Method and system for monitoring equipment process processing data of hydraulic cylinder

Abstract

The invention relates to the technical field of hydraulic cylinder test data processing, in particular to a hydraulic cylinder equipment process processing data monitoring method and system. The method comprises the following steps: acquiring an initial abnormal point in a pressure sequence corresponding to each hydraulic cylinder in each test; constructing a reference straight line corresponding to the primary test; according to the distance from the initial abnormal point to the reference straight line in the test and the data points before and after the initial abnormal point, obtaining the spatial distribution characteristics of the initial abnormal point; calculating time distribution characteristics of an initial abnormal point according to the corresponding pressure sequences of the hydraulic cylinder corresponding to the initial abnormal point in other tests; obtaining an abnormal degree according to the spatial distribution feature and the time distribution feature of the initial abnormal point, and further obtaining an abnormal point; and obtaining the defect probability of the hydraulic cylinder corresponding to the pressure sequence based on the number of the abnormal points in the pressure sequence and the spacing distance between the two farthest abnormal points. The accuracy of abnormal detection of the hydraulic cylinder can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hydraulic cylinder test data processing, and in particular to a method and system for monitoring equipment process processing data of a hydraulic cylinder. Background Art

[0002] A hydraulic cylinder is a power device that uses a hydraulic system to generate linear mechanical motion. It is widely used in engineering machinery, automobiles, aviation, mining, ships and other industries. It drives and controls external loads by pushing the piston to move linearly through the pressure of hydraulic oil in a closed cylinder. Pressure deviation in the hydraulic system may cause large errors in the engineering process and pose great safety hazards. Therefore, monitoring hydraulic process data (such as pressure, flow, temperature, oil level, etc.) is crucial for the normal operation of the hydraulic system. Before the hydraulic cylinder is produced and shipped, it usually needs to undergo a series of monitoring and testing to ensure that its performance, reliability and safety meet the expected requirements.

[0003] The existing technology usually uses a threshold monitoring method to monitor the process data of hydraulic cylinders. However, this method requires manual screening based on a large amount of monitoring experience data, and the thresholds are different for hydraulic cylinders of different cylinder diameters. It is difficult to obtain an accurate threshold to evaluate the process performance of the hydraulic cylinder. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a method and system for monitoring equipment process data of a hydraulic cylinder. The technical solutions adopted are as follows: In a first aspect, an embodiment of the present invention provides a method for monitoring equipment process data of a hydraulic cylinder, the method comprising: Perform a preset number of tests on each hydraulic cylinder to obtain a pressure sequence corresponding to each hydraulic cylinder in each test; obtain an initial abnormal point in the pressure sequence of each hydraulic cylinder in a test; The reference straight line is obtained by linear fitting using the data points except the initial abnormal point in the pressure sequence of each hydraulic cylinder in the same test; the spatial distribution characteristics of the initial abnormal point are obtained according to the distance from the initial abnormal point to the reference straight line in the test and the data points before and after the initial abnormal point; Calculate the time distribution characteristics of an initial abnormal point according to the pressure sequences corresponding to the hydraulic cylinder corresponding to the initial abnormal point in other tests; obtain the abnormal degree according to the spatial distribution characteristics and time distribution characteristics of the initial abnormal point; Whether the initial abnormal point is an abnormal point is determined based on the abnormal degree of the initial abnormal point; and the defect probability of the hydraulic cylinder corresponding to the pressure sequence is obtained based on the number of abnormal points in a pressure sequence and the interval distance between the two most distant abnormal points.

[0005] Preferably, obtaining the initial abnormal point in the pressure sequence of each hydraulic cylinder in a test includes: The local anomaly factor algorithm is used to obtain the initial abnormal points in the pressure sequence of each hydraulic cylinder in a test.

