C in FIG. 2The nearest point distance is set to be 0; when the included angle between the movable arm 1 and the cantilever arm 2 at the point F is minimized, the initial state of the mechanical arm of the excavator is set as +.>The method comprises the steps of carrying out a first treatment on the surface of the During the operation, let point C and +.>The displacement between the points increases from a minimum ramp rate, the displacement value reaches a maximum value at the end of the operating cycle, the maximum value is set to +.>The method comprises the steps of carrying out a first treatment on the surface of the The included angle between the movable arm at the point F and the extending arm is increased from the minimum uniform speed, the angle is changed to the maximum at the end of the operation period, and the maximum value of the included angle is set as +.>Ending the operation period; before the next operation period is carried out, the state of the mechanical arm is restored to the state of the most initial operation period;
the method comprises the steps of setting an operation period as T, collecting data of g time points in each period, collecting sensor data of n operation periods in the method, taking a checked value 20 for g, setting interval time for obtaining data in one period as T, collecting data of a plurality of indexes at each time point, and collecting the data from three dimensions of displacement, angle and force respectively.
In the movement process of the mechanical arm, the movable arm oil cylinder 3 generates displacement along with the movement of the mechanical arm, and a displacement sensor is arranged on the movable arm oil cylinder 3 to acquire displacement data of the movable arm oil cylinder, namely the distance of a C point in fig. 2Displacement change data of the points; the mechanical arm generates vibration in the moving process, force sensors are arranged at the point D and the point F in the figure 2, and the data of hinge point forces on the point D and the point F are obtained through the vibration frequency of the mechanical arm; an angle sensor is installed at the point F to acquire angle data between the boom 1 and the arm 2 during the periodic operation.
Preprocessing the acquired data, and removing the data with obvious abnormality, for example, on the boom cylinder 3The displacement of the point is defined within a specified range, the length of the movable arm oil cylinder is not exceeded, and if displacement data exceeding the length of the movable arm oil cylinder exists, the data are removed; the angle between the movable arm and the arm is also within a certain range, and the minimum angle between the movable arm and the arm is +.>The maximum angle is +.>If the adopted angle data exceeds the range, rejecting the data, and obtaining data (displacement data, angle data and stress data) of all time nodes in a set period according to the data;
obtaining a displacement linear polymerization degree by using fitting data of the displacement-time image and the number of displacement data contained in a sampling area of each sampling point: performing linear function fitting by using the displacement-time image to obtain displacement linear offset; obtaining the displacement abnormality degree of each sampling point in the displacement-time image by using the number of the displacement data contained in the sampling area of each sampling point in the displacement-time image; and obtaining the displacement linear polymerization degree of each sampling point in the displacement-time image by using the displacement abnormality degree and the displacement linear offset degree of each sampling point in the displacement-time image.
The method for obtaining the linear aggregation degree of the included angle by utilizing the fitting data of the included angle-time image and the number of included angle data contained in the sampling area of each sampling point comprises the following steps: fitting by utilizing included angle data contained in the included angle-time image to obtain an included angle linear regression function; obtaining the linear offset of the included angle by utilizing the slope and intercept of the linear regression function of the included angle obtained by fitting and the linear regression function of the theoretical included angle; acquiring the abnormal degree of the included angle of each sampling point in the included angle-time image by using the number of included angle data contained in the sampling area of each sampling point in the included angle-time image; and obtaining the included angle linear aggregation degree of each sampling point in the included angle-time image by using the included angle abnormality degree and the included angle linear offset degree of each sampling point in the included angle-time image.
Obtaining displacement linear offset and included angle linear offset by using linear functions fitted by the displacement data and the included angle data of each time node in a set period respectively;
in order to facilitate the person skilled in the art how to obtain the displacement linear offset and the included angle linear offset, the following exemplifies the obtained displacement linear offset and included angle linear offset;
the displacement sensor obtains displacement data in the operation process in the period T, and the first operation period is set asDraw +.>Displacement time image of displacement of boom cylinder 3 in period as time varies, in period +.>Reading data every interval time t, and counting the acquired data into a displacement-time image shown in fig. 3, wherein the abscissa in the displacement-time image is a time node, the ordinate is displacement data, and the same applies to the displacement-time image to obtain the displacement-time image in ≡>The included angle-time image and the stress-time image in the period are respectively taken as the ordinate by taking the displacement data and the angle data acquired in n periods, and the acquisition time of each data is taken as the abscissa, so that the corresponding displacement-time image and the included angle-time image are obtainedThe method comprises the steps of carrying out a first treatment on the surface of the And taking the data of the force acquired at the point D in n periods as an ordinate and the data of the force acquired at the point F as an abscissa, and obtaining a stress-time image of the corresponding relationship of the force change of the point D and the point F along with time. The angle-time image and the force-time image in this embodiment are similar to the displacement-time image (the abscissa is a time node), and belong to coordinate images known in the art, and specific images of the angle-time image and the force-time image are not described in detail herein.
