CN117798744A - Method for monitoring running state of numerical control machine tool - Google Patents
Method for monitoring running state of numerical control machine tool Download PDFInfo
- Publication number
- CN117798744A CN117798744A CN202410224010.7A CN202410224010A CN117798744A CN 117798744 A CN117798744 A CN 117798744A CN 202410224010 A CN202410224010 A CN 202410224010A CN 117798744 A CN117798744 A CN 117798744A
- Authority
- CN
- China
- Prior art keywords
- sampling
- sampling moment
- temperature
- range
- moment
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000012544 monitoring process Methods 0.000 title claims abstract description 40
- 238000005070 sampling Methods 0.000 claims abstract description 397
- 230000008859 change Effects 0.000 claims abstract description 97
- 238000001914 filtration Methods 0.000 claims abstract description 33
- 238000012937 correction Methods 0.000 claims abstract description 28
- 230000005856 abnormality Effects 0.000 claims abstract description 15
- 230000000875 corresponding effect Effects 0.000 claims description 103
- 230000002159 abnormal effect Effects 0.000 claims description 42
- 238000012545 processing Methods 0.000 claims description 14
- 238000010606 normalization Methods 0.000 claims description 11
- 230000002596 correlated effect Effects 0.000 claims description 10
- 230000002093 peripheral effect Effects 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 description 12
- 238000004458 analytical method Methods 0.000 description 11
- 238000000605 extraction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 230000003321 amplification Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000002277 temperature effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/12—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring vibration
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to the technical field of numerical control machine tool data monitoring, in particular to a numerical control machine tool running state monitoring method. According to the method, the temperature influence degree is obtained according to the deviation condition of temperature data change in a preset neighborhood range at each sampling time, the similarity condition of the vibration data and the temperature data in the change trend and the difference condition of the vibration data and the temperature data in the change degree; according to the distribution difference of vibration data of sampling time in a preset neighborhood range and a preset surrounding range, adjusting the deviation condition of the vibration data of the sampling time to obtain the noise possibility; and obtaining a Kalman gain correction coefficient of the sampling moment according to the temperature influence degree and the noise possibility degree of the sampling moment and the front part sampling moment, and obtaining filtering vibration data through Kalman filtering for monitoring. According to the invention, the influence and the noise abnormality similarity during vibration data acquisition are synthesized, the Kalman filtering algorithm is optimized, the denoising accuracy is improved, and the follow-up accuracy of vibration data monitoring in the running state is higher.
Description
Technical Field
The invention relates to the technical field of numerical control machine tool data monitoring, in particular to a numerical control machine tool running state monitoring method.
Background
The numerical control machine tool is an automatic machine tool provided with a program control system, and can better solve the problems of complex, precise, small batch and multiple kinds of part processing. In order to ensure the processing quality and the operation efficiency of the operation of the numerical control airport, data monitoring is required to be carried out on the operation state, such as vibration data with wider influence, potential machine tool faults can be found through the monitoring of the vibration data, and the abnormal processing or abnormal change in the operation in the processing process can be found, so that the normal operation of the machine tool is ensured, the processing quality is improved, and therefore, the monitoring on the operation state of the numerical control machine tool is extremely important.
When vibration data is monitored on the running state of the numerical control machine tool, abnormal noise data can be monitored due to interference of other environmental factors such as impact collision or electromagnetic interference, so that the abnormal vibration data cannot be accurately monitored, when the data is denoised through a Kalman filtering algorithm, the method combines historical information and current observation, and can process data acquired by a sensor in real time, but due to interference of the environmental temperature, the performance of some electronic elements of the vibration sensor is limited, the signal amplification and processing of the sensor are affected, the sensitivity of the sensor is reduced, abnormal noise data cannot be detected, the noise reduction result of Kalman filtering on the vibration data is inaccurate, and the follow-up accurate monitoring on the vibration data is affected.
Disclosure of Invention
In order to solve the technical problems that the noise reduction result of Kalman filtering on vibration data is inaccurate and the follow-up accurate monitoring on the vibration data is affected in the prior art, the invention aims to provide a method for monitoring the running state of a numerical control machine tool, and the adopted technical scheme is as follows:
the invention provides a method for monitoring the running state of a numerical control machine tool, which comprises the following steps:
vibration data and temperature data of each sampling time are obtained in a preset sampling time period;
obtaining a range change difference value of each sampling moment according to the change deviation condition of the temperature data corresponding to each sampling moment in the preset neighborhood range and the difference condition of the vibration data and the temperature data corresponding to each sampling moment in the preset neighborhood range in the change degree; obtaining the temperature influence degree of each sampling moment according to the similarity degree of the variation trend of the vibration data and the temperature data in the preset neighborhood range of each sampling moment, the range variation difference value of the corresponding sampling moment and the variation degree difference condition of the vibration data and the temperature data at the corresponding sampling moment;
obtaining the noise probability of each sampling moment according to the deviation degree of the corresponding vibration data of each sampling moment in a preset neighborhood range and the distribution deviation condition of the vibration data in the preset neighborhood range, and according to the distribution difference condition of all vibration data in the preset neighborhood range of the corresponding sampling moment and other vibration data in the preset surrounding range of the corresponding sampling moment; the preset surrounding range is larger than the preset neighborhood range;
Obtaining a Kalman gain correction coefficient of each sampling moment according to the temperature influence degree and the noise possibility degree of each sampling moment and a preset number of sampling moments before the corresponding sampling moment; and monitoring the vibration data at all sampling moments based on the Kalman gain correction coefficient by obtaining filtering vibration data through a Kalman filtering algorithm.
Further, the method for obtaining the range variation difference value includes:
taking each sampling time as a reference time in sequence, and for any sampling time in a preset neighborhood range of the reference time, obtaining the temperature credibility weight of the sampling time according to the value of temperature data of the sampling time and the deviation condition of the variation degree in the preset neighborhood range;
calculating the difference between the slope of the temperature data corresponding to the sampling moment and the slope of the vibration data corresponding to the sampling moment to obtain a variation difference index of the sampling moment; taking the product of the variation difference index of the sampling moment and the temperature credible weight as an adjustment difference index of the sampling moment;
and calculating the average value of the adjustment difference indexes of all sampling moments in the preset neighborhood range of the reference moment to obtain the range change difference value of the reference moment.
Further, the method for acquiring the temperature credible weight comprises the following steps:
calculating the average value of the values of the temperature data corresponding to all the sampling moments in a preset neighborhood range of the reference moment to obtain the average value of the temperature data of the reference moment; calculating the difference between the value of the temperature data corresponding to the sampling time and the temperature average value to obtain a temperature data difference value of the sampling time;
calculating the average value of the slopes of the temperature data corresponding to all sampling moments, and obtaining the average value of the temperature slopes at the reference moment; calculating the difference between the slope of the temperature data corresponding to the sampling time and the average value of the temperature slope to obtain a temperature slope difference value of the sampling time;
carrying out negative correlation mapping and normalization processing on the product of the temperature data difference value and the temperature slope difference value at the sampling moment to obtain a temperature similarity index at the sampling moment;
calculating accumulated values of temperature similarity indexes of all sampling moments in a preset neighborhood range of the reference moment to obtain a total temperature similarity index of the reference moment; and taking the ratio of the temperature similarity index at the sampling moment to the total temperature similarity index as the temperature credible weight at the sampling moment.