[0006] Preferably, the spatial distribution characteristics of the initial outlier point are obtained according to the distance from the initial outlier point to the reference straight line and the data points before and after the initial outlier point in the test, including: Obtain the ratio of the distance from the initial outlier point to the reference straight line to the maximum value of the distances from each initial outlier point to the reference straight line, recorded as the normalized distance; obtain four data points adjacent to the initial outlier point in the sequence where the initial outlier point is located, wherein the data point closest to the initial outlier point among the adjacent data points before the initial outlier point is the first data point, and the other data point is the second data point, and the data point closest to the initial outlier point among the adjacent data points after the initial outlier point is the third data point, and the other data point is the fourth data point; calculate the Euclidean distances between the first data point, the third data point and the initial outlier point respectively and sum them to obtain the first distance sum; calculate the Euclidean distance between the first data point and the second data point and the Euclidean distance between the third data point and the fourth data point respectively, and sum them to obtain the second distance sum; use an exponential function with a natural constant as the base to negatively correlate the ratio of the first distance sum to the second distance sum to obtain a mapping result; the sum of the mapping result and the normalized distance is the spatial distribution feature of the initial outlier point.

[0007] Preferably, calculating the time distribution characteristics of an initial abnormal point according to each pressure sequence corresponding to the hydraulic cylinder corresponding to the initial abnormal point in other tests includes: The pressure sequences corresponding to the hydraulic cylinder corresponding to an initial abnormal point in other tests are recorded as pressure sequences to be compared; the data points in the same position as the initial abnormal point in each pressure sequence to be compared are obtained and recorded as data points to be compared; the sum of the absolute values ​​of the differences between the initial abnormal point and the corresponding data points to be compared is negatively correlated using an exponential function with a natural constant as the base to obtain the pressure value similarity; the sum of the absolute values ​​of the differences in the spatial distribution characteristics between the initial abnormal point and the corresponding data points to be compared is negatively correlated using an exponential function with a natural constant as the base to obtain the spatial distribution characteristic similarity; the sum of the pressure value similarity and the spatial distribution characteristic similarity is the temporal distribution characteristic degree of the initial abnormal point.

[0008] Preferably, obtaining the abnormality degree according to the spatial distribution characteristics and temporal distribution characteristics of the initial abnormal point includes: The spatial distribution characteristics and temporal distribution characteristics of the initial abnormal point are added and compared with the preset value to obtain the abnormal degree of the initial abnormal point.

[0009] Preferably, judging whether the initial abnormal point is an abnormal point based on the abnormality degree of the initial abnormal point includes: A first threshold is set. If the abnormality degree of an initial abnormal point is greater than or equal to the first threshold, the initial abnormal point is an abnormal point.

[0010] Preferably, obtaining the defect probability of the hydraulic cylinder corresponding to a pressure sequence based on the number of abnormal points in a pressure sequence and the interval distance between two abnormal points that are farthest apart includes: The ratio of the number of abnormal points in a pressure sequence to the number of data points in the pressure sequence is obtained, which is recorded as the number ratio. The ratio of the interval distance between the two abnormal points that are farthest apart in the pressure sequence to the length of the pressure sequence is obtained, which is recorded as the range ratio. The product of the number ratio and the range ratio is the defect probability of the hydraulic cylinder corresponding to the pressure sequence. Among them, the interval distance between the two abnormal points that are farthest apart in the pressure sequence is the number of data points between the two abnormal points that are farthest apart.

[0011] In a second aspect, the present invention also provides a hydraulic cylinder equipment process processing data monitoring system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of a hydraulic cylinder equipment process processing data monitoring method are implemented.

[0012] The embodiments of the present invention have at least the following beneficial effects: the present invention analyzes the pressure sequence of each hydraulic cylinder in a test to obtain the initial abnormal point, and can preliminarily detect the abnormal data points for subsequent analysis; compared with the prior art that uses a threshold to segment abnormal outliers and has a single usage scenario, the present solution constructs a monitoring space by integrating the monitoring data of hydraulic cylinders of the same caliber, and is suitable for abnormal monitoring of process processing data of hydraulic cylinders of any multiple calibers; further, a reference straight line is constructed, and the spatial distribution characteristics of the initial abnormal point are obtained according to the distance from the initial abnormal point to the reference straight line in a test and the data points before and after the initial abnormal point, and then the time distribution characteristics of the initial abnormal point are calculated according to the pressure sequences corresponding to the hydraulic cylinder corresponding to an initial abnormal point in other tests, and then the abnormal degree of the initial abnormal point is obtained by combining the spatial distribution characteristics and the time distribution characteristics, and a more accurate abnormal degree can be obtained by analyzing from multiple aspects, and then the defect probability of the hydraulic cylinder is obtained according to the number of abnormal points in the pressure sequence, so that the defects of the hydraulic cylinder can be detected, and the staff can be reminded to conduct a more detailed inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0014] Figure 1 A method flow chart of a method for monitoring equipment process processing data of a hydraulic cylinder provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the equipment process processing data monitoring method and system of a hydraulic cylinder proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0016] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0017] The following is a detailed description of a method and system for monitoring equipment process data of a hydraulic cylinder provided by the present invention in conjunction with the accompanying drawings.