When the linear displacement offset is obtained, as the displacement of the movable arm oil cylinder 3 is larger along with the time change in a single operation period of the excavator, the displacement of the movable arm oil cylinder 3 is in direct proportion to the time change; if no anomaly occurs in the data in a single period in n periods of the statistical data, the data in the current period and the data in the previous period are very similar. Meanwhile, some variable data are inevitably generated in the operation process of the excavator, for example, although the operation process of the excavator in each period is identical, the operation of the excavator in the actual operation process is not identical and may deviate slightly due to the factors of manual operation, and when the sensor data are sampled at intervals t in a single period, the obtained data deviate slightly from the data in the previous and later periods; or during the operation of the excavator, the speed is slightly higher or lower according to the original rule, so that the time of the next period of one operation flow can be slightly longer or shorter, and the sampling interval t is also influenced.
Therefore, the collected displacement data are analyzed, and in one period, the change relation between the displacement data of the normal movable arm oil cylinder and the time accords with a linear regression model, and after the data in the period are combined together, the change relation between the normal displacement and the time also accords with the linear regression model; combining the periodic displacement-time images into a displacement-time image formed by all displacement data, wherein each time node is provided with a plurality of sampling points which are gathered together, and the normal sampling points have higher local density;
in the displacement-time image, when the displacement data are all normal points, the data acquired in the displacement-time image accord with a linear regression model, and the local density of the sampling points is relatively close; if abnormal data appear in the process of the excavator operation, the abnormal sampling points do not accord with the linear regression model, and the local density of the abnormal points also changes.
Fitting according to the acquired displacement data, and fitting a linear regression model of the displacement data by a least square method; the data may contain abnormal data points, the abnormal data points are also taken as statistical samples into sample points of the linear regression in the fitting process, the sampling points with obvious abnormality are removed in the preprocessing process of the data, the rest sampling points represent the real situation in the operation process of the excavator, and the fitted linear regression model has better effect. Because the excavator is manually controlled in the operation process, the displacement of the movable arm oil cylinder is increased along with the time in the period, and the minimum value 0 and the maximum value of the displacement are calculated on the basis of theoryAll are fixed values, the operation period T is fixed value, and the change process is uniform, so that the displacement time linear function is determined under normal conditions. Linear regression function obtained from fitting sample points>And a theoretically calculated displacement time linear function +.>The degree of abnormality of the data points is analyzed.
Let us set a linear regression function derived by sample point fittingThe method comprises the following steps:
theoretically calculated linear function of displacement timeThe method comprises the following steps:
constructing a displacement linear offsetThe formula is:
wherein,is a linear regression function->Slope of>Linear function of displacement time->Slope of>Is a linear regression function->The larger the displacement linear offset between the linear regression function and the displacement time linear function is, the larger the offset degree of the linear regression function is, and the more abnormal data points are collected; the smaller the linear offset between the linear regression function and the displacement time linear function, the smaller the displacement offset of the linear regression function, and the fewer abnormal data points in the acquired data.
Obtaining the angular linear offset according to the method for obtaining the displacement linear offset；
The characteristic value is set according to the density of the sampling points to reflect the abnormality degree of each sampling point, and the sampling points at the same sequence sampling time in each period are gathered together at a certain density in a displacement-time image/an included angle-time image from the density analysis of the sampling points, for example, the first sampling time in the first period and the first sampling time in the nth period are the same sequence sampling time. The local density of the normal sampling point and the local density of the abnormal point have certain deviation. From the image analysis, the number of the collected sampling points in the neighborhood range of the normal sampling points is more, the sampling points are denser, and the local density is higher; the number of sampling points in the set in the neighborhood range of the abnormal sampling points is smaller, the abnormal sampling points are scattered, and the local density is smaller.
Respectively determining the sampling area of each sampling point in the displacement-time image and the included angle-time image by using the Euclidean distance between each sampling point in the displacement-time image and the included angle-time image and the nearest sampling point of the sampling point;
the method comprises the steps of obtaining Euclidean distance between each sampling point and the nearest sampling point in a displacement-time image and an included angle-time image;
obtaining the sampling radius of each sampling point according to the Euclidean distance between each sampling point and the nearest sampling point in the displacement-time image/included angle-time image on each time node;
and acquiring the sampling area of each sampling point in the displacement-time image and the included angle-time image by using the sampling radius of each sampling point in the displacement-time image and the included angle-time image.