Further, the method for obtaining the temperature influence degree comprises the following steps:
For any sampling moment, calculating the difference between the change difference index of the sampling moment and the range change difference value of the sampling moment, performing negative correlation mapping and normalizing to obtain the change correlation degree of the sampling moment;
counting the number of the same sign between the slope of vibration data and the slope of temperature data at each sampling moment in a preset neighborhood range of the sampling moment to obtain the trend correlation degree of the sampling moment;
obtaining the temperature influence degree of the sampling moment according to the change correlation degree and the trend correlation degree of the sampling moment; both the change correlation and the trend correlation are positively correlated with the temperature influence.
Further, the method for obtaining the noise probability comprises the following steps:
for any sampling moment, calculating the average value of the values of vibration data corresponding to all sampling moments in a preset neighborhood range of the sampling moment, and obtaining the vibration data average value of the sampling moment; taking the ratio of the value of the vibration data corresponding to the sampling moment to the average value of the vibration data as the value abnormality index of the sampling moment;
obtaining extreme points in a preset neighborhood range of the sampling moment; calculating the difference between the value of each extreme point and the mean value of the vibration data to obtain the extreme deviation difference of each extreme point; taking the average value of extreme value deviation differences of all extreme value points in a preset neighborhood range as the range distribution degree of the sampling moment;
Calculating the average value of maximum values in all vibration data corresponding to the preset surrounding range of the sampling moment except the preset neighborhood range, and obtaining the high surrounding average value of the sampling moment; calculating the average value of minimum values in all vibration data corresponding to the preset surrounding range of the sampling moment except the preset neighborhood range, and obtaining the low surrounding average value of the sampling moment; counting the number of the vibration data corresponding to the preset neighborhood range of the sampling moment, wherein the number is smaller than the high peripheral average value and larger than the low peripheral average value, so as to obtain the distribution concentration degree of the sampling moment;
obtaining a range concentration index of the sampling moment according to the distribution concentration degree and the range distribution degree of the sampling moment; the distribution concentration degree is positively correlated with the range concentration index, and the range distribution degree is negatively correlated with the range concentration index; the range concentration index is a normalized value;
and taking the product of the numerical value abnormal index and the range concentration index at the sampling time as the noise possibility of the sampling time.
Further, the obtaining the extreme point in the preset neighborhood range of the sampling time includes:
performing curve fitting on vibration data corresponding to a preset neighborhood range of the sampling moment to obtain a vibration curve; and taking the point when the first order derivative on the vibration curve is zero as an extreme point.
Further, the method for acquiring the kalman gain correction coefficient comprises the following steps:
taking the product of the temperature influence degree and the noise possibility degree of each sampling moment as the influence degree of each sampling moment;
for any sampling moment, taking the influence degree of the inverse proportion of the sampling moment as an adjustment index of the sampling moment; taking the accumulated value of the influence of a preset number of sampling moments before the sampling moment as the adjustment credibility of the sampling moment;
and calculating the product of the adjustment credible weight and the adjustment index of the sampling moment, and adjusting the value range to obtain the Kalman gain correction coefficient of the sampling moment.
Further, the monitoring of the filtered vibration data of the vibration data at all sampling moments based on the kalman gain correction coefficient by a kalman filtering algorithm includes:
when the vibration data at all sampling moments are filtered through a Kalman filtering algorithm, taking the product of a Kalman gain correction coefficient and a corresponding Kalman gain at each sampling moment as the optimized Kalman gain to filter, and obtaining the filtered vibration data at each sampling moment;
when the filtering vibration data is larger than a preset abnormal threshold value, the corresponding sampling time is marked as an abnormal time; when the continuous number of abnormal moments is greater than or equal to the preset abnormal number, the running state is recorded as an abnormal state and early warning is carried out; otherwise, the running state is recorded as a normal state.
Further, the preset neighborhood range is set to be a neighborhood range with the sampling time as the center and the side length of 50.
Further, the preset surrounding range is set to be 4 times as large as the preset neighborhood range.
The invention has the following beneficial effects:
according to the method, through analysis on the influence of temperature and the possibility of noise of vibration data, firstly, according to the deviation condition of temperature data change and the difference condition of vibration data and temperature data in the change degree in a preset neighborhood range of each sampling time, a range change difference value of each sampling time is obtained, and the influence of temperature data with abnormal conditions is eliminated, so that consistency data of the change degree difference condition in the range of each sampling time is obtained, and whether the difference on the corresponding change of each sampling time accords with consistency is conveniently analyzed. Further, according to the consistency of vibration data and temperature data in the preset neighborhood range of each sampling time in the change trend, the range change difference value and the change degree difference condition, the temperature influence degree of the sampling time is obtained, and the influence of the temperature data on the vibration data is obtained through the relevance according to the consistency of the vibration data and the temperature data in the change trend and the change degree. And secondly, analyzing according to data distribution deviation of the vibration data possibly of the abnormal situation at each sampling moment, and adjusting through distribution differences of the vibration data in a preset neighborhood range and a preset surrounding range corresponding to the sampling moment, so that noise possibility which is more likely to be noise is obtained from the abnormal situation, and denoising of the vibration data is more accurate. Finally, the temperature influence degree and the noise possibility degree of the sampling time and the historical sampling time are combined to obtain the self-adaptive Kalman gain correction coefficient of each sampling time, so that the accuracy of a Kalman filtering algorithm can be improved, and the filtering vibration data can be obtained for monitoring. According to the method, according to the influence condition of the temperature data during vibration data acquisition and the condition that noise in the vibration data is similar to abnormality, the Kalman gain in the Kalman filtering algorithm is adaptively adjusted, the denoising accuracy of the Kalman filtering algorithm is improved, and the follow-up accuracy of monitoring the vibration data of the running state of the numerical control machine tool is higher.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for monitoring an operation state of a numerically-controlled machine tool according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of a specific implementation, structure, characteristics and effects of a method for monitoring the operation state of a numerical control machine tool according to the invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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.
The following specifically describes a specific scheme of the method for monitoring the running state of the numerical control machine tool provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for monitoring an operation state of a numerically-controlled machine tool according to an embodiment of the present invention is shown, and the method includes the following steps:
s1: vibration data and temperature data of each sampling time are obtained in a preset sampling time period.
The performance of some of the electronic components of the vibration sensor is limited due to the disturbance of the ambient temperature, thereby affecting the signal amplification and processing of the sensor, resulting in reduced sensitivity. If the sensitivity of the sensor is reduced, the amplitude of the output signal generated by the sensor will be smaller for the same input vibration signal, so that the vibration data will be lower, and thus some abnormal noise data cannot be detected, so that the noise reduction result of the vibration data is not accurate enough. The Kalman gain in the Kalman filtering is corrected according to the influence between each vibration data and the temperature data and the abnormality possibility degree of each vibration data.