[0018] Embodiment 1: The main application scenario of the present invention is: after the production of the hydraulic cylinder is completed, the process processing data of the hydraulic cylinder is monitored to ensure that the product is qualified. The present invention mainly targets the pressure monitoring data of the hydraulic cylinder, and takes the no-load pressure monitoring data as a benchmark to obtain the abnormal monitoring results of the hydraulic cylinder to evaluate whether the process performance of the hydraulic cylinder is qualified.

[0019] See also Figure 1 , which shows a method flow chart of a method for monitoring equipment process data of a hydraulic cylinder provided by an embodiment of the present invention, the method comprising the following steps: Step S1, testing each hydraulic cylinder a preset number of times to obtain a pressure sequence corresponding to each hydraulic cylinder in each test; and obtaining an initial abnormal point in the pressure sequence of each hydraulic cylinder in a test.

[0020] A no-load pressure test is performed on multiple factory equipment of the same caliber, and the number of tests is set. The number of tests is a preset number. Preferably, the preset number in the embodiment of the present invention is 5 times, and the period of each test is 10 minutes. Specifically, a pressure sensor is installed at the rod end of each hydraulic cylinder, the hydraulic system is turned on, and the hydraulic pump is started to ensure that the hydraulic cylinder is in a no-load working state. After the hydraulic cylinder system runs smoothly, the hydraulic cylinder piston rod is extended and retracted without load, and the pressure data is monitored and recorded. The frequency of monitoring and collecting pressure data is 10Hz, and the test period and sampling frequency can be adjusted by the implementer according to actual conditions. The pressure data corresponding to each hydraulic cylinder in each test will form a pressure sequence, that is, each hydraulic cylinder will correspond to a pressure sequence in each test. After the test is completed, the pressure sequence obtained by each hydraulic cylinder for each test is transmitted to the data processing terminal, and the terminal stores multiple pressure data corresponding to multiple hydraulic cylinders of the same caliber monitored this time.

[0021] Process the pressure data in the terminal. Project the pressure sequences of multiple hydraulic cylinders monitored in the same test into a two-dimensional rectangular coordinate system, and record the original data sequence to which each data point belongs. Under normal circumstances, during the no-load test, the piston rod of the hydraulic cylinder is extended and retracted without load, and the pressure data in the hydraulic cylinder fluctuates slowly and very slightly. Therefore, the pressure monitoring data of a normal hydraulic cylinder should be similar, and the data values ​​of the data points in the coordinate system are close to each other and have clustering. When the hydraulic cylinder has equipment process problems, such as uneven inner diameter flatness, it will cause pressure differences when the piston passes through uneven places. This type of problem is a low-probability event. Compared with normal data points, the pressure data points obtained will show relatively outlier abnormal characteristics. Therefore, by integrating the data of other hydraulic cylinders of the same caliber into the same coordinate system, relatively outlier points can be obtained as abnormal data points.

[0022] Here, the data of the pressure sequences of multiple hydraulic cylinders monitored in the same test are projected into a two-dimensional rectangular coordinate system. The data in the coordinate system is scattered, and the abnormal monitoring space of multiple pressure sequence data is constructed. The multiple pressure sequences monitored in the same measurement are taken to control the same environmental variables in the same measurement and monitoring process, so as to ensure that the obtained abnormal data is not interfered by environmental variables.