The embodiment uses the sampling area of each sampling point in the acquired displacement-time image for detailed description;
in the displacement-time image, randomly selecting one sampling point on the displacement-time image for illustration, setting the sampling point as a sampling point i, solving the Euclidean distance between sampling points closest to the sampling point i, traversing all the sampling points in the displacement-time image in sequence, calculating the corresponding Euclidean distance by each sampling point, averaging the obtained Euclidean distances, and setting the average of the obtained Euclidean distancesThe value isThen the sampling point i is taken as the center, and the Euclidean distance average value is taken as +>Is a radius; taking the sampling point as the center point of the sampling area, and the Euclidean distance average value +.>In order to obtain a sampling area of each sampling point for a sampling radius, a euclidean distance algorithm conventional to those skilled in the art is adopted in the embodiment, and a detailed description of the euclidean distance algorithm is omitted here; the method for acquiring the sampling range of each sampling point in the displacement-time image is similar to the method for acquiring the sampling area of each sampling point in the included angle-time image;
after the sampling range of each sampling point in the displacement-time image is acquired, acquiring the displacement abnormality degree of each sampling point in the displacement-time image by utilizing the number of displacement data contained in the sampling area of each sampling point in the displacement-time image;
specifically, the number of sampling points in each sampling area is counted, and the number of sampling points in the area is set as. If the sampling point i is a normal point, the Euclidean distance between the sampling point i and the nearest neighbor point is smaller; if the sampling point i is an abnormal point, the Euclidean distance between the sampling point i and the nearest neighbor point is larger. After the average value is obtained for all Euclidean distances, the area of the normal sampling point with the Euclidean distance as the radius is increased, the area of the abnormal sampling point with the Euclidean distance as the radius is reduced, the number of the sampling points in the adjacent areas of the normal sampling point is increased, and the number of the sampling points in the adjacent areas of the abnormal sampling point is reduced; the dense index is used for measuring the displacement abnormality degree of the sampling point i.
Wherein,for the total number of sampling points in each time node, +.>The number of sampling points in the sampling area of the sampling point i. The larger the number of sampling points in the adjacent area of the sampling point i is, the larger the density index of the sampling point i is, and the displacement abnormality degree is +.>The smaller the size; the smaller the number of sampling points in the adjacent area of the sampling point i is, the smaller the density index of the sampling point i is, and the displacement abnormality degree is +.>The larger.
Similarly, the displacement abnormality degree of each sampling point in the displacement-time image is obtainedThe method of (1) obtains the abnormal degree of the included angle of each sampling point in the included angle-displacement image>。
Obtaining the displacement linear polymerization degree of each sampling point in the displacement-time image by utilizing the displacement abnormality degree and the displacement linear offset degree of each sampling point in the displacement-time image;
specifically, in the dimension in which the boom cylinder 3 is displaced, the linear deviation is obtained by displacementAnd degree of displacement abnormalityConstruction of Displacement Linear polymerization degree->Linear degree of polymerization according to displacement->The weight of the abnormal sampling point in the displacement dimension of the movable arm oil cylinder 3 is set, the sampling point with larger displacement abnormality degree is given higher weight, and the sampling point with smaller displacement abnormality degree is given lower weight.
Wherein,is a linear regression function->Is a displacement linear offset of ∈10->For the displacement abnormality degree of the sampling point i, the larger the displacement linear offset degree is, the more the number of abnormal sampling points in the sampling points is, and the smaller the displacement linear aggregation degree is; the smaller the degree of displacement abnormality of the sampling point i, the greater the degree of abnormality thereof, and the smaller the degree of displacement linear aggregation. In the displacement dimension, the linear degree of polymerization of the displacement of the sampling point i +.>The smaller the weight that should be given, the greater the weight; linear degree of polymerization of displacement->The larger the weight that should be given, the smaller the weight.
Similarly, in the angular dimension, the function of angular time is likewise a linear function, but the initial value is not zero, the angle of point F is at a minimumI.e. the intercept of the linear function of angular time is +.>. Similarly, fitting the sampling points in the included angle-time image to obtain a linear regression function of the included angle, and constructingAnd (5) establishing an included angle linear offset to measure the included angle linear offset of the linear regression function. And constructing a dense index according to each sampling point in the included angle-time image to measure the included angle abnormality degree of the sampling points, and constructing the included angle linear aggregation degree through the included angle linear offset degree and the included angle abnormality degree. Let the angular linear concentration of the sampling point i in the angular dimension be +.>. In the angular dimension, the angular linear concentration of the sampling point i +.>The smaller the weight that should be given, the greater the weight; angular linear concentration +.>The larger the weight that should be given, the smaller the weight.