Firstly, vibration data and temperature data of each sampling time are required to be acquired in a preset time period, so that the change relation between the two data and the distribution change condition of the data can be analyzed conveniently. In the embodiment of the invention, a vibration sensor and a high-precision thermometer are adopted for sampling and recording in the running process of the numerical control machine tool, the preset time period is 30 minutes, the sampling frequency is once of 0.5 second, a specific numerical value implementation person can adjust according to specific implementation conditions, and each sampling moment corresponds to one vibration data and one temperature data.
The data acquisition process is completed, the relation and distribution situation among the data are further analyzed, and the denoising degree is adjusted in a self-adaptive mode.
S2: obtaining a range change difference value of each sampling moment according to the change deviation condition of the temperature data corresponding to each sampling moment in the preset neighborhood range and the difference condition of the vibration data and the temperature data corresponding to each sampling moment in the preset neighborhood range in the change degree; and obtaining the temperature influence degree of each sampling moment according to the similarity degree of the variation trend of the vibration data and the temperature data in the preset neighborhood range of each sampling moment, the range variation difference value of the corresponding sampling moment and the variation degree difference condition of the vibration data and the temperature data at the corresponding sampling moment.
Considering the influence condition of the vibration data on the temperature relationship, when the change of the temperature data and the change of the vibration data are more close, the influence of the temperature data is larger, the corresponding vibration data are more likely to be influenced by the environment temperature, so that the influence degree of the temperature data at each sampling time is analyzed through the correlation between the change condition of the vibration data in the range and the change condition of the temperature data.
When analysis is carried out according to the similar conditions among the changes, the conditions of the sensor subjected to the environmental influence are staged, namely when the change conditions are relatively consistent in a certain period of time, the influence of the temperature on the vibration sensor can be reflected, so that the detection of vibration data is inaccurate, the change degree in a period of time is taken as a reference, the consistency degree of each acquisition time is analyzed, and the specific influence degree is obtained. When the change degree over a period of time is analyzed integrally, whether the temperature data has abnormal data or not is considered, when the collected temperature data has abnormality, the reliability of the change difference condition of the corresponding analysis is low, so that the range change difference value of each sampling time is obtained in the preset neighborhood range of each sampling time according to the change difference condition of the temperature data corresponding to each sampling time in the preset neighborhood range and the change condition of the vibration data and the temperature data corresponding to each sampling time in the preset neighborhood range, and the range change difference value represents the integral trend condition of the change difference in the corresponding range of the sampling time.
Preferably, each sampling time is sequentially taken as a reference time, for any one sampling time in a preset neighborhood range of the reference time, the change condition of each sampling time is analyzed in the preset neighborhood range corresponding to the reference time, in the embodiment of the invention, the preset neighborhood range is set to be a neighborhood range with the sampling time as the center and the side length of 50, and a specific numerical value implementer can adjust according to specific implementation conditions.
Further, according to the deviation condition of the numerical value and the variation degree of the temperature data at the sampling time in a preset neighborhood range, obtaining the temperature credibility weight at the sampling time, and considering the unreliable condition of calculating the variation correlation degree under the abnormal condition of the temperature data, adjusting the analysis temperature credibility weight at each sampling time in the preset neighborhood range.
In one embodiment of the invention, in a preset neighborhood range of a reference time, calculating an average value of values of temperature data corresponding to all sampling times, obtaining a temperature data average value of the reference time, calculating a difference between the values of the temperature data corresponding to the sampling times and the temperature average value, obtaining a temperature data difference value of the sampling times, and reflecting abnormal conditions of temperature by the temperature data difference value through the deviation of the values. Calculating the average value of the slopes of the temperature data corresponding to all the sampling moments, obtaining the average value of the temperature slopes of the reference moments, reflecting the change degree of the temperature data at the corresponding sampling moments through the slopes, calculating the difference between the slopes of the temperature data corresponding to the sampling moments and the average value of the temperature slopes, and obtaining the temperature slope difference value at the sampling moments, wherein the temperature slope difference reflects the abnormal condition of the temperature through the deviation of the change degree. And carrying out negative correlation mapping and normalization processing on the product of the temperature data difference value and the temperature slope difference value at the sampling moment to obtain a temperature similarity index at the sampling moment, wherein the temperature similarity index reflects the credibility of the temperature data through the negative correlation of the comprehensive deviation on the numerical value and the change. It should be noted that, the method for calculating the corresponding slope of the data point is a technical means known to those skilled in the art, for example, a first order difference method is adopted, and will not be described herein.
Further, the accumulated value of the temperature similarity indexes of all sampling moments in a preset neighborhood range of the reference moment is calculated, the total temperature similarity index of the reference moment is obtained, the ratio of the temperature similarity index of the sampling moment to the total temperature similarity index is used as the temperature credibility weight of the sampling moment, the similarity degree of each sampling moment is standardized in a duty ratio mode, and in the embodiment of the invention, the expression of the temperature credibility weight is as follows:
in the method, in the process of the invention,denoted as +.>The first part of the preset neighborhood range of the sampling time>Temperature confidence weights for the individual sampling instants, +.>Denoted as +.>Slope of temperature data corresponding to each sampling instant, +.>Denoted as +.>Temperature gradient mean value of each sampling time, +.>Denoted as +.>The value of the temperature data corresponding to the sampling instants, +.>Denoted as +.>Temperature data mean value at each sampling instant, +.>Denoted as +.>Total number of all sampling instants in a preset neighborhood of the individual sampling instants, < >>Expressed as absolute value extraction function,/->Represented as an exponential function with a base of natural constant.
Wherein,denoted as +.>The first part of the preset neighborhood range of the sampling time>Temperature gradient difference value of each sampling moment, +. >Denoted as +.>The first part of the preset neighborhood range of the sampling time>Temperature data difference values at each sampling instant, +.>Denoted as +.>The first part of the preset neighborhood range of the sampling time>Temperature similarity index at each sampling instant, +.>Denoted as the firstAnd the temperature at each sampling moment is a total similarity index. When the sampling time in the neighborhood range corresponds to the smaller temperature deviation degree, the higher the temperature similarity is, the larger the temperature similarity index is in all the similar conditions, the higher the temperature at the sampling time in the neighborhood range corresponds to the temperature change under the normal condition, the greater the credibility degree of the sampling time when the vibration change and the temperature change are analyzed is, and therefore the larger the credibility weight of the temperature is.
After the analysis of the temperature credible weight is completed, the integral correlation degree between the vibration data change and the temperature data change in the neighborhood range corresponding to the reference time is further analyzed, the difference between the slope of the temperature data corresponding to the sampling time and the slope of the vibration data corresponding to the sampling time is calculated, the change difference index of the sampling time is obtained, and the change difference index reflects the change difference degree at the time through the slope difference of each sampling time. Taking the product of the variation difference index of the sampling time and the temperature credible weight as an adjustment difference index of the sampling time, and obtaining more accurate difference degree of each sampling time through temperature adjustment by the adjustment difference index.