[0023] Further use LOF outlier monitoring to obtain the initial outlier factor: use Euclidean distance to calculate the distance between each data point in the pressure sequence corresponding to each hydraulic cylinder in the same test. For each data point, calculate the data points in its neighborhood. This neighborhood is usually defined by a distance metric, and methods such as the K nearest neighbor algorithm or the radius neighborhood can be used. Then calculate the local reachable density of each local point, and finally obtain the local outlier factor (LOF) of each data point. LOF is an indicator to measure the degree of abnormality of a data point relative to its neighborhood. It is the ratio of the local reachable density to the local reachable density of other points in the neighborhood. Take a commonly used empirical threshold of LOF as the threshold of the LOF value of all data points obtained to identify outliers. The data points with LOF values ​​greater than the threshold in the coordinate system are taken as initial outliers, and the initial outliers are recorded as initial outliers. Therefore, the local outlier factor algorithm (Local Outlier Factor) can be used to obtain the initial outliers in the pressure sequence of each hydraulic cylinder in a test.

[0024] Step S2, using the data points in the pressure sequence of each hydraulic cylinder in the same test except the initial abnormal point to perform straight line fitting to obtain a reference straight line; according to the distance from the initial abnormal point to the reference straight line in the test and the data points before and after the initial abnormal point, the spatial distribution characteristics of the initial abnormal point are obtained.

[0025] Because LOF outlier detection obtains outlier factors based on local density, the initial outliers (initial outliers) obtained may have a relatively high degree of local outliers, but compared with the overall low degree of outliers, these initial outliers cannot clearly reflect the abnormality of the hydraulic cylinder pressure data. At the same time, because the initial outliers are obtained only based on local density division, the outlier data points obtained by LOF outlier detection may contain sensor noise error data in addition to the data caused by hydraulic cylinder abnormalities. In view of the various factors of the above abnormal situations, it is necessary to further analyze the distribution characteristics of the data to obtain relatively accurate hydraulic cylinder abnormal data points.

[0026] Here, a spatial distribution characteristic model of the initial outlier point is constructed to obtain the spatial characteristic degree of the initial outlier point. Because the normal data fluctuation of the hydraulic cylinder is a slight fluctuation with linear distribution, the data points except the initial outlier point in the pressure sequence of each hydraulic cylinder in the same test are used for straight line fitting to obtain the reference straight line, and the possibility of the initial outlier point being a hydraulic cylinder process outlier point is judged according to the distance from the initial outlier point to the reference straight line. The farther from the reference straight line, the more likely the initial outlier point is a hydraulic cylinder process outlier point, eliminating the defect interference of LOF outlier points with high local outliers but low overall outlier degree.

[0027] On the other hand, other data points closest to the initial abnormal point in the pressure sequence where the initial abnormal point is located are obtained. The noise data points caused by poor sensor reception are sudden, and the distance between the initial abnormal point and the adjacent data points will suddenly change dramatically. Here, the Euclidean distance between the initial abnormal point and the adjacent data points is obtained, and the ratio of the subsequent adjacent data points of the initial abnormal point is obtained. The larger the ratio, the more drastic the change of the initial abnormal point at this point is, and the more likely it is a noise point. Otherwise, it is more likely to be a non-noise hydraulic cylinder abnormal outlier, thereby eliminating noise interference.

[0028] The spatial distribution characteristics of the initial outlier point are obtained according to the distance between the initial outlier point and the reference straight line in the test and the data points before and after the initial outlier point. Specifically, the ratio of the distance between the initial outlier point and the reference straight line to the maximum value of the distance between each initial outlier point and the reference straight line is obtained, which is recorded as the normalized distance; four data points adjacent to the initial outlier point in the sequence where the initial outlier point is located are obtained, the data point closest to the initial outlier point among the adjacent data points before the initial outlier point is the first data point, and the other data point is the second data point; the data point closest to the initial outlier point among the adjacent data points after the initial outlier point is the third data point, and the other data point is the fourth data point; the Euclidean distances between the first data point, the third data point and the initial outlier point are calculated respectively and summed to obtain the first distance sum; the Euclidean distance between the first data point and the second data point and the Euclidean distance between the third data point and the fourth data point are calculated respectively, and the sum is obtained to obtain the second distance sum; the exponential function with a natural constant as the base is used to negatively correlate the ratio of the first distance sum to the second distance sum to obtain the mapping result; the sum of the mapping result and the normalized distance is the spatial distribution characteristics of the initial outlier point.