Analyzing each stress data in the obtained stress-time image; detecting the limit stress conditions of the two ends of the movable arm 1 through force sensors arranged at the two ends of the movable arm 1, namely acquiring forces at a point D and a point F as shown in fig. 2, wherein stress data of the point D and the point F show a certain functional relation in the operation period of the excavator, and a stress linear function is obtained by using stress data level reverse fitting at the point D and the point F; at the position ofIn the period, stress analysis is carried out on hinge point forces of the point D and the point F every t, the stress is analyzed by neglecting unbalanced load and friction force of a hinge point pin shaft, the point D and the point F only bear forces in the X direction and the Y direction, the stress of the point D and the point F in the X direction and the Y direction is analyzed, and the magnitude of resultant force is obtained to serve as ideal stress data of the point D and the point F; for period->After the hinge point forces of the points D and F in the inner part are analyzed, fitting the analyzed ideal stress data to obtain an ideal stress linear function; performing function fitting using a least squares method when performing the fitting of the stressed linear function and the ideal stressed linear functionIs a function fitting method which is conventional in the art, and a detailed description of how it is fitted is not provided herein;
by using a fitted force linear functionAnd ideal stress linear function->A cross-correlation function E is constructed, which is represented by the following expression:
the average value of the hinge point force of the point F obtained by the first sampling in n periods T is obtained and is set asThe average value of the hinge point force of the point F obtained by the last sampling in n periods T is obtained by the same principle and is set as +.>。
Will be%) As a function->And function->Upper and lower limits of convolution are obtained to obtain corresponding values; the larger the value of the cross-correlation function E, the function +.>Sum function->The greater the correlation between the two, the fewer the number of abnormal sampling points in the sampling points; the smaller the value of the cross-correlation function E, the function +.>Sum function->The smaller the correlation between the two, the greater the number of abnormal sampling points among the sampling points.
In the stress-time image, firstly, the stress abnormality degree of each sampling point is obtained by utilizing the distance from the sampling point to the original point;
specifically, the coordinates of the sampling point i are set asIs provided with
In the method, in the process of the invention,representing the distance from the sampling point i to the coordinate origin, wherein the distances from the normal sampling point at the same sampling moment to the coordinate origin are similar, solving the mean value and standard deviation of the distances from n sampling points on each time node to the coordinate origin, and setting the mean value as +.>Standard deviation of->The number of the difference standard deviation of the average value of the distance between the sampling point i and the sampling time is obtained through a Z-Score formula, the stress abnormality degree of the sampling point i is reflected, and the Z-Score formula is as follows:
wherein,the value of (2) reflects the degree of stress abnormality of the sampling point i,/-, and>for the distance of the sampling point i from the origin of coordinates, +.>Is the average value of the distances from all sampling points to the origin in the sampling time of the sampling point i, +.>The standard deviation of the distances from the origin to all the sampling points in the sampling time of the sampling point i. />The greater the absolute value of the sampling point i is, the greater the degree of the distance from the average value is, and the greater the degree of stress abnormality is; />The smaller the absolute value of the sampling point i is, the smaller the degree of the distance from the average value is at the sampling moment, and the smaller the degree of stress abnormality is.
Then, the mean value of the Euclidean distance between each sampling point and other sampling points on each time node and the abnormal degree of the sampling point are utilized to obtain the fluctuation abnormal degree of each sampling point in each stress-time image;
specifically, the average value of Euclidean distances between the sampling point i and other sampling points on the same time node is calculated, and the larger the average value of Euclidean distances is, the more the sampling point i is separated from the group, and the greater the degree of abnormality is; the smaller the mean value of the Euclidean distance is, the more the sampling points i are clustered, and the smaller the degree of abnormality is. The mean value of Euclidean distances between the sampling point i and other sampling points at the same moment is set asBuild up of heave anomalyThe formula is:
wherein the abnormal wavinessReflecting the function obtained after fitting the sampling point i>Influence of production->Reflecting the degree to which the sampling point i deviates from the mean value at its sampling instant, +.>The greater the degree of deviation from the mean, the greater the degree of deviation from the meanThe greater the heave effect of (c); />The smaller the degree of deviation from the mean, the +.>The smaller the heave effect of (c);the degree of outlier, < +.>The larger the sampling point i is, the more outlier is, for the function +.>The greater the heave effect of (c);the smaller the sampling points i the more clustered the function +.>The less the heave effect of (c).