And finally, obtaining the range variation difference value of the reference time by the average value of variation difference indexes of all sampling times in a preset neighborhood range of the reference time, wherein the range variation difference value reflects the overall average trend condition of variation difference in time by the average value of more accurate difference degrees corresponding to each sampling time in the neighborhood range corresponding to the reference time. In the embodiment of the present invention, the expression of the range change difference value is:
in the method, in the process of the invention,denoted as +.>The range change difference value of the individual sampling instants, +.>Denoted as +.>Total number of all sampling instants in a preset neighborhood of the individual sampling instants, < >>Denoted as +.>Slope of vibration data corresponding to the sampling time points, < >>Denoted as +.>Slope of temperature data corresponding to each sampling instant, +.>Denoted as +.>The first part of the preset neighborhood range of the sampling time>Temperature confidence weights for the individual sampling instants, +.>Represented as an absolute value extraction function.
Wherein,denoted as +.>Variation difference index of each sampling instant, +.>Denoted as +.>The adjustment difference indexes of the sampling moments are reflected on the local time periods corresponding to the sampling moments through range change difference values, and when the vibration data and the temperature data have the change correlation, the difference degrees are more consistent, so that the correlation at the sampling moments can be further judged according to the difference conditions of average trends in the sampling moments and the time ranges.
Meanwhile, when the vibration data and the temperature data are more consistent and similar in change trend, the stronger the correlation degree is, the larger the vibration data are influenced, so that the influence condition of the temperature at each sampling moment is analyzed by combining the change trend and the deviation degree of the average difference condition in a range, namely, the temperature influence degree at each sampling moment is obtained according to the change trend similarity degree of the vibration data and the temperature data in the preset neighborhood range of each sampling moment, the range change difference value of the corresponding sampling moment and the difference condition of the vibration data and the temperature data in the change degree at the corresponding sampling moment.
Preferably, for any one sampling time, the difference between the change difference index of the sampling time and the range change difference value of the sampling time is calculated, negative correlation mapping is performed, normalization processing is performed to obtain the change correlation degree of the sampling time, and for any one sampling time, when the change difference index is larger relative to the overall trend deviation degree in the range time period, the fluctuation of the change difference is stronger, and the vibration data and the temperature data at the sampling time do not have stronger correlation degree.
And counting the number of the same sign between the slope of vibration data and the slope of temperature data at each sampling time, namely the number of times of the same sign, in a preset neighborhood range of the sampling time, so as to obtain the trend correlation degree of the sampling time, and reflecting the degree of closeness between the changes through the consistent degree of the same trend change.
And obtaining the temperature influence degree of the sampling time according to the change correlation degree and the trend correlation degree of the sampling time, wherein the temperature influence degree comprehensively reflects the correlation degree on the change difference and the correlation degree on the change trend. The change correlation and the trend correlation are both positively correlated with the temperature influence, and in the embodiment of the invention, the expression of the temperature influence is:
in the method, in the process of the invention,denoted as +.>Temperature influence at each sampling instant, +.>Denoted as +.>The range change difference value of the individual sampling instants, +.>Denoted as +.>Slope of vibration data corresponding to the sampling time points, < >>Denoted as +.>Slope of temperature data corresponding to each sampling instant, +.>Denoted as +.>Trend correlation at each sampling instant, +.>Expressed as absolute value extraction function,/->It should be noted that, normalization is a technical means well known to those skilled in the art, and the normalization function may be selected by linear normalization or standard normalization, and the specific normalization method is not limited herein.
Wherein,denoted as +.>The variation of the difference index at each sampling instant,denoted as +.>The change correlation degree of each sampling time point shows that the higher the change correlation degree is, the higher the trend correlation degree is, and the higher the change correlation between the temperature data and the vibration data isThe greater the effect that the temperature data will have on the vibration data at this time, the greater the temperature effect. In other embodiments of the present invention, other basic mathematical operations may be applied to reflect that the change correlation degree and the trend correlation degree are both positively correlated with the temperature influence degree, such as addition or power operation, without limitation.
So far, by analyzing the change correlation of the temperature and the vibration data, the inaccuracy of the temperature influence degree reflection vibration data caused by the temperature influence of the acquisition environment is obtained.
S3: obtaining the noise probability of each sampling moment according to the deviation degree of the corresponding vibration data of each sampling moment in a preset neighborhood range and the distribution deviation condition of the vibration data in the preset neighborhood range, and according to the distribution difference condition of all vibration data in the preset neighborhood range of the corresponding sampling moment and other vibration data in the preset surrounding range of the corresponding sampling moment; the preset surrounding range is larger than the preset neighborhood range.
In the process of changing vibration data, some abrupt noise points also need to be corrected, when the vibration data is larger, the abnormal behavior is more likely to be detected, but when the vibration data is more intensively distributed in a neighborhood range, the abnormal value is more likely to influence the judged noise points, and the degree of correcting the noise points is also larger. And obtaining the noise probability of each sampling moment according to the deviation degree of the corresponding vibration data in the preset neighborhood range and the distribution deviation condition of the vibration data in the preset neighborhood range of each sampling moment and the distribution difference condition of all vibration data in the preset neighborhood range of the corresponding sampling moment and other vibration data in the preset surrounding range of the corresponding sampling moment.
Preferably, for any one sampling time, calculating the average value of the values of vibration data corresponding to all sampling times in a preset neighborhood range of the sampling time, obtaining the vibration data average value of the sampling time, reflecting the average value condition of the vibration data corresponding to the sampling time in the range through the vibration data average value, using the ratio of the value of the vibration data corresponding to the sampling time to the vibration data average value as the value abnormality index of the sampling time, and when the value of the vibration data corresponding to the sampling time is larger, indicating that the noise condition is more similar to the abnormality condition, and judging that correction is needed.
Further, an extreme point in a preset neighborhood range of the sampling time is obtained, in the embodiment of the invention, curve fitting is performed on vibration data corresponding to the preset neighborhood range of the sampling time to obtain a vibration curve, and a point with zero first order derivative on the vibration curve is used as the extreme point. It should be noted that, the method of curve fitting and first-order extremum derivation is a technical means well known to those skilled in the art, such as curve fitting by using least square method, and the like, and will not be described herein. Calculating the difference between the numerical value of each extreme point and the mean value of vibration data to obtain the extreme value deviation difference of each extreme point, reflecting the overall deviation condition of the local vibration data extreme value through the deviation condition of the extreme value, taking the average value of the extreme value deviation differences of all the extreme points in a preset neighborhood range as the range distribution degree of the sampling moment, reflecting the concentration degree of the overall distribution of the vibration data in the local range at the sampling moment through the range distribution degree, and when the data distribution in the corresponding range at the sampling moment is concentrated, namely, the smaller the range distribution degree is, indicating that the vibration at the moment is not abnormal, and when the numerical value deviation degree at the sampling moment is larger, further indicating that the abnormality at the sampling moment is more likely to be noise.
Considering that there may be a situation that the sampling time is similar to the overall trend caused by abnormality in the corresponding local range, but the actual situation is still abnormal, the preset neighborhood range is enlarged, and the deviation of vibration data in a wider range is further limited.