[0029] The calculation formula for the spatial distribution characteristics of the initial outlier points is: , in, Represents the spatial distribution characteristics of the qth data point that is the initial abnormal point in the pressure sequence corresponding to a hydraulic cylinder in a test; is the distance between the initial abnormal point and the reference straight line. The farther the distance is, the more the data point representing the initial abnormal point deviates from the normal linear distribution of the hydraulic cylinder pressure fluctuation trend, and the more likely it is a hydraulic cylinder process abnormal point. Indicates the maximum value of the distance from all initial abnormal points to the reference straight line in this test. It is for normalization. , They represent the Euclidean distances between the q-1th data point, the q+1th data point and the qth data point which is the initial outlier, that is, the Euclidean distances between the first data point, the third data point and the initial outlier. is the first distance and; , They represent the Euclidean distance between the q-1th data point and the q-2th data point, and the Euclidean distance between the q+1th data point and the q+2th data point, respectively. is the second distance and, It represents the ratio of the distance between the nearest data points on both sides of the qth data point which is the initial outlier to the distance between the nearest data points one data point forward and backward. The smaller the ratio is, the lower the suddenness of the qth data point which is the initial outlier is, the less likely it is a noise point, and the more likely it is a hydraulic cylinder process outlier. It is the result of negative correlation mapping of the ratio of the first distance sum to the second distance sum.

[0030] Therefore, the spatial distribution characteristic degree can be obtained by combining the spatial distribution characteristics of the initial abnormal point in the scattered point space. The greater the characteristic degree, the more likely it is that the corresponding initial abnormal point is a data abnormal point caused by the process abnormality of the hydraulic cylinder equipment.

[0031] Step S3, calculating the time distribution characteristics of an initial abnormal point according to the pressure sequences corresponding to the hydraulic cylinder corresponding to the initial abnormal point in other tests; and obtaining the abnormal degree according to the spatial distribution characteristics and time distribution characteristics of the initial abnormal point.

[0032] When boring and honing the inner diameter of the cylinder, the inner wall flatness is uneven due to processing errors caused by clamping and extrusion. The unevenness will be located on the fixed inner wall of the hydraulic cylinder. Therefore, during multiple no-load tests, as the hydraulic cylinder piston rod moves in extension and retraction, the fluctuation position of the pressure anomaly should be located in the same place, that is, the same time series monitoring point, which is manifested in the two-dimensional plane coordinate system as the similarity of multiple historical data points corresponding to the same horizontal axis coordinate value. Here, the time series data (pressure sequence) of multiple tests of the same hydraulic cylinder is used to obtain the position of the current initial abnormal point in the pressure sequence, corresponding to the pressure data monitored at this position in other test times. If the pressure value difference at the same time is smaller and the data distribution is more similar, it means that the initial abnormal point is more likely to be a pressure data anomaly caused by a hydraulic cylinder equipment process problem, thereby further obtaining more accurate hydraulic cylinder abnormal data points.

[0033] The time distribution characteristics of an initial abnormal point are calculated according to the pressure sequences corresponding to the hydraulic cylinder corresponding to an initial abnormal point in other tests. Specifically, the pressure sequences corresponding to the hydraulic cylinder corresponding to an initial abnormal point in other tests are respectively recorded as pressure sequences to be compared; the data points in the same position as the initial abnormal point in each pressure sequence to be compared are obtained and recorded as data points to be compared; the sum of the absolute values ​​of the differences between the initial abnormal point and the corresponding data points to be compared is negatively correlated using an exponential function with a natural constant as the base to obtain the pressure value similarity; the sum of the absolute values ​​of the differences in spatial distribution characteristics between the initial abnormal point and the corresponding data points to be compared is negatively correlated using an exponential function with a natural constant as the base to obtain the spatial distribution characteristic similarity; the sum of the pressure value similarity and the spatial distribution characteristic similarity is the time distribution characteristic degree of the initial abnormal point.