By the value of the cross-correlation function E and the degree of fluctuation anomalyThe relevant intensity is constructed and the relevant concentration is built,let the related concentration->The formula is
Wherein the larger the value of E, the functionSum function->The bigger the correlation between the two is, the fewer the number of abnormal sampling points is, and the bigger the correlation density is; relief abnormality->The smaller the value of (c) is, the smaller the degree of abnormality of the sampling point i is, and the greater the correlation density is. The greater the correlation density, the smaller the degree of abnormality of the sampling point i, and the smaller the weight given; the smaller the correlation density, the greater the degree of abnormality of the sampling point i, and the greater the weight given.
Then, obtaining a weighting factor of each sampling point by using the displacement linear aggregation degree, the included angle linear aggregation degree and the fluctuation anomaly degree of each sampling point;
the sampling points contained in each time node comprise sampling points in a displacement-time image, an included angle-time image and a stress-time image in the same sampling time node;
specifically, the expression for obtaining the weighting factor for each time node is as follows:
/>
wherein:the displacement dimension weight of the ith sampling point; />The included angle dimension weight of the ith sampling point; />The stress dimension weight of the ith sampling point; />The displacement linear polymerization degree is the ith; />Linear polymerization degree of included angle of the ith; />Force-bearing linear degree of polymerization of the ith; />Acquiring the total number of sampling points in n periods; />Representing the sum of the displacement linear aggregation degrees of all sampling points; />Representing the sum of the linear aggregation degrees of the included angles of the sampling points; />Representing the sum of the stress linear polymerization degrees of all sampling points;
obtaining a weighting factor of each sampling point by using the displacement dimension weight, the included angle dimension weight and the stress dimension weight of each sampling point;
the expression of the weighting factor is as follows:
wherein: p is the weighting factor of each sampling point;a displacement dimension weight for each sampling point; />The included angle dimension weight of each sampling point; />Stress dimension weight for each sampling point, +.>Is a constant;
as can be seen from the expression, the greater the linear aggregation degree of the sampling point i is, the smaller the degree of abnormality is, and the weight isThe smaller; the greater the included angle aggregation degree of the sampling point i, the smaller the degree of abnormality, and the weight +.>The smaller; the greater the correlation density of the sampling points i, the smaller the degree of abnormality, the weight +.>The smaller. Weighting of three dimensions as weighting factor for each time data>I.e. the greater the degree of abnormality of the sampling point i, +.>The larger the value of (2), the larger the weight given by the sampling point i; the smaller the degree of abnormality of the sampling point i is +.>The smaller the value of (c), the smaller the weight given to the sampling point i.
Weighting the LOF algorithm by using the weighting factors of the obtained sampling points;
specifically, the weighting factor P improves the local reachable density, and when abnormal data is detected by the LOF local abnormality factor, the higher the local reachable density of the sampling points is, the more likely the data points are in the same cluster; the lower the local reachable density, the more likely it is an outlier. Therefore, the local reachable density is improved by the weighting factor according to the embodiment, and the improved formula is that
Wherein,representing the inverse of the average reachable distance from the point in the kth neighborhood of the sampling point i to i, wherein P is a weighting factor; />Representing the distance from a point o to the kth field of the sampling point i, wherein the point o is a point within the reachable distance of the kth field of the sampling point i, the greater the weighting factor, the lower the density, the greater the degree of abnormality and the more likely to be an outlier; the smaller the weighting factor, the higher the density, the smaller the degree of abnormality, and the more likely to be a normal point. After the improved local reachable density is obtained, the local outlier factor of the sampling point i is obtained, the calculation of the local outlier factor is a known technology, and the detailed process is not repeated.
Detecting each sampling point by using a weighted LOF algorithm, judging the stability of the excavator movable arm in a set period according to the detection result of the sampling point, specifically, acquiring a local outlier factor LOF of the sampling point i through the steps, and if the local outlier factor LOF of the sampling point i is closer to 1, the local outlier factor LOF is more likely to belong to the same cluster as the neighborhood; the density of the sampling points i is higher than the density of the neighborhood points of the sampling points i, namely the dense points, the smaller the sampling points i are smaller than 1; the greater 1 indicates that the density of the sampling points i is less than the density of the neighborhood points, and the more likely to be abnormal points. Sampling points with local outlier factors LOF greater than 1.2 are abnormal sampling points.
Setting abnormality rateThe formula is:
wherein p is the number of abnormal sampling points, and l is the total number of sampling points acquired in n periods. The more the number of the abnormal sampling points is, the higher the abnormal rate is, and the lower the stability of the movable arm of the excavator is; the fewer the number of abnormal sampling points, the lower the abnormal rate, and the higher the stability of the excavator boom.
Thus, the detection of the stability of the movable arm of the excavator is completed.