Further, the average value of maximum values in all vibration data corresponding to the preset surrounding range except the preset neighborhood range of the sampling moment is calculated, the high surrounding average value of the sampling moment is obtained, the average value of minimum values in all vibration data corresponding to the preset surrounding range except the preset neighborhood range of the sampling moment is calculated, the low surrounding average value of the sampling moment is obtained, the analysis of the extremum average value is carried out through the vibration data outside the preset neighborhood range in the preset surrounding range, and the overall distribution condition of the local vibration data of the preset neighborhood range at each sampling moment is reflected. Counting the number of the vibration data corresponding to the preset neighborhood range of the sampling time, wherein the number is smaller than the high peripheral average value and larger than the low peripheral average value, obtaining the distribution concentration degree of the sampling time, reflecting the overall distribution range condition of the local vibration data outside the preset neighborhood range through the high peripheral average value and the low peripheral average value, and when the vibration data in the preset neighborhood range are more in the distribution range, indicating that the vibration data in the preset neighborhood range are more likely to be vibration data under normal conditions, and further indicating that the vibration data are more likely to be noise conditions when numerical value abnormality occurs.
And combining the distribution concentration degree and the range distribution degree to comprehensively reflect the normal condition of the whole data in the range of the corresponding time period of the sampling time, and obtaining the range concentration index of the sampling time according to the distribution concentration degree and the range distribution degree of the sampling time, wherein the distribution concentration degree and the range concentration index are positively correlated, the range distribution degree and the range concentration index are negatively correlated, and the range concentration index is a normalized value. The smaller the range distribution degree is, the more concentrated the distribution of vibration data in the neighborhood range at the sampling moment is, the larger the distribution concentration degree is, the more similar the distribution of vibration data in the neighborhood range at the sampling moment is to the distribution of local vibration data outside the neighborhood range is, and the more likely the distribution condition of the vibration data in the neighborhood range at the sampling moment is, namely the larger the range concentration index is.
Finally, taking the product of the numerical value abnormality index and the range concentration index at the sampling time as the noise probability at the sampling time, when the vibration data is larger than the average level and the distribution normality of the vibration data of the local range is higher, the abnormality at the sampling time is the noise probability is larger, and in the embodiment of the invention, the expression of the noise probability is as follows:
In the method, in the process of the invention,denoted as +.>Noise probability at each sampling instant, +.>Denoted as +.>The values of the vibration data corresponding to the sampling moments, < >>Denoted as +.>Vibration data mean value of each sampling moment, +.>Denoted as +.>Distribution concentration of individual sampling instants, +.>Denoted as +.>Total number of all extreme points in preset neighborhood range of each sampling time, < >>Denoted as +.>The first part of the preset neighborhood range of the sampling time>Numerical value of each extreme point->Represented as an absolute value extraction function,expressed as a normalization function>Expressed as a preset adjustment coefficient, which is set to 0.001 in the embodiment of the present invention, the purpose is to prevent the situation where the distribution is zero making the formula meaningless.
Wherein,denoted as +.>The first part of the preset neighborhood range of the sampling time>Extreme deviation difference of the extreme points +.>Denoted as +.>The extent of distribution of the individual sampling instants,denoted as +.>The range concentration index of each sampling moment reflects the positive correlation between the distribution concentration degree and the range concentration index in the form of the ratio, and the negative correlation between the range distribution degree and the range concentration index, in other embodiments of the invention, the distribution concentration degree can be reflected by other basic mathematical operations The range distribution degree is in positive correlation with the range concentration index, such as power operation, and the like, and the range distribution degree is in negative correlation with the range concentration index, such as subtraction, and the like, without limitation.
Denoted as +.>The larger the numerical abnormality index is, the more likely the vibration data at the sampling time is to be abnormal, and when the range concentration index is larger, the more likely the vibration data at the sampling time is to be normal, the more likely the abnormal condition at the sampling time is to be noise data is reflected, so the noise probability is higher.
So far, the analysis of the local distribution condition of the data at the time is completed, the noise probability is obtained to distinguish the noise data condition similar to the abnormal data, and then the noise is corrected more highly.
S4: obtaining a Kalman gain correction coefficient of each sampling moment according to the temperature influence degree and the noise possibility degree of each sampling moment and a preset number of sampling moments before the corresponding sampling moment; and monitoring the vibration data at all sampling moments based on the Kalman gain correction coefficient by obtaining filtering vibration data through a Kalman filtering algorithm.
For the kalman filter algorithm, the magnitude of the kalman gain determines the weight distribution of the measured value and the prior estimate in the state update, when the kalman gain is larger, the system is more dependent on the measured value, and when the kalman gain is smaller, the system is more dependent on the prior estimate, that is, the magnitude of the kalman gain determines the contribution degree of the historical data and the current data to the state update of the system.
Therefore, when the relation between the vibration data and the temperature data at the sampling moment is closer, the probability that the vibration data is influenced by the temperature and is false is higher, and when the probability of noise corresponding to the sampling moment is higher, the degree that the vibration data needs to be corrected is larger, and the Kalman gain needs to be smaller to be adjusted. However, since the kalman filtering algorithm is composed of the historical measurement data and the current estimate, the reliability of the data at the historical moment will also affect the adjustment of the kalman gain at the current moment, that is, when the historical moment for analysis needs to be corrected to a larger extent, more weight should be replayed on the current estimate, and the larger the kalman gain needs to be. Therefore, the Kalman gain correction coefficient of each sampling time is obtained according to the temperature influence degree and the noise possibility degree of each sampling time and the preset number of sampling times before the corresponding sampling time.
Preferably, the product of the temperature influence and the noise probability of each sampling time is taken as the influence of each sampling time, and the influence reflects the degree to which each sampling time needs to be modified. For any sampling time, the influence degree of the inverse proportion of the sampling time is used as an adjustment index of the sampling time, and the larger the correction degree required for the current sampling time is, the smaller the Kalman gain is required. And taking the accumulated value of the influence of a preset number of sampling moments before the sampling moment as the adjustment credibility weight of the sampling moment, wherein the larger the integral historical modification degree corresponding to the current sampling moment is, the larger the Kalman gain is required. In the embodiment of the present invention, the preset number is set to 50, and the specific numerical value implementation can be adjusted according to the specific implementation, which is not limited herein.
Further, calculating the product of the adjustment credibility weight and the adjustment index at the sampling moment, and adjusting the value range to obtain a Kalman gain correction coefficient at the sampling moment, wherein the Kalman gain correction coefficient is a weight obtained by correcting and optimizing the local analysis of the vibration data so as to adaptively adjust the Kalman gain of the vibration data corresponding to each sampling moment, and in the embodiment of the invention, the Kalman gain correction coefficient is expressed as follows:
in the method, in the process of the invention,denoted as +.>Kalman gain correction factor for each sampling instant, < ->Denoted as +.>Temperature influence at each sampling instant, +.>Denoted as +.>Noise probability at each sampling instant, +.>Denoted as +.>The total number of the preset number of sampling instants before the sampling instant,/->Denoted as +.>The first of a preset number of sampling instants before the sampling instant>Temperature influence at each sampling instant, +.>Denoted as +.>The first of a preset number of sampling instants before the sampling instant>Noise probability at each sampling instant, +.>Represented as a normalization function.