[0034] It should be noted that, in the embodiment of the present invention, an initial abnormal point has a total of four pressure sequences to be compared, because a total of five tests were performed, including four tests in addition to the test corresponding to the initial abnormal point. At the same time, the position of the initial abnormal point refers to its position in the pressure sequence.

[0035] The specific calculation formula for the time distribution characteristic degree is: , in, It represents the time distribution characteristics of the data point with the qth initial abnormal point in the pressure sequence corresponding to a hydraulic cylinder in a test; exp() represents the exponential function with the natural constant e as the base; Indicates the number of data points to be compared corresponding to the qth data point which is the initial outlier, which is 4; represents the pressure data value of the qth data point which is the initial abnormal point; represents the pressure data value of the ith data point to be compared among the data points to be compared corresponding to the qth data point which is the initial abnormal point, The pressure difference between two tests of a hydraulic cylinder at the same time. It is the cumulative sum of pressure differences. The smaller the cumulative sum is, the more similar the qth data point, which is the initial abnormal point, is to the pressure values ​​at the same time in other tests, and the more likely it is to be an abnormal outlier of the hydraulic cylinder equipment process abnormality.

[0036] It represents the spatial distribution characteristics of the data point with the qth initial abnormal point in the pressure sequence corresponding to a hydraulic cylinder in a test. It represents the spatial distribution characteristics of the ith data point to be compared among the data points to be compared corresponding to the qth data point which is the initial outlier. It represents the cumulative sum of the differences in the spatial distribution characteristics of the data corresponding to two tests of a hydraulic cylinder at the same time. The smaller the cumulative sum, the more similar the data distribution of the qth data point as the initial abnormal point is to the data distribution at the same time in other tests, and the points of other tests at the same time are more likely to be abnormal points, and the initial abnormal point is more likely to be an abnormal outlier point of the hydraulic cylinder equipment process abnormality. In this way, more accurate abnormal outlier data points of hydraulic cylinders are obtained by combining the data distribution in the time dimension.

[0037] Furthermore, the spatial distribution characteristics and time distribution characteristics of the initial abnormal point are combined to obtain the abnormal degree of the initial abnormal point. Specifically, the spatial distribution characteristics and time distribution characteristics of the initial abnormal point are added and compared with the preset value to obtain the abnormal degree of the initial abnormal point, where the preset value is 4. The specific calculation formula is: , in, It indicates the abnormal degree of the qth data point which is the initial abnormal point in the pressure sequence corresponding to a hydraulic cylinder in a test; and They respectively represent the spatial distribution characteristics and temporal distribution characteristics of the qth initial abnormal point in the pressure sequence corresponding to a hydraulic cylinder in a test; The control value range is in the range [0,1].

[0038] In this way, the abnormality degree of all initial abnormal points can be obtained.

[0039] Step S4, judging whether the initial abnormal point is an abnormal point based on the abnormal degree of the initial abnormal point; obtaining the defect probability of the hydraulic cylinder corresponding to the pressure sequence based on the number of abnormal points in a pressure sequence and the interval distance between the two most distant abnormal points.

[0040] In step S3, the abnormality degree of all initial abnormal points is obtained. Further, a first threshold is set. Preferably, the value of the first threshold in the embodiment of the present invention is 0.6. The implementer can adjust the first threshold according to actual conditions. If the abnormality degree of an initial abnormal point is greater than or equal to the first threshold, the initial abnormal point is an abnormal point, and the initial abnormal point is an abnormal data point caused by the process abnormality of the hydraulic cylinder equipment.

[0041] Then, all abnormal points are restored to the original pressure sequence, and the existence of abnormal points in the pressure sequence is used to determine whether the sequence data is abnormal. The more abnormal points there are in the sequence and the larger the distribution range of the abnormal points, the more serious the abnormality of the hydraulic cylinder's identification process is, and the stronger the abnormality of its pressure sequence is.