According to the scheme provided by the embodiment, displacement data, included angle data and stress data are obtained by utilizing each sampling point, weights in different dimensions are obtained, weighting factors are obtained by utilizing the weights in multiple dimensions, an LOF algorithm is weighted by utilizing the obtained weighting factors, each sampling point is detected by utilizing the weighted LOF algorithm, and the stability of the excavator movable arm in a set period is judged according to the detection result; the LOF algorithm is weighted through the multidimensional data, the weighted LOF algorithm is used for detecting the data of the sampling points, whether the data on the sampling points are abnormal or not can be accurately obtained, the stability of the excavator is judged through the number of the obtained data abnormality, the stability is judged through whether the data of the excavator are abnormal or not after the multidimensional data of the excavator are weighted, and the accuracy of stability judgment is effectively improved.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (6)
1. The excavator movable arm stability detection method is characterized by comprising the following steps:
acquiring a plurality of displacement data of the movable arm of the excavator, which is close to the end part of the bucket, a plurality of included angle data of the movable arm and the extending arm and a plurality of stress data of the two ends of the movable arm at each time node in a plurality of periods; respectively constructing a displacement-time image, an included angle-time image and a stress-time image by using the displacement data, the included angle data and the stress data;
obtaining displacement linear polymerization degree by utilizing fitting data of the displacement-time image and the number of displacement data contained in a sampling area of each sampling point;
obtaining the linear aggregation degree of the included angles by using the fitting data of the included angle-time images and the number of included angle data contained in the sampling area of each sampling point;
performing cross-correlation function fitting according to stress data of each sampling point in the stress-time image to obtain a cross-correlation value; the fluctuation anomaly degree of each sampling point is obtained by utilizing the distance from the sampling point to the origin of coordinates in the stress-time image;
respectively utilizing the displacement linear aggregation degree, the included angle linear aggregation degree and the fluctuation anomaly degree of each sampling point to obtain the weighting factors of the sampling points;
weighting the LOF algorithm by using a weighting factor of the sampling point, detecting the sampling point by using the weighted LOF algorithm, and judging the stability of the movable arm of the excavator in a set period according to a detection result;
the method for obtaining the displacement linear polymerization degree comprises the following steps:
performing linear function fitting by using the displacement-time image to obtain displacement linear offset;
obtaining the displacement abnormality degree of each sampling point in the displacement-time image by using the number of the displacement data contained in the sampling area of each sampling point in the displacement-time image;
obtaining the displacement linear polymerization degree of each sampling point in the displacement-time image by utilizing the displacement abnormality degree and the displacement linear offset degree of each sampling point in the displacement-time image;
the method for obtaining the linear aggregation degree of the included angle by utilizing the fitting data of the included angle-time image and the number of included angle data contained in the sampling area of each sampling point comprises the following steps:
fitting by utilizing included angle data contained in the included angle-time image to obtain an included angle linear regression function;
obtaining the linear offset of the included angle by utilizing the slope and intercept of the linear regression function of the included angle obtained by fitting and the linear regression function of the theoretical included angle;
acquiring the abnormal degree of the included angle of each sampling point in the included angle-time image by using the number of included angle data contained in the sampling area of each sampling point in the included angle-time image;
obtaining the included angle linear aggregation degree of each sampling point in the included angle-time image by utilizing the included angle abnormality degree and the included angle linear offset degree of each sampling point in the included angle-time image;
the method for determining the sampling area of each sampling point in the included angle-time image comprises the following steps:
acquiring Euclidean distance between each sampling point and the nearest sampling point in the included angle-time image;
obtaining the sampling radius of each sampling point according to the included angle on each time node, namely the Euclidean distance between each sampling point and the nearest sampling point in the time image;
acquiring a sampling area of each sampling point in the included angle-time image by using the sampling radius of each sampling point in the included angle-time image;
the method for obtaining the cross-correlation value by performing cross-correlation function fitting according to the stress data of each sampling point in the stress-time image comprises the following steps:
fitting by using stress data at two ends of a movable arm on each sampling point to obtain a stress linear function;
carrying out resultant force analysis on the stress data at the two ends of the movable arm on each sampling point, and taking the stress data after the resultant force analysis on each sampling point as ideal stress data of the movable arm on the sampling point;
fitting the obtained ideal stress data of the movable arm on each sampling point to obtain an ideal stress linear function;
constructing a stress cross-correlation function by using the obtained stress linear function and the ideal linear function;
and obtaining a cross-correlation value by using the stress cross-correlation function and the ideal cross-correlation function.
2. The excavator boom stability detection method of claim 1 wherein the method of obtaining displacement linear excursions comprises:
fitting by using displacement data contained in the displacement-time image to obtain a displacement linear regression function;
and obtaining the displacement linear offset by utilizing the slope and intercept of the displacement linear regression function obtained by fitting and the theoretical displacement linear regression function.