Wherein,denoted as +.>Influence of the individual sampling instants, +.>Denoted as +.>Adjustment index for individual sampling instants,/->Denoted as +. >The first of a preset number of sampling instants before the sampling instant>Influence of the individual sampling instants, +.>Denoted as +.>The adjustment of the trusted weights for the individual sampling instants,the numerical range expressed as a product is mapped into 0 to 1 by normalizing the range, and the numerical range is adjusted to between 0 and 2 by multiplying with 2. When the adjustment reliability weight is larger, the adjustment index is larger, which means that the current sampling time needs to be corrected to a large extent and the reliability degree of the historical data is high, the obtained Kalman gain can be larger, so that the Kalman gain correction coefficient is larger.
Finally, the vibration data at all sampling moments can be monitored through a Kalman filtering algorithm based on Kalman gain correction coefficients, and in one embodiment of the invention, when the vibration data at all sampling moments are filtered through the Kalman filtering algorithm, the product of the Kalman gain correction coefficients at each sampling moment and the corresponding Kalman gain is used as the optimized Kalman gain to be filtered, so that the filtered vibration data at each sampling moment is obtained. The Kalman filtering process mainly comprises the following steps: establishing a mathematical model for the signal to be processed, determining state variables and observables, and defining a dynamic equation and an observation equation of the system; estimating a state vector and an error covariance matrix of the system according to the initial state and the observed value of the system; predicting a state vector and an error covariance matrix at the next moment according to a dynamics equation of the system; according to the observation equation, the predicted value and the Kalman gain correction coefficient, calculating Kalman gain, and updating a state vector and an error covariance matrix; the prediction and update operations are repeated until all the data has been processed. It should be noted that, the kalman filtering is a well-known technique known to those skilled in the art, and specific applications will not be further described herein.
In the embodiment of the invention, the filtering vibration data can be transmitted to the monitoring unit for data monitoring, when the filtering vibration data is larger than the preset abnormal threshold value, the fact that the vibration data is larger at the moment is indicated, abnormal faults or processing behaviors possibly occur in the numerical control machine tool is possible, and the corresponding sampling time is recorded as the abnormal time. In the embodiment of the invention, the preset abnormal threshold is the upper boundary value of the normal vibration data, the normal vibration data can be collected through the normal running state of the numerical control machine, and the specific numerical value is specifically set according to the implementation scene of the running environment of the numerical control machine, so that the numerical control machine is not limited. When the continuous number of abnormal moments is greater than or equal to the preset abnormal number, the abnormal state is not a possible noise condition, manual processing is needed, the running state is marked as the abnormal state and early warning is carried out, and otherwise, the running state is marked as the normal state. In the embodiment of the present invention, the preset continuous number is set to 3, and the practitioner can adjust according to the specific implementation situation, which is not limited herein.
In summary, according to the analysis of the temperature influence and the noise possibility of the vibration data, the range change difference value of each sampling moment is obtained according to the deviation condition of the temperature data change and the difference condition of the vibration data and the temperature data in the change degree in a preset neighborhood range according to each sampling moment, and the consistency data of the change degree difference condition in the range of each sampling moment is obtained by eliminating the influence of the temperature data with abnormal conditions, so that whether the difference in the corresponding change of each sampling moment accords with the consistency is conveniently analyzed. Further, according to the consistency of vibration data and temperature data in the preset neighborhood range of each sampling time in the change trend, the range change difference value and the change degree difference condition, the temperature influence degree of the sampling time is obtained, and the influence of the temperature data on the vibration data is obtained through the relevance according to the consistency of the vibration data and the temperature data in the change trend and the change degree. And secondly, analyzing according to data distribution deviation of the vibration data possibly of the abnormal situation at each sampling moment, and adjusting through distribution differences of the vibration data in a preset neighborhood range and a preset surrounding range corresponding to the sampling moment, so that noise possibility which is more likely to be noise is obtained from the abnormal situation, and denoising of the vibration data is more accurate. Finally, the temperature influence degree and the noise possibility degree of the sampling time and the historical sampling time are combined to obtain the self-adaptive Kalman gain correction coefficient of each sampling time, so that the accuracy of a Kalman filtering algorithm can be improved, and the filtering vibration data can be obtained for monitoring. According to the invention, the influence condition of the temperature data during vibration data acquisition and the condition that noise in the vibration data is similar to abnormality are taken into consideration for comprehensive analysis, the Kalman gain in the Kalman filtering algorithm is adaptively adjusted, the denoising accuracy of the Kalman filtering algorithm is improved, and the follow-up monitoring accuracy of the vibration data of the running state of the numerical control machine tool is higher.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. The method for monitoring the running state of the numerical control machine tool is characterized by comprising the following steps of:
vibration data and temperature data of each sampling time are obtained in a preset sampling time period;
obtaining a range change difference value of each sampling moment according to the change deviation condition of the temperature data corresponding to each sampling moment in the preset neighborhood range and the difference condition of the vibration data and the temperature data corresponding to each sampling moment in the preset neighborhood range in the change degree; obtaining the temperature influence degree of each sampling moment according to the similarity degree of the variation trend of the vibration data and the temperature data in the preset neighborhood range of each sampling moment, the range variation difference value of the corresponding sampling moment and the variation degree difference condition of the vibration data and the temperature data at the corresponding sampling moment;
Obtaining the noise probability of each sampling moment according to the deviation degree of the corresponding vibration data of each sampling moment in a preset neighborhood range and the distribution deviation condition of the vibration data in the preset neighborhood range, and according to the distribution difference condition of all vibration data in the preset neighborhood range of the corresponding sampling moment and other vibration data in the preset surrounding range of the corresponding sampling moment; the preset surrounding range is larger than the preset neighborhood range;
obtaining a Kalman gain correction coefficient of each sampling moment according to the temperature influence degree and the noise possibility degree of each sampling moment and a preset number of sampling moments before the corresponding sampling moment; and monitoring the vibration data at all sampling moments based on the Kalman gain correction coefficient by obtaining filtering vibration data through a Kalman filtering algorithm.
2. The method for monitoring the operation state of a numerical control machine according to claim 1, wherein the method for acquiring the range variation difference value comprises the steps of:
taking each sampling time as a reference time in sequence, and for any sampling time in a preset neighborhood range of the reference time, obtaining the temperature credibility weight of the sampling time according to the value of temperature data of the sampling time and the deviation condition of the variation degree in the preset neighborhood range;
Calculating the difference between the slope of the temperature data corresponding to the sampling moment and the slope of the vibration data corresponding to the sampling moment to obtain a variation difference index of the sampling moment; taking the product of the variation difference index of the sampling moment and the temperature credible weight as an adjustment difference index of the sampling moment;
and calculating the average value of the adjustment difference indexes of all sampling moments in the preset neighborhood range of the reference moment to obtain the range change difference value of the reference moment.