[0042] Then, based on the number of abnormal points in a pressure sequence and the interval distance between the two abnormal points that are farthest apart, the defect probability of the hydraulic cylinder corresponding to the pressure sequence is obtained. Specifically, the ratio of the number of abnormal points in a pressure sequence to the number of data points in the pressure sequence is obtained, which is recorded as the number ratio; the ratio of the interval distance between the two abnormal points that are farthest apart in the pressure sequence to the length of the pressure sequence is obtained, which is recorded as the range ratio; the product of the number ratio and the range ratio is the defect probability of the hydraulic cylinder corresponding to the pressure sequence; wherein the interval distance between the two abnormal points that are farthest apart in the pressure sequence is the number of data points between the two abnormal points that are farthest apart.

[0043] The specific calculation formula is: , in, It represents the defect probability of the equipment process processing data obtained in a test of the p-th hydraulic cylinder; The number of pressure sequence data points corresponding to the pth hydraulic cylinder in this test; It represents the number of abnormal points in the pressure sequence corresponding to p hydraulic cylinders in this test; Indicates the length of the pressure sequence corresponding to the pth hydraulic cylinder in this test; It indicates the distance between the two most distant abnormal points in the pressure sequence corresponding to the p-th hydraulic cylinder in this test, that is, the number of data points between the two most distant abnormal points. The larger the proportion of abnormal data points, the greater the probability of defect in the corresponding process data. It is the number of intervals between the two farthest abnormal data points in the sequence, that is, the distribution range of the abnormal points. The more data points there are, the longer the interval distance is, the larger the corresponding abnormal distribution range is, and the greater the probability of defects in the process data is.

[0044] Finally, the second threshold is set. Preferably, the value of the second threshold in the embodiment of the present invention is 0.4, which can be adjusted according to the scene requirements. When the probability of a hydraulic cylinder defect in a test is greater than or equal to the second threshold, it is considered that the hydraulic cylinder has equipment process processing problems and needs to be checked in time for subsequent processing. Through the above monitoring and analysis method, it is possible to handle the abnormal monitoring task of equipment process processing data of hydraulic cylinders of various calibers, and obtain relatively accurate abnormal monitoring results.

[0045] Embodiment 2: This embodiment provides a hydraulic cylinder equipment process processing data monitoring system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of a hydraulic cylinder equipment process processing data monitoring method when executed by the processor. Since the embodiment 1 has already described a hydraulic cylinder equipment process processing data monitoring method in detail, it will not be described in detail here.

[0046] To summarize, the present invention simultaneously monitors and collects pressure data of multiple groups of hydraulic cylinders, constructs an abnormal monitoring space, and analyzes the abnormal distribution characteristics of data points, so as to obtain a relatively accurate degree of abnormality of the hydraulic cylinder monitoring data and make an accurate and quick evaluation of the production process processing data of the hydraulic cylinder.

[0047] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0049] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for monitoring equipment process data of a hydraulic cylinder, characterized in that: The method includes: Perform a preset number of tests on each hydraulic cylinder to obtain a pressure sequence corresponding to each hydraulic cylinder in each test; obtain an initial abnormal point in the pressure sequence of each hydraulic cylinder in a test; The reference straight line is obtained by linear fitting using the data points except the initial abnormal point in the pressure sequence of each hydraulic cylinder in the same test; the spatial distribution characteristics of the initial abnormal point are obtained according to the distance from the initial abnormal point to the reference straight line in the test and the data points before and after the initial abnormal point; Calculate the time distribution characteristics of an initial abnormal point according to the pressure sequences corresponding to the hydraulic cylinder corresponding to the initial abnormal point in other tests; obtain the abnormal degree according to the spatial distribution characteristics and time distribution characteristics of the initial abnormal point; Whether the initial abnormal point is an abnormal point is determined based on the abnormal degree of the initial abnormal point; and the defect probability of the hydraulic cylinder corresponding to the pressure sequence is obtained based on the number of abnormal points in a pressure sequence and the interval distance between the two most distant abnormal points.

2. The method for monitoring equipment process data of a hydraulic cylinder according to claim 1, characterized in that: The step of obtaining the initial abnormal point in the pressure sequence of each hydraulic cylinder in a test comprises: The local anomaly factor algorithm is used to obtain the initial abnormal points in the pressure sequence of each hydraulic cylinder in a test.