3. The excavator boom stability detection method of claim 1 wherein the method of determining the sampling region for each sampling point in the displacement-time image comprises:
acquiring Euclidean distance between each sampling point and the nearest sampling point in the displacement-time image;
obtaining the sampling radius of each sampling point according to the Euclidean distance between each sampling point and the nearest sampling point in the displacement-time image on each time node;
and acquiring a sampling area of each sampling point in the displacement-time image by using the sampling radius of each sampling point in the displacement-time image.
4. The method for detecting the stability of a boom of an excavator according to claim 1, wherein the method for acquiring the heave anomaly of each sampling point by using the distance from the sampling point to the origin of coordinates in the stress-time image comprises:
obtaining the stress abnormality degree of each sampling point in the stress-time image by utilizing the distance from each sampling point in the stress-time image to the origin of coordinates; and obtaining the fluctuation anomaly degree of each sampling point in the stress-time image by using the stress anomaly degree of each sampling point in the stress-time image.
5. The method for detecting the stability of a boom of an excavator according to claim 4, wherein the step of obtaining the undulation anomaly degree of each sampling point in the force-time image using the force anomaly degree of each sampling point in the force-time image comprises:
and obtaining the fluctuation anomaly degree of each sampling point in each stress-time image by using the mean value of the Euclidean distance between each sampling point and other sampling points on each time node and the anomaly degree of the sampling point.
6. The method for detecting the stability of a boom of an excavator according to claim 1, wherein the method for obtaining the weighting factor of the sampling point comprises:
obtaining displacement dimension weight, included angle dimension weight and stress dimension weight of each sampling point by utilizing the displacement linear aggregation degree, included angle linear aggregation degree and fluctuation anomaly degree of each sampling point respectively;
obtaining a weighting factor of each sampling point by using the displacement dimension weight, the included angle dimension weight and the stress dimension weight of each sampling point;
the expression of the weighting factor is as follows:
wherein P is the weighting factor of each sampling point;a displacement dimension weight for each sampling point; />The included angle dimension weight of each sampling point; />Stress dimension weight for each sampling point, +.>Is constant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311481983.0A CN117216728B (en) | 2023-11-09 | 2023-11-09 | Excavator movable arm stability detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311481983.0A CN117216728B (en) | 2023-11-09 | 2023-11-09 | Excavator movable arm stability detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117216728A CN117216728A (en) | 2023-12-12 |
| CN117216728B true CN117216728B (en) | 2024-02-02 |
Family
ID=89046622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311481983.0A Active CN117216728B (en) | 2023-11-09 | 2023-11-09 | Excavator movable arm stability detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117216728B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106874914A (en) * | 2017-01-12 | 2017-06-20 | 华南理工大学 | A kind of industrial machinery arm visual spatial attention method based on depth convolutional neural networks |
| CN211401395U (en) * | 2020-03-26 | 2020-09-01 | 徐州徐工矿业机械有限公司 | Loading and weighing device for bucket of hydraulic excavator |
| WO2020228310A1 (en) * | 2019-10-17 | 2020-11-19 | 浙江大学 | Interval and bounded probability mixed uncertainty-based mechanical arm robustness optimization design method |
| CN113605483A (en) * | 2021-07-19 | 2021-11-05 | 江苏徐工工程机械研究院有限公司 | Automatic operation control method and device for excavator |
| CN114266457A (en) * | 2021-12-14 | 2022-04-01 | 国网山东省电力公司枣庄供电公司 | Method for detecting different loss inducement of distribution line |
| CN115059127A (en) * | 2022-07-25 | 2022-09-16 | 柳州柳工挖掘机有限公司 | Excavator stability evaluation system and method, controller and excavator |
| CN115077958A (en) * | 2022-05-27 | 2022-09-20 | 江苏徐工工程机械研究院有限公司 | Method and system for evaluating stroke stability of excavator working device |
| CN116108339A (en) * | 2023-02-21 | 2023-05-12 | 大连理工大学 | Actual measurement data outlier detection and correction method for tunnel boring machine |
| CN116852347A (en) * | 2023-03-14 | 2023-10-10 | 哈尔滨理工大学 | State estimation and decision control method for non-cooperative target autonomous grabbing |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101614013B1 (en) * | 2011-05-16 | 2016-04-29 | 스미도모쥬기가이고교 가부시키가이샤 | Shovel, monitoring device therefor, and shovel output device |