3. The method for monitoring the operation state of the numerical control machine according to claim 2, wherein the method for acquiring the temperature credible weight comprises the following steps:
calculating the average value of the values of the temperature data corresponding to all the sampling moments in a preset neighborhood range of the reference moment to obtain the average value of the temperature data of the reference moment; calculating the difference between the value of the temperature data corresponding to the sampling time and the temperature average value to obtain a temperature data difference value of the sampling time;
calculating the average value of the slopes of the temperature data corresponding to all sampling moments, and obtaining the average value of the temperature slopes at the reference moment; calculating the difference between the slope of the temperature data corresponding to the sampling time and the average value of the temperature slope to obtain a temperature slope difference value of the sampling time;
Carrying out negative correlation mapping and normalization processing on the product of the temperature data difference value and the temperature slope difference value at the sampling moment to obtain a temperature similarity index at the sampling moment;
calculating accumulated values of temperature similarity indexes of all sampling moments in a preset neighborhood range of the reference moment to obtain a total temperature similarity index of the reference moment; and taking the ratio of the temperature similarity index at the sampling moment to the total temperature similarity index as the temperature credible weight at the sampling moment.
4. The method for monitoring the operation state of a numerically-controlled machine tool according to claim 2, wherein the method for acquiring the temperature influence degree comprises the following steps:
for any sampling moment, calculating the difference between the change difference index of the sampling moment and the range change difference value of the sampling moment, performing negative correlation mapping and normalizing to obtain the change correlation degree of the sampling moment;
counting the number of the same sign between the slope of vibration data and the slope of temperature data at each sampling moment in a preset neighborhood range of the sampling moment to obtain the trend correlation degree of the sampling moment;
obtaining the temperature influence degree of the sampling moment according to the change correlation degree and the trend correlation degree of the sampling moment; both the change correlation and the trend correlation are positively correlated with the temperature influence.
5. The method for monitoring the operation state of a numerical control machine according to claim 1, wherein the method for acquiring the noise probability comprises the steps of:
for any sampling moment, calculating the average value of the values of vibration data corresponding to all sampling moments in a preset neighborhood range of the sampling moment, and obtaining the vibration data average value of the sampling moment; taking the ratio of the value of the vibration data corresponding to the sampling moment to the average value of the vibration data as the value abnormality index of the sampling moment;
obtaining extreme points in a preset neighborhood range of the sampling moment; calculating the difference between the value of each extreme point and the mean value of the vibration data to obtain the extreme deviation difference of each extreme point; taking the average value of extreme value deviation differences of all extreme value points in a preset neighborhood range as the range distribution degree of the sampling moment;
calculating the average value of maximum values in all vibration data corresponding to the preset surrounding range of the sampling moment except the preset neighborhood range, and obtaining the high surrounding average value of the sampling moment; calculating the average value of minimum values in all vibration data corresponding to the preset surrounding range of the sampling moment except the preset neighborhood range, and obtaining the low surrounding average value of the sampling moment; counting the number of the vibration data corresponding to the preset neighborhood range of the sampling moment, wherein the number is smaller than the high peripheral average value and larger than the low peripheral average value, so as to obtain the distribution concentration degree of the sampling moment;
Obtaining a range concentration index of the sampling moment according to the distribution concentration degree and the range distribution degree of the sampling moment; the distribution concentration degree is positively correlated with the range concentration index, and the range distribution degree is negatively correlated with the range concentration index; the range concentration index is a normalized value;
and taking the product of the numerical value abnormal index and the range concentration index at the sampling time as the noise possibility of the sampling time.
6. The method for monitoring the operation state of a numerically-controlled machine tool according to claim 5, wherein the obtaining the extreme point in the preset neighborhood range at the sampling time comprises:
performing curve fitting on vibration data corresponding to a preset neighborhood range of the sampling moment to obtain a vibration curve; and taking the point when the first order derivative on the vibration curve is zero as an extreme point.
7. The method for monitoring the operation state of a numerical control machine according to claim 1, wherein the method for acquiring the kalman gain correction coefficient comprises the steps of:
taking the product of the temperature influence degree and the noise possibility degree of each sampling moment as the influence degree of each sampling moment;
for any sampling moment, taking the influence degree of the inverse proportion of the sampling moment as an adjustment index of the sampling moment; taking the accumulated value of the influence of a preset number of sampling moments before the sampling moment as the adjustment credibility of the sampling moment;
And calculating the product of the adjustment credible weight and the adjustment index of the sampling moment, and adjusting the value range to obtain the Kalman gain correction coefficient of the sampling moment.
8. The method for monitoring the operation state of the numerically-controlled machine tool according to claim 1, wherein the monitoring the vibration data at all sampling moments based on the kalman gain correction coefficient by the kalman filter algorithm includes:
when the vibration data at all sampling moments are filtered through a Kalman filtering algorithm, taking the product of a Kalman gain correction coefficient and a corresponding Kalman gain at each sampling moment as the optimized Kalman gain to filter, and obtaining the filtered vibration data at each sampling moment;
when the filtering vibration data is larger than a preset abnormal threshold value, the corresponding sampling time is marked as an abnormal time; when the continuous number of abnormal moments is greater than or equal to the preset abnormal number, the running state is recorded as an abnormal state and early warning is carried out; otherwise, the running state is recorded as a normal state.
9. The method for monitoring the running state of the numerical control machine according to claim 1, wherein the preset neighborhood range is set to be a neighborhood range with a side length of 50 by taking a sampling time as a center.
10. The method for monitoring the operation state of a numerical control machine according to claim 1, wherein the preset surrounding range is set to be 4 times as large as the preset neighborhood range.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410224010.7A CN117798744B (en) | 2024-02-29 | 2024-02-29 | Method for monitoring running state of numerical control machine tool |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410224010.7A CN117798744B (en) | 2024-02-29 | 2024-02-29 | Method for monitoring running state of numerical control machine tool |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117798744A true CN117798744A (en) | 2024-04-02 |
CN117798744B CN117798744B (en) | 2024-05-10 |
Family
ID=90420315
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410224010.7A Active CN117798744B (en) | 2024-02-29 | 2024-02-29 | Method for monitoring running state of numerical control machine tool |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117798744B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118114071A (en) * | 2024-04-23 | 2024-05-31 | 中铁建工集团第二建设有限公司 | Drill bit drilling state monitoring system for cutting underground anchor cable |
CN118114082A (en) * | 2024-04-29 | 2024-05-31 | 海易(大连)实业有限公司 | Centrifuge data remote monitoring and analyzing system based on cloud computing |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101549543B1 (en) * | 2014-04-29 | 2015-09-07 | 충북대학교 산학협력단 | A power state diagnosis method using kalman estimation process and measuring the relative probability by the metric defined by functional mapping |
CN110977614A (en) * | 2019-12-18 | 2020-04-10 | 常州机电职业技术学院 | Health diagnosis method for numerical control machine tool |
CN112699926A (en) * | 2020-12-25 | 2021-04-23 | 浙江中控技术股份有限公司 | Method for identifying saturated grinding abnormity of cement raw material vertical mill based on artificial intelligence technology |
CN112710358A (en) * | 2021-03-29 | 2021-04-27 | 南京诚远高新科技有限公司 | Intelligent machine tool state monitoring device and state monitoring method thereof |
WO2023000532A1 (en) * | 2021-03-17 | 2023-01-26 | 江苏中科君达物联网股份有限公司 | New vibration noise test method |
WO2023045425A1 (en) * | 2021-09-23 | 2023-03-30 | 重庆山外山血液净化技术股份有限公司 | Method and system for evaluating risk of intradialytic hypotension event |
WO2023061850A1 (en) * | 2021-10-11 | 2023-04-20 | Reishauer Ag | Method for monitoring the state of a machine tool |
CN116861313A (en) * | 2023-07-07 | 2023-10-10 | 昆明理工大学 | Kalman filtering working condition identification method and system based on vibration energy trend |
CN116871982A (en) * | 2023-06-21 | 2023-10-13 | 中国第一汽车股份有限公司 | Device and method for detecting spindle of numerical control machine tool and terminal equipment |
CN117131747A (en) * | 2023-10-26 | 2023-11-28 | 深圳市路桥建设集团有限公司 | State estimation method and device based on sampling point Kalman filtering |
CN117236084A (en) * | 2023-11-16 | 2023-12-15 | 青岛永强木工机械有限公司 | Intelligent management method and system for woodworking machining production |
CN117313029A (en) * | 2023-09-27 | 2023-12-29 | 云南电网有限责任公司楚雄供电局 | Multi-sensor data fusion method based on Kalman filtering parameter extraction and state updating |
DE112021007281T5 (en) * | 2021-05-26 | 2024-01-04 | Fanuc Corporation | Vibration detection device |
CN117349611A (en) * | 2023-12-06 | 2024-01-05 | 山东清控生态环境产业发展有限公司 | Water quality fluctuation instrument monitoring method based on big data analysis |
-
2024
- 2024-02-29 CN CN202410224010.7A patent/CN117798744B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101549543B1 (en) * | 2014-04-29 | 2015-09-07 | 충북대학교 산학협력단 | A power state diagnosis method using kalman estimation process and measuring the relative probability by the metric defined by functional mapping |
CN110977614A (en) * | 2019-12-18 | 2020-04-10 | 常州机电职业技术学院 | Health diagnosis method for numerical control machine tool |
CN112699926A (en) * | 2020-12-25 | 2021-04-23 | 浙江中控技术股份有限公司 | Method for identifying saturated grinding abnormity of cement raw material vertical mill based on artificial intelligence technology |
WO2023000532A1 (en) * | 2021-03-17 | 2023-01-26 | 江苏中科君达物联网股份有限公司 | New vibration noise test method |
CN112710358A (en) * | 2021-03-29 | 2021-04-27 | 南京诚远高新科技有限公司 | Intelligent machine tool state monitoring device and state monitoring method thereof |
DE112021007281T5 (en) * | 2021-05-26 | 2024-01-04 | Fanuc Corporation | Vibration detection device |
WO2023045425A1 (en) * | 2021-09-23 | 2023-03-30 | 重庆山外山血液净化技术股份有限公司 | Method and system for evaluating risk of intradialytic hypotension event |
WO2023061850A1 (en) * | 2021-10-11 | 2023-04-20 | Reishauer Ag | Method for monitoring the state of a machine tool |
CN116871982A (en) * | 2023-06-21 | 2023-10-13 | 中国第一汽车股份有限公司 | Device and method for detecting spindle of numerical control machine tool and terminal equipment |
CN116861313A (en) * | 2023-07-07 | 2023-10-10 | 昆明理工大学 | Kalman filtering working condition identification method and system based on vibration energy trend |
CN117313029A (en) * | 2023-09-27 | 2023-12-29 | 云南电网有限责任公司楚雄供电局 | Multi-sensor data fusion method based on Kalman filtering parameter extraction and state updating |
CN117131747A (en) * | 2023-10-26 | 2023-11-28 | 深圳市路桥建设集团有限公司 | State estimation method and device based on sampling point Kalman filtering |
CN117236084A (en) * | 2023-11-16 | 2023-12-15 | 青岛永强木工机械有限公司 | Intelligent management method and system for woodworking machining production |
CN117349611A (en) * | 2023-12-06 | 2024-01-05 | 山东清控生态环境产业发展有限公司 | Water quality fluctuation instrument monitoring method based on big data analysis |
Non-Patent Citations (1)
Title |
---|
石荣波;: "基于状态监测的高档数控机床智能保障***研究", 自动化与仪器仪表, no. 05, 25 September 2011 (2011-09-25), pages 23 - 25 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118114071A (en) * | 2024-04-23 | 2024-05-31 | 中铁建工集团第二建设有限公司 | Drill bit drilling state monitoring system for cutting underground anchor cable |
CN118114082A (en) * | 2024-04-29 | 2024-05-31 | 海易(大连)实业有限公司 | Centrifuge data remote monitoring and analyzing system based on cloud computing |
Also Published As
Publication number | Publication date |
---|---|
CN117798744B (en) | 2024-05-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117798744B (en) | Method for monitoring running state of numerical control machine tool | |
US7233884B2 (en) | Methodology for temporal fault event isolation and identification | |
CN117196353B (en) | Environmental pollution assessment and monitoring method and system based on big data | |
CN117421610B (en) | Data anomaly analysis method for electric energy meter running state early warning | |
CN117407700A (en) | Method for monitoring working environment in live working process | |
CN117236084B (en) | Intelligent management method and system for woodworking machining production | |
CN116976684B (en) | Risk model predictive control method and system for logistics conveyor | |
CN116985183B (en) | Quality monitoring and management method and system for near infrared spectrum analyzer | |
CN116861313B (en) | Kalman filtering working condition identification method and system based on vibration energy trend | |
JP2014035282A (en) | Abnormality diagnostic device | |
CN117571107B (en) | Intelligent unattended wagon balance anomaly monitoring system | |
CN117420346A (en) | Circuit protection board overcurrent value detection method and system | |
CN117349664A (en) | On-line monitoring method and system for sprouting vegetable growth environment | |
CN116258087B (en) | Matte grade soft measurement method and device, electronic equipment and storage medium | |
CN117470529A (en) | Static friction fault detection method and system for valve of process control system | |
CN112329258B (en) | Test box part aging judgment method and device, electronic equipment and storage medium | |
JP7437163B2 (en) | Diagnostic equipment, diagnostic methods and programs | |
CN114611633A (en) | Health monitoring method of electromagnetic valve | |
CN117648537B (en) | Atmospheric pollution real-time monitoring method and system based on hyperspectral technology | |
Musaeva et al. | Algorithms for constructing the confidence interval of the mathematical expectation of the noise and its application for calculating statistical characteristics | |
CN117906937B (en) | Hydrostatic test method, test system and storage medium | |
CN118090078B (en) | Leakage online monitoring method for closed circulation water cooling system | |
CN116952354B (en) | Data optimization acquisition method for driving measurement sensor | |
CN117172431B (en) | Food processing apparatus and equipment management method and system | |
CN118136139B (en) | Solid-liquid phase balance production management method for doxifluridine based on predictive model |
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 |