3. The method for monitoring equipment process data of a hydraulic cylinder according to claim 1, characterized in that: The spatial distribution characteristics of the initial outlier point are obtained according to the distance from the initial outlier point to the reference straight line and the data points before and after the initial outlier point in the test, including: Obtain the ratio of the distance from the initial outlier point to the reference straight line to the maximum value of the distances from each initial outlier point to the reference straight line, recorded as the normalized distance; obtain four data points adjacent to the initial outlier point in the sequence where the initial outlier point is located, wherein the data point closest to the initial outlier point among the adjacent data points before the initial outlier point is the first data point, and the other data point is the second data point, and the data point closest to the initial outlier point among the adjacent data points after the initial outlier point is the third data point, and the other data point is the fourth data point; calculate the Euclidean distances between the first data point, the third data point and the initial outlier point respectively and sum them to obtain the first distance sum; calculate the Euclidean distance between the first data point and the second data point and the Euclidean distance between the third data point and the fourth data point respectively, and sum them to obtain the second distance sum; use an exponential function with a natural constant as the base to negatively correlate the ratio of the first distance sum to the second distance sum to obtain a mapping result; the sum of the mapping result and the normalized distance is the spatial distribution feature of the initial outlier point.

4. The method for monitoring equipment process data of a hydraulic cylinder according to claim 1, characterized in that: The calculating of the time distribution characteristics of an initial abnormal point according to each pressure sequence corresponding to the hydraulic cylinder corresponding to the initial abnormal point in other tests includes: The pressure sequences corresponding to the hydraulic cylinder corresponding to an initial abnormal point in other tests are recorded as pressure sequences to be compared; the data points in the same position as the initial abnormal point in each pressure sequence to be compared are obtained and recorded as data points to be compared; the sum of the absolute values ​​of the differences between the initial abnormal point and the corresponding data points to be compared is negatively correlated using an exponential function with a natural constant as the base to obtain the pressure value similarity; the sum of the absolute values ​​of the differences in the spatial distribution characteristics between the initial abnormal point and the corresponding data points to be compared is negatively correlated using an exponential function with a natural constant as the base to obtain the spatial distribution characteristic similarity; the sum of the pressure value similarity and the spatial distribution characteristic similarity is the temporal distribution characteristic degree of the initial abnormal point.

5. The method for monitoring equipment process data of a hydraulic cylinder according to claim 1, characterized in that: The step of obtaining the abnormality degree according to the spatial distribution characteristics and the temporal distribution characteristics of the initial abnormal point includes: The spatial distribution characteristics and temporal distribution characteristics of the initial abnormal point are added and compared with the preset value to obtain the abnormal degree of the initial abnormal point.

6. The method for monitoring equipment process data of a hydraulic cylinder according to claim 1, characterized in that: The determining whether the initial abnormal point is an abnormal point based on the abnormality degree of the initial abnormal point includes: A first threshold is set. If the abnormality degree of an initial abnormal point is greater than or equal to the first threshold, the initial abnormal point is an abnormal point.

7. The method for monitoring equipment process data of a hydraulic cylinder according to claim 1, characterized in that: The method of obtaining the defect probability of the hydraulic cylinder corresponding to a pressure sequence based on the number of abnormal points in the pressure sequence and the interval distance between two abnormal points that are farthest apart comprises: The ratio of the number of abnormal points in a pressure sequence to the number of data points in the pressure sequence is obtained, which is recorded as the number ratio. The ratio of the interval distance between the two abnormal points that are farthest apart in the pressure sequence to the length of the pressure sequence is obtained, which is recorded as the range ratio. The product of the number ratio and the range ratio is the defect probability of the hydraulic cylinder corresponding to the pressure sequence. Among them, the interval distance between the two abnormal points that are farthest apart in the pressure sequence is the number of data points between the two abnormal points that are farthest apart.

8. A hydraulic cylinder equipment process data monitoring system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of a method for monitoring equipment process data of a hydraulic cylinder as described in any one of claims 1 to 7 are implemented.

Images