-
2023
- 2023-11-09 CN CN202311481983.0A patent/CN117216728B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106874914A (en) * | 2017-01-12 | 2017-06-20 | 华南理工大学 | A kind of industrial machinery arm visual spatial attention method based on depth convolutional neural networks |
| WO2020228310A1 (en) * | 2019-10-17 | 2020-11-19 | 浙江大学 | Interval and bounded probability mixed uncertainty-based mechanical arm robustness optimization design method |
| CN211401395U (en) * | 2020-03-26 | 2020-09-01 | 徐州徐工矿业机械有限公司 | Loading and weighing device for bucket of hydraulic excavator |
| CN113605483A (en) * | 2021-07-19 | 2021-11-05 | 江苏徐工工程机械研究院有限公司 | Automatic operation control method and device for excavator |
| CN114266457A (en) * | 2021-12-14 | 2022-04-01 | 国网山东省电力公司枣庄供电公司 | Method for detecting different loss inducement of distribution line |
| CN115077958A (en) * | 2022-05-27 | 2022-09-20 | 江苏徐工工程机械研究院有限公司 | Method and system for evaluating stroke stability of excavator working device |
| CN115059127A (en) * | 2022-07-25 | 2022-09-16 | 柳州柳工挖掘机有限公司 | Excavator stability evaluation system and method, controller and excavator |
| CN116108339A (en) * | 2023-02-21 | 2023-05-12 | 大连理工大学 | Actual measurement data outlier detection and correction method for tunnel boring machine |
| CN116852347A (en) * | 2023-03-14 | 2023-10-10 | 哈尔滨理工大学 | State estimation and decision control method for non-cooperative target autonomous grabbing |
Non-Patent Citations (3)
| Title |
|---|
| Ning Sun ; Tong Yang ; Yongchun Fang ; Biao Lu ; Yuzhe Qian.Nonlinear Motion Control of Underactuated Three-Dimensional Boom Cranes With Hardware Experiments. IEEE Transactions on Industrial Informatics .2017,第887-897页. * |
| 挖掘机实际挖掘仿真及动臂结构改进;蔡立;贾春玉;刘伟;;机械设计与制造(04);第59-62页 * |
| 胡均平 ; 乔榆凯 ; 李科军 ; 胡慧雨.直接提钻工况下双斜撑长螺旋钻机桅杆稳定性分析.工程设计学报.2018,第151-158页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117216728A (en) | 2023-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112629647B (en) | Real-time identification, monitoring and early warning method for vortex vibration event of large-span suspension bridge | |
| CN105787901B (en) | A kind of multiple dimensioned velocity field measurement method for adjacent two interframe in sun full resolution pricture sequence | |
| CN102840829B (en) | Manual mark-basedsystem and manual mark-based method for measuring displacement field inside high temperature object area | |
| CN108573224B (en) | Bridge structure damage positioning method for mobile reconstruction of principal components by using single sensor information | |
| CN111678698A (en) | Rolling bearing fault detection method based on sound and vibration signal fusion | |
| CN112284575B (en) | On-line bearing load monitoring method based on eddy current displacement sensor | |
| CN115033042B (en) | Intelligent control system of food dehydrator based on multi-mode Internet of things | |
| CN114627115B (en) | Dynamic balance detection system and fluid pump impeller production line using same | |
| CN117216728B (en) | Excavator movable arm stability detection method | |
| CN112179653A (en) | Rolling bearing vibration signal blind source separation method and device and computer equipment | |
| CN113340598A (en) | Rolling bearing intelligent fault diagnosis method based on regularization sparse model | |
| Shen et al. | Rate-dependent characteristics of damage and roughness degradation of three-dimensional artificial joint surfaces | |
| CN107726991B (en) | Digital image related strain field calculation scheme based on weak form | |
| JP2012073182A (en) | Method for testing material | |
| CN111412864B (en) | Automatic grinding crack angle detection method based on grinding crack gray scale similarity | |
| CN108844469B (en) | Method and system for testing workpiece step height based on laser | |
| CN115795341A (en) | Two-dimensional piston pump health state assessment method based on variable rotating speed | |
| US4334282A (en) | Testing apparatus | |
| KR102292602B1 (en) | Deep learning and image processing based bolt loosening detection method | |
| CN117491422B (en) | Method for detecting high heat-conducting property of aluminum alloy material | |
| CN117557574B (en) | Material parameter detection method and system based on image processing | |
| CN117109909B (en) | Detection and early warning method and system for mechanical sealing performance of large-shaft-diameter semi-split type machine | |
| CN112001234B (en) | Rotary mechanical sampling data clustering method based on affine invariant measurement | |
| CN114720110B (en) | Frequency identification method based on cyclic multi-scale reverse optimization layout of leaf-end timing sensor | |
| CN114720041B (en) | Arch bridge suspender construction cable force change monitoring method based on three-dimensional laser scanning